A calculation method and computer-readable medium for evaluating the adjustable potential of air conditioning load in a substation considering end-user usage behavior

Abstract

The present application discloses a calculation method and computer-readable medium for evaluating the adjustable potential of the air-conditioning load in a substation, which takes into account the usage behavior of the end users and the differences in the location of the air conditioners. By combining a deep learning algorithm, the method takes into account the usage behavior of the end users and the differences in the location of the air conditioners. The method uses a cluster analysis method to classify the time series characteristic distribution of the air-conditioning load with the same user usage behavior pattern and air-conditioning location characteristics into one category to form a matching database as benchmark information. The time series correlation distribution pattern of the air-conditioning load of the single air conditioner is then matched and associated with the benchmark patterns of each type in the matching database to serve as the basis for evaluating the adjustable potential of the air-conditioning load, thereby realizing the intelligent generation of the upper limit value and the lower limit value of the temperature adjustment of the air conditioner. In this way, the upper and lower limits of the adjustable potential of the air-conditioning load are obtained, providing important technical support for temperature control and optimization of energy distribution.

Description

Technical Field

[0001] The present application relates to the field of intelligent evaluation, and more specifically, to a calculation method for evaluating the adjustable potential of air-conditioning load in a substation taking into account the usage behavior of end users and a computer-readable medium. Background Art

[0002] With global climate change and accelerating urbanization, air conditioners, as essential equipment for regulating indoor environments, are facing increasing energy consumption challenges. Evaluating the adjustable potential of air conditioning loads is crucial for optimizing energy distribution, reducing energy costs, and improving energy efficiency.

[0003] However, existing studies evaluating the adjustable potential of air conditioning load often overlook the diversity and complexity of end-user usage behaviors, as well as the varying demands placed on air conditioning by different users and spaces. Therefore, a computational method and computer-readable medium for evaluating the adjustable potential of air conditioning load in a substation that considers end-user usage behaviors is desired. Summary of the Invention

[0004] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a calculation method and computer-readable medium for evaluating the adjustable potential of the air-conditioning load in a substation taking into account the usage behavior of the terminal users. By combining a deep learning algorithm, the method takes into account the differences in the usage behavior of the terminal users and the location of the air conditioners. The air-conditioning load time series characteristic distributions with the same user usage behavior pattern and air-conditioning location characteristics are classified into one category through a cluster analysis method to form a matching database as benchmark information. The air-conditioning load time series correlation distribution pattern of the single air conditioner is then matched and associated with the benchmark patterns of each type in the matching database to serve as the basis for evaluating the adjustable potential of the air-conditioning load, thereby realizing the intelligent generation of the upper limit value and the lower limit value of the temperature adjustment of the air-conditioning. In this way, the upper and lower limits of the adjustable potential of the air-conditioning load are obtained, providing important technical support for temperature control and optimization of energy allocation.

[0005] According to one aspect of the present application, a method for evaluating and calculating the adjustable potential of air conditioning load in a substation taking into account the usage behavior of end users is provided, which includes:

[0006] Get a collection of air conditioning load data;

[0007] Based on the differences in user usage behaviors and air conditioner locations in the low-voltage area, cluster analysis is performed on the set of air conditioner load data to obtain a set of cluster subsets of air conditioner load data;

[0008] Performing air conditioning load time series pattern feature extraction on the set of the air conditioning load data cluster subsets to obtain a set of air conditioning load data cluster subset time series pattern feature vectors;

[0009] Obtain the time series of air conditioning load data of a single air conditioner;

[0010] Extracting time series pattern features of the air conditioning load of the individual air conditioner from the time series of the air conditioning load data to obtain a time series associated implicit feature vector of the individual air conditioning load data;

[0011] Performing a matching query on the implicit feature vector associated with the time series of the single air-conditioning load data and the set of the time series pattern feature vectors of the cluster subset of the air-conditioning load data to obtain a semantic coding vector of the adjustable potential of the air-conditioning load;

[0012] Based on the air-conditioning load adjustable potential semantic coding vector, an upper temperature adjustment limit value and a lower temperature adjustment limit value are determined.

[0013] According to another aspect of the present application, a computer-readable medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes the above-mentioned method for evaluating the adjustable potential of the air-conditioning load in the substation considering the usage behavior of the end users.

[0014] Compared with the prior art, the present application provides a method for calculating the potential for evaluating the adjustable air-conditioning load in a substation, which takes into account the usage behavior of end users and the differences in the location of the air conditioners, by combining a deep learning algorithm. The method uses a cluster analysis method to classify the time series characteristic distribution of air-conditioning loads with the same user usage behavior pattern and air-conditioning location characteristics into one category to form a matching database as benchmark information. The time series correlation distribution pattern of the air-conditioning load of a single air conditioner is then matched and associated with the benchmark patterns of each type in the matching database to serve as the basis for evaluating the adjustable potential of the air-conditioning load, thereby realizing the intelligent generation of the upper and lower limits of the air-conditioning temperature adjustment. In this way, the upper and lower limits of the adjustable potential of the air-conditioning load are obtained, providing important technical support for temperature control and optimizing energy allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The above and other purposes, features, and advantages of the present application will become more apparent through a more detailed description of the embodiments of the present application in conjunction with the accompanying drawings. The accompanying drawings are intended to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0016] Figure 1 Flowchart of a method for evaluating and calculating the adjustable potential of air-conditioning load in a substation considering the usage behavior of end users according to an embodiment of the present application;

[0017] Figure 2A system architecture diagram of a method for evaluating and calculating the adjustable potential of air-conditioning load in a substation considering the usage behavior of end users according to an embodiment of the present application;

[0018] Figure 3 This is a flowchart of sub-step S6 of the method for evaluating the adjustable potential of air-conditioning load in a substation considering the usage behavior of end users according to an embodiment of the present application;

[0019] Figure 4 This is a flowchart of an embodiment of the present application. DETAILED DESCRIPTION

[0020] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0021] As used in this application and the claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not intended to refer to the singular but may include the plural. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list. A method or apparatus may also include other steps or elements.

[0022] Although the present application makes various references to certain modules in the system according to embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are illustrative only, and different aspects of the system and method can use different modules.

[0023] Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the various steps may be processed in reverse order or simultaneously, as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0024] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0025] Existing studies evaluating the adjustable potential of air conditioning loads often overlook the diversity and complexity of end-user usage behaviors, as well as the varying demands placed on air conditioning by different users and spaces. Air conditioning usage behavior is influenced by numerous factors, including but not limited to user comfort preferences, indoor and outdoor temperature differences, user activity patterns, and lifestyle habits. These factors collectively determine users' demand for and ability to respond to air conditioning adjustments. For example, air conditioning usage patterns in offices and homes can differ significantly, and the same user's air conditioning needs may vary across time periods. Therefore, analyzing from a single user type may not fully capture the adjustable potential of air conditioning loads.

[0026] Furthermore, existing technologies often use user sensitivity to electricity prices as the basis for evaluation. While this approach is simple and intuitive, it overlooks the individual needs of users regarding air conditioning usage and their willingness to respond. A user's sensitivity to electricity prices does not always directly reflect their ability to respond to air conditioning load adjustments, especially when faced with complex usage environments and diverse user needs.

[0027] In the technical solution of the present application, a calculation method for evaluating the adjustable potential of air-conditioning load in a substation is proposed, which takes into account the usage behavior of end users. Figure 1 The flowchart is a calculation method for evaluating the adjustable potential of air-conditioning load in a substation considering the usage behavior of end users according to an embodiment of the present application. Figure 2 This is a system architecture diagram of a method for evaluating the potential for adjusting the air conditioning load in a substation considering the end-user usage behavior according to an embodiment of the present application. Figure 1 and Figure 2 As shown, according to an embodiment of the present application, a calculation method for evaluating the adjustable potential of the air-conditioning load in a substation considering the usage behavior of terminal users includes the following steps: S1, obtaining a set of air-conditioning load data; S2, performing cluster analysis on the set of air-conditioning load data based on the differences in user usage behavior and the location of the air-conditioning in the low-voltage substation to obtain a set of cluster subsets of the air-conditioning load data; S3, performing air-conditioning load time series pattern feature extraction on the set of cluster subsets of the air-conditioning load data to obtain a set of time series pattern feature vectors of the cluster subsets of the air-conditioning load data; S4, obtaining the time series of the air-conditioning load data of a single air-conditioner; S5, performing air-conditioning load time series pattern feature extraction on the time series of the air-conditioning load data of the single air-conditioner to obtain a time series associated implicit feature vector of the single air-conditioning load data; S6, performing a matching query on the time series associated implicit feature vector of the single air-conditioning load data and the set of time series pattern feature vectors of the cluster subsets of the air-conditioning load data to obtain a semantic coding vector of the adjustable potential of the air-conditioning load; S7, determining the upper limit value and the lower limit value of the temperature adjustment based on the semantic coding vector of the adjustable potential of the air-conditioning load.

[0028] In particular, the S1 and S2 obtain a set of air-conditioning load data; and based on the differences in user usage behavior and the location of the air conditioner under the low-voltage station, cluster analysis is performed on the set of air-conditioning load data to obtain a set of cluster subsets of air-conditioning load data. In an embodiment of the present application, users in residential areas are divided into three categories according to working hours: regular work, non-routine work, and non-work. In the same type of residents, air conditioners are divided into two categories according to the installation location: living room and bedroom; users in commercial areas are divided according to industry; and users in office areas are divided according to working hours. Air-conditioning load data with similar behavioral habits are grouped together according to grouping indicators to obtain a set of cluster subsets of air-conditioning load data. In this way, the air-conditioning usage behavior patterns of different user groups can be identified by dividing by working hours and industries. For example, regular working users may use air conditioners less during the daytime period on weekdays, while non-working users may have more uniform air-conditioning usage time throughout the day. It should be understood that cluster analysis can be used to group air-conditioning load data with similar characteristics, which helps to conduct refined data analysis and form a more organized matching database, providing an important data basis for the intelligent evaluation of the upper and lower limits of the adjustable potential of air-conditioning loads.

[0029] In particular, the step S3 extracts air conditioning load time series pattern features from the set of air conditioning load data cluster subsets to obtain a set of air conditioning load data cluster subset time series pattern feature vectors. In particular, in a specific example of the present application, each air conditioning load data cluster subset in the set of air conditioning load data cluster subsets is subjected to an air conditioning load time series pattern feature extractor based on an RNN model to obtain a set of air conditioning load data cluster subset time series pattern feature vectors. That is, an air conditioning load time series pattern feature extractor is constructed using an RNN model to capture the air conditioning load time series change trend and dynamic fluctuation pattern of each of the air conditioning load data cluster subsets.

[0030] It's worth noting that RNNs (Recurrent Neural Networks) are deep learning models specifically designed for processing sequential data, such as time series data, text, speech, and other data with temporal relationships. Compared to traditional feedforward neural networks, RNNs have a memory function that can capture temporal dependencies in sequential data.

[0031] Specifically, step S4 obtains a time series of air conditioning load data for a single air conditioner. The time series of air conditioning load data for the single air conditioner includes unique operating characteristics of the single air conditioner. These unique operating characteristics may vary depending on factors such as user habits, the air conditioner's own characteristics, and the installation environment.

[0032] In particular, the S5 performs air-conditioning load time series pattern feature extraction on the time series of the air-conditioning load data of the individual air-conditioner to obtain the implicit feature vector of the time series association of the air-conditioning load data of the individual air-conditioner. In particular, in a specific example of the present application, the time series of the air-conditioning load data of the individual air-conditioner is passed through the air-conditioning load time series pattern feature extractor based on the RNN model to obtain the implicit feature vector of the time series association of the air-conditioning load data of the individual air-conditioner. That is, the air-conditioning load time series pattern feature extractor based on the RNN model is similarly used to capture the air-conditioning load time series change pattern and fluctuation law contained in the time series of the air-conditioning load data of the individual air-conditioner, so as to understand the air-conditioning usage behavior and air-conditioning operation status characteristics of a specific user.

[0033] In particular, the S6 performs a matching query on the implicit feature vector associated with the time series of the single air-conditioning load data and the set of the time series pattern feature vectors of the cluster subset of the air-conditioning load data to obtain the semantic coding vector of the adjustable potential of the air-conditioning load. The implicit feature vector associated with the time series of the single air-conditioning load data represents the unique usage mode and operating status characteristics of the single air-conditioning. By using it as a query feature vector and performing a matching query analysis with the set of the time series pattern feature vectors of the cluster subset of the air-conditioning load data, the cluster subset similar to the usage mode of the specific single air-conditioning can be searched in the matching database, thereby deeply understanding the adjustable potential of the air-conditioning load. In other words, the obtained semantic coding vector of the adjustable potential of the air-conditioning load is a comprehensive abstract representation of the dynamic interaction between the single air-conditioning and each subset in the matching database. In particular, in a specific example of the present application, such as Figure 3 As shown, the S6 includes: using the implicit feature vector associated with the time series of the single air-conditioning load data as a query feature vector, and inputting the query feature vector and the set of the time series pattern feature vector of the air-conditioning load data cluster subset into a feature query module based on information flow interaction to obtain the semantic coding vector of the adjustable potential of the air-conditioning load. More specifically, the implicit feature vector of the temporal association of the single air-conditioning load data is used as the query feature vector, and the query feature vector and the set of the temporal pattern feature vectors of the air-conditioning load data cluster subset are input into the feature query module based on information flow interaction to obtain the semantic coding vector of the adjustable potential of the air-conditioning load, including: S61, calculating the semantic association between each temporal pattern feature vector of the air-conditioning load data cluster subset in the set of the temporal pattern feature vectors of the air-conditioning load data cluster subset and the query feature vector to obtain a sequence of semantic associations; S62, normalizing each semantic association in the sequence of semantic associations to obtain a sequence of information interaction matching query factors; S63, using the sequence of information interaction matching query factors as weights, calculating the position-weighted sum of the set of temporal pattern feature vectors of the air-conditioning load data cluster subset to obtain the semantic coding vector of the adjustable potential of the air-conditioning load.

[0034] Specifically, the S61 calculates the semantic association between each air-conditioning load data cluster subset time series pattern feature vector in the set of the air-conditioning load data cluster subset time series pattern feature vectors and the query feature vector to obtain a sequence of semantic associations. In a specific example of the present application, after calculating the position-by-element difference between the query feature vector and each air-conditioning load data cluster subset time series pattern feature vector in the set of the air-conditioning load data cluster subset time series pattern feature vectors, the norm of each feature vector obtained is calculated to obtain the sequence of semantic associations. That is, the semantic association quantifies the similarity and degree of association between different air-conditioning load data cluster subsets and the individual air-conditioning behavior patterns, providing a basis for subsequent information interaction.

[0035] Specifically, the S62 normalizes each semantic correlation in the sequence of semantic correlation to obtain a sequence of information interaction matching query factors. In a specific example of the present application, each semantic correlation in the sequence of semantic correlation is subjected to soft maximum normalization processing based on a softmax activation function to obtain a sequence of information interaction matching query factors. That is, each semantic correlation in the sequence of semantic correlation is normalized to ensure the comparability of each semantic correlation, thereby obtaining a sequence of information interaction matching query factors. Wherein, the sequence of information interaction matching query factors reflects the relative importance between different cluster subsets and the query feature vector.

[0036] Specifically, in S63, using the sequence of information interaction matching query factors as weights, the position-weighted sum of the set of time series pattern feature vectors of the air conditioning load data cluster subsets is calculated to obtain the semantic encoding vector of the adjustable potential of the air conditioning load. It should be understood that by using the sequence of information interaction matching query factors as weights to calculate the weighted sum of the time series pattern feature vectors of the air conditioning load data cluster subsets, the importance of each time series pattern feature vector to the query feature vector can be more accurately captured. In this way, sufficient fluid interaction can be achieved between the time series distribution pattern of the individual air conditioning load and each cluster subset, realizing information integration and dynamic search.

[0037] In summary, in the above embodiment, the implicit feature vector associated with the time series of the single air-conditioning load data is used as a query feature vector, and the query feature vector and the set of the time series pattern feature vectors of the cluster subset of the air-conditioning load data are input into a feature query module based on information flow interaction to obtain the semantic coding vector of the adjustable potential of the air-conditioning load. The method includes: performing a feature query on the query feature vector and the set of the time series pattern feature vectors of the cluster subset of the air-conditioning load data using the following formula to obtain the semantic coding vector of the adjustable potential of the air-conditioning load through the feature query module based on information flow interaction; wherein, the formula is:

[0038] r i =‖v q -v i ‖1

[0039]

[0040] Among them, v q is the query feature vector, v i is the time series pattern feature vector of the ith air conditioning load data cluster subset in the set of time series pattern feature vectors of the air conditioning load data cluster subset, r i is the i-th semantic relevance, r k is the kth semantic association, ‖·‖1 represents the norm of the vector, m represents the total number of the time series pattern feature vectors of the air conditioning load data cluster subset in the set of the time series pattern feature vectors of the air conditioning load data cluster subset, exp(·) represents the exponential function with a natural constant as the base, and a is the semantic coding vector of the adjustable potential of the air conditioning load.

[0041] It is worth mentioning that in other specific examples of the present application, the temporal association implicit feature vector of the individual air-conditioning load data can be used as a query feature vector in other ways, and the query feature vector and the set of the temporal pattern feature vector of the cluster subset of the air-conditioning load data are input into the feature query module based on information flow interaction to obtain the semantic coding vector of the adjustable potential of the air-conditioning load. For example: the temporal association implicit feature vector of the individual air-conditioning load data and the set of the temporal pattern feature vector of the cluster subset of the air-conditioning load data are input; the query feature vector and the set of the cluster subset temporal pattern feature vector are interacted through the feature query module based on information flow interaction to obtain the semantic coding vector of the adjustable potential of the air-conditioning load; the query feature vector and the set of the cluster subset temporal pattern feature vector are input into the feature query module; in the feature query module, the correlation and mutual influence between the query feature vector and each temporal pattern feature vector are considered through information flow interaction; in the feature query module, the query feature vector and the temporal pattern feature vector are fused to capture the semantic correlation between them; and the semantic coding vector of the adjustable potential of the air-conditioning load is obtained through information flow interaction.

[0042] In particular, S7 determines the upper and lower temperature adjustment limits based on the semantically encoded vector of the air conditioning load adjustable potential. In particular, in a specific example of the present application, the semantically encoded vector of the air conditioning load adjustable potential is input into a decoder-based temperature upper and lower limit generator to obtain the upper and lower temperature adjustment limits. In this way, the load time series distribution pattern of the individual air conditioners, which is integrated with the benchmark information in the matching database, is used as the basis for judgment, achieving automated evaluation of the upper and lower limits of the air conditioning load adjustable potential.

[0043] Here, in the technical solution of the present application, in the process of using the implicit feature vector associated with the temporal sequence of the individual air-conditioning load data as the query feature vector and inputting the query feature vector and the set of temporal pattern feature vectors of the cluster subsets of the air-conditioning load data into the feature query module based on information flow interaction, the essence is to use the association information between the implicit feature vector associated with the temporal sequence of the individual air-conditioning load data and the temporal pattern feature vectors of the cluster subsets of the air-conditioning load data as the query matching confidence, and to integrate the query matching confidence based on the attention mechanism to obtain the semantic encoding vector of the adjustable potential of the air-conditioning load. However, considering the difference in feature distribution probability density between the implicit feature vector associated with the temporal sequence of the individual air-conditioning load data and the set of temporal pattern feature vectors of the cluster subsets of the air-conditioning load data, this will lead to local feature overflow due to probability density misalignment during the process of fusing the query matching confidence using the attention mechanism, thereby affecting the accuracy of the decoding result obtained by the decoder-based upper and lower temperature limit generator of the semantic encoding vector of the adjustable potential of the air-conditioning load.

[0044] Preferably, the step of passing the air-conditioning load adjustable potential semantic coding vector through a decoder-based temperature upper and lower limit generator to obtain a temperature adjustment upper limit value and a temperature adjustment lower limit value comprises the following steps:

[0045] Calculating an air conditioning load adjustable potential semantic coding and representation matrix and an air conditioning load adjustable potential semantic coding difference representation matrix of the air conditioning load adjustable potential semantic coding vector, wherein the value of each position of the air conditioning load adjustable potential semantic coding and representation matrix is ​​the mean of a pair of eigenvalues ​​of two positions of the air conditioning load adjustable potential semantic coding vector corresponding to the position coordinates, and the value of each position of the air conditioning load adjustable potential semantic coding difference representation matrix is ​​the absolute value of the difference between a pair of eigenvalues ​​of two positions of the air conditioning load adjustable potential semantic coding vector corresponding to the position coordinates;

[0046] Performing matrix multiplication on the transpose vector of the air-conditioning load adjustable potential semantic coding vector and the air-conditioning load adjustable potential semantic coding and representation matrix to obtain an air-conditioning load adjustable potential semantic coding and representation vector, and performing matrix multiplication on the air-conditioning load adjustable potential semantic coding difference representation matrix and the air-conditioning load adjustable potential semantic coding vector to obtain an air-conditioning load adjustable potential semantic coding difference representation vector, wherein the air-conditioning load adjustable potential semantic coding vector is a column vector;

[0047] Calculating the point sum of the air conditioning load adjustable potential semantic coding sum representation vector and the transposed vector of the air conditioning load adjustable potential semantic coding difference representation vector to obtain a first air conditioning load adjustable potential semantic coding sum difference representation vector;

[0048] Calculating the matrix product of the air conditioning load adjustable potential semantic coding sum representation matrix and the air conditioning load adjustable potential semantic coding difference representation matrix, and performing matrix multiplication on the transposed vector of the air conditioning load adjustable potential semantic coding vector and the matrix product to obtain a second air conditioning load adjustable potential semantic coding sum difference representation vector;

[0049] Calculating the point sum of the first air conditioning load adjustable potential semantic coding sum difference representation vector and the transposed vector of the second air conditioning load adjustable potential semantic coding sum difference representation vector to obtain a corrected air conditioning load adjustable potential semantic coding vector;

[0050] The corrected air-conditioning load adjustable potential semantic coding vector is passed through a decoder-based temperature upper and lower limit generator to obtain the temperature adjustment upper limit value and the temperature adjustment lower limit value.

[0051] Based on this, the air-conditioning load adjustable potential semantic coding sum representation matrix and the air-conditioning load adjustable potential semantic coding difference representation matrix used for the group aggregation statistical evaluation of the local distribution of the eigenvalue granularity of the air-conditioning load adjustable potential semantic coding vector are used as the retrieval and response distribution enhancements of the air-conditioning load adjustable potential semantic coding vector, and a reference-free distribution retrieval response framework based on the group aggregation sum-difference feature distribution open domain of the air-conditioning load adjustable potential semantic coding vector is constructed to avoid the distribution response redundancy caused by the local overflow characteristics of the air-conditioning load adjustable potential semantic coding vector through response superposition, so as to realize the faithfulness constraint of the response self-aggregation statistical correlation of the air-conditioning load adjustable potential semantic coding vector to the target decoding regression domain, and improve the accuracy of the decoding result of the air-conditioning load adjustable potential semantic coding vector obtained by the decoder-based temperature upper and lower limit generator.

[0052] In summary, according to the embodiment of the present application, a calculation method for evaluating the adjustable potential of the air-conditioning load in a substation considering the usage behavior of the terminal users is explained. It combines a deep learning algorithm, considers the differences in the usage behavior of the terminal users and the location of the air conditioners, and classifies the time series characteristic distribution of the air-conditioning load with the same user usage behavior pattern and air-conditioning location characteristics into one category through a cluster analysis method to form a matching database as benchmark information. Then, the time series correlation distribution pattern of the air-conditioning load of the single air conditioner is matched and associated with the benchmark patterns of each type in the matching database to serve as the basis for evaluating the adjustable potential of the air-conditioning load, thereby realizing the intelligent generation of the upper limit value and the lower limit value of the temperature adjustment of the air conditioner. In this way, the upper and lower limits of the adjustable potential of the air-conditioning load are obtained, providing important technical support for temperature control and optimization of energy distribution.

[0053] In addition, an embodiment of the present application may also be a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, enables the processor to execute the steps in the functions of the calculation method for evaluating the adjustable potential of the air-conditioning load in the substation considering the usage behavior of the end user according to various embodiments of the present application described in the above "Exemplary Method" section of this specification.

[0054] The computer-readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can, for example, include but is not limited to a system, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable 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 thereof.

[0055] The technical solution of this application is a calculation method and computer-readable medium for evaluating the adjustable potential of air-conditioning load in a substation considering the usage behavior of end users. Figure 4 The technical solution of the embodiment of the present invention is a method and computer-readable medium for evaluating the adjustable potential of air conditioning load in a substation, taking into account the usage behavior of end users, as follows:

[0056] Step 1: Consider the differences in usage behavior of different types of users in the low-voltage area and the location of the air conditioners, and perform cluster analysis on the air conditioning load. Residential users are divided into three categories based on working hours: regular work, non-routine work, and non-working. Within the same resident type, air conditioners are divided into two categories based on installation location: living room and bedroom. Commercial users are divided by industry; and office users are divided by working hours. Statistics are collected on the usage habits and proportions of different types of users. Based on the grouping index, air conditioning loads with similar behavior habits are grouped together. An improved iterative self-organizing clustering algorithm is used to perform cluster analysis on the initial air conditioning load curves within each group to obtain the overall power consumption patterns of each type of air conditioner.

[0057] Step 2: Based on the cluster analysis results from Step 1, taking into account user behavior and the differences in air conditioning equipment parameters, we conduct air conditioning load classification modeling by combining the first-order equivalent thermodynamic parameter model and air conditioning operating characteristics. Using Monte Carlo simulation, we randomly generate the rated power, room equivalent thermal resistance, and room equivalent heat capacity of individual air conditioners. All equipment parameters follow a normal distribution.

[0058] Step 3: For each type of air-conditioning load model in step 2, the FangerPMV thermal comfort model is introduced to obtain the indoor temperature range corresponding to the user's optimal comfort level. This range is set as the temperature adjustment margin, and the air-conditioning temperature is controlled. The air-conditioning set temperature is adjusted within the adjustable range to change the air-conditioning operating power, thereby calculating the adjustment potential of each type of user at each set temperature. The air-conditioning load models of each type of user are aggregated to obtain the upper and lower limits of the overall adjustable potential of the air-conditioning load in the substation.

[0059] The improved iterative self-organizing clustering algorithm (ISODATA) described in step 1 is specifically:

[0060] The ISODATA algorithm adds splitting and merging to the K-means clustering algorithm, solving the problem that the K-means clustering algorithm needs to customize the number of clusters, but the initial cluster center is still randomly selected. The improved ISODATA algorithm randomly selects the first cluster center in the data during the initial cluster center selection process, and calculates the shortest distance d(x i ), the probability of being selected as the next cluster center is p. The greater the distance from the existing cluster center, the greater the probability of being selected as the next cluster center. Repeat this process until all initial cluster centers are selected, so that the similarity of the selected initial centers is smaller and the clustering is clearer. Among them:

[0061]

[0062] The first-order thermal equivalent parameter air conditioning model described in step 2 is specifically:

[0063]

[0064] Where: T i is the indoor temperature, ℃; R is the indoor equivalent thermal resistance; C is the indoor equivalent heat capacity; T o is the outdoor temperature; Q AC is the cooling capacity of the air conditioner.

[0065] The upper and lower limits of the air conditioner operating temperature are as follows:

[0066] T min =T set -δ / 2

[0067] T max =T set +δ / 2

[0068] Where: T set is the temperature setting value; δ is the air conditioning temperature control hysteresis interval; T min 、T max The upper and lower limits of the air conditioner operating temperature.

[0069] In the cooling state, when the indoor temperature is higher than the upper limit of the operating temperature, the air conditioner is in operation; when the indoor temperature is lower than the lower limit of the operating temperature, the air conditioner is in the off state. Based on the intermittent characteristics of the air conditioner operation time, the corresponding switch control model is established:

[0070]

[0071] Where: Δt is the sampling time interval of switch control; s t is the on / off status of the air conditioner at time t.

[0072] The air conditioning cooling capacity and real-time power loss can be expressed as:

[0073]

[0074] Where: is the power and cooling capacity of the air conditioner at time t; P N , Q N It is the rated power and rated cooling capacity of the air conditioner.

[0075]

[0076] Where: are the indoor temperature at time t, the indoor temperature at time t+1, and the outdoor temperature respectively.

[0077] The aggregate adjustable potential model of the overall air conditioning load in step 3 is specifically:

[0078]

[0079] Where n is the total number of air conditioners; ΔP(t) is the power difference before and after changing the set temperature at time t, i.e., the control potential; The power of air conditioner i at time t before and after changing the set temperature.

[0080] The PWV model described in step 3 can be interpreted as:

[0081] The Fanger PMV model calculates and predicts a person's perceived thermal comfort level based on various building and interior parameters, such as indoor air temperature, relative humidity, air velocity, radiant temperature, and clothing insulation. The PMV value represents a person's perception of thermal comfort, ranging from -3 to +3, with 0 representing ideal thermal comfort. The PMV value can be used to guide air conditioning setpoint adjustments.

[0082] In the FangerPMV model, when all other parameters inside and outside the building are in a comfortable state, the functional relationship between the PWV value and the indoor temperature x is:

[0083]

[0084] Research shows that when the PWV value is between -0.5 and 0.5, the user is in the best comfort state, and the corresponding air conditioning set temperature adjustment range is 24℃-27℃.

[0085] The method proposed in this application can improve the shortcomings of existing research that rarely consider the impact of user usage behavior characteristics on the accuracy of air-conditioning load adjustable potential evaluation. Based on the cluster analysis of air-conditioning loads of different users, an air-conditioning load classification aggregation model is proposed, which can well grasp the air-conditioning power consumption patterns of different types of users. On the basis of considering the analysis of usage characteristics differences, it can carry out research on the calculation method for evaluating the adjustable potential of air-conditioning loads in substations considering the usage behavior of end users, improve the accuracy of the evaluation of the adjustable potential of air-conditioning loads, and provide the prerequisite for the participation of air-conditioning loads in demand response.

[0086] A specific embodiment of the present application also provides a computer-readable medium.

[0087] The computer readable medium is a server workstation;

[0088] The server workstation stores a computer program executed by an electronic device. When the computer program runs on the electronic device, the electronic device executes the steps of the method of the embodiment of the present invention.

[0089] While various embodiments of the present disclosure have been described above, the above descriptions are illustrative, non-exhaustive, and not intended to be limiting of the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or improvements to existing technologies, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for evaluating and calculating the adjustable potential of air conditioning load in a substation considering the end-user usage behavior, characterized in that: include: Get a collection of air conditioning load data; Based on the differences in user usage behaviors and air conditioner locations in the low-voltage area, cluster analysis is performed on the set of air conditioner load data to obtain a set of cluster subsets of air conditioner load data; Performing air conditioning load time series pattern feature extraction on the set of the air conditioning load data cluster subsets to obtain a set of air conditioning load data cluster subset time series pattern feature vectors; Obtain the time series of air conditioning load data of a single air conditioner; Extracting time series pattern features of the air conditioning load of the individual air conditioner from the time series of the air conditioning load data to obtain a time series associated implicit feature vector of the individual air conditioning load data; Performing a matching query on the implicit feature vector associated with the time series of the single air-conditioning load data and the set of the time series pattern feature vectors of the cluster subset of the air-conditioning load data to obtain a semantic coding vector of the adjustable potential of the air-conditioning load; The air conditioning load adjustable potential semantic coding vector is passed through a decoder-based temperature upper and lower limit generator to obtain a temperature adjustment upper limit value and a temperature adjustment lower limit value, including the following steps: Calculating an air conditioning load adjustable potential semantic coding and representation matrix and an air conditioning load adjustable potential semantic coding difference representation matrix of the air conditioning load adjustable potential semantic coding vector, wherein the value of each position of the air conditioning load adjustable potential semantic coding and representation matrix is ​​the mean of a pair of eigenvalues ​​of two positions of the air conditioning load adjustable potential semantic coding vector corresponding to the position coordinates, and the value of each position of the air conditioning load adjustable potential semantic coding difference representation matrix is ​​the absolute value of the difference between a pair of eigenvalues ​​of two positions of the air conditioning load adjustable potential semantic coding vector corresponding to the position coordinates; Performing matrix multiplication on the transpose vector of the air-conditioning load adjustable potential semantic coding vector and the air-conditioning load adjustable potential semantic coding and representation matrix to obtain an air-conditioning load adjustable potential semantic coding and representation vector, and performing matrix multiplication on the air-conditioning load adjustable potential semantic coding difference representation matrix and the air-conditioning load adjustable potential semantic coding vector to obtain an air-conditioning load adjustable potential semantic coding difference representation vector, wherein the air-conditioning load adjustable potential semantic coding vector is a column vector; Calculating the point sum of the air conditioning load adjustable potential semantic coding sum representation vector and the transposed vector of the air conditioning load adjustable potential semantic coding difference representation vector to obtain a first air conditioning load adjustable potential semantic coding sum difference representation vector; Calculating the matrix product of the air conditioning load adjustable potential semantic coding sum representation matrix and the air conditioning load adjustable potential semantic coding difference representation matrix, and performing matrix multiplication on the transposed vector of the air conditioning load adjustable potential semantic coding vector and the matrix product to obtain a second air conditioning load adjustable potential semantic coding sum difference representation vector; Calculating the point sum of the first air conditioning load adjustable potential semantic coding sum difference representation vector and the transposed vector of the second air conditioning load adjustable potential semantic coding sum difference representation vector to obtain a corrected air conditioning load adjustable potential semantic coding vector; The corrected air-conditioning load adjustable potential semantic coding vector is passed through a decoder-based temperature upper and lower limit generator to obtain the temperature adjustment upper limit value and the temperature adjustment lower limit value.

2. The method for evaluating and calculating the adjustable potential of air conditioning load in a substation considering the end-user usage behavior according to claim 1 is characterized in that: Performing air conditioning load time series pattern feature extraction on the set of the air conditioning load data cluster subsets to obtain a set of air conditioning load data cluster subset time series pattern feature vectors, including: Each air-conditioning load data cluster subset in the set of the air-conditioning load data cluster subsets is respectively passed through an air-conditioning load time series pattern feature extractor based on an RNN model to obtain a set of time series pattern feature vectors of the air-conditioning load data cluster subsets.

3. The method for evaluating and calculating the adjustable potential of air conditioning load in a substation considering the end-user usage behavior according to claim 2 is characterized in that: Extracting air conditioning load time series pattern features from the time series of the air conditioning load data of the individual air conditioner to obtain an implicit feature vector associated with the time series of the air conditioning load data of the individual air conditioner includes: The time series of the air-conditioning load data of the individual air-conditioner is passed through the air-conditioning load time series pattern feature extractor based on the RNN model to obtain the time series associated implicit feature vector of the individual air-conditioning load data.

4. The method for evaluating and calculating the adjustable potential of air conditioning load in a substation considering the end-user usage behavior according to claim 3 is characterized in that: Performing a matching query on the implicit feature vector associated with the time series of the single air-conditioning load data and the set of the time series pattern feature vectors of the cluster subset of the air-conditioning load data to obtain the semantic coding vector of the adjustable potential of the air-conditioning load, including: The implicit feature vector associated with the time series of the single air-conditioning load data is used as a query feature vector, and the query feature vector and the set of the time series pattern feature vector of the cluster subset of the air-conditioning load data are input into a feature query module based on information flow interaction to obtain the semantic coding vector of the adjustable potential of the air-conditioning load.

5. The method for evaluating and calculating the adjustable potential of air conditioning load in a substation considering the end-user usage behavior according to claim 4 is characterized in that: The implicit feature vector associated with the time series of the single air-conditioning load data is used as a query feature vector, and the query feature vector and the set of the time series pattern feature vectors of the cluster subset of the air-conditioning load data are input into a feature query module based on information flow interaction to obtain the semantic encoding vector of the adjustable potential of the air-conditioning load, including: Calculating the semantic association between each air conditioning load data cluster subset time series pattern feature vector in the set of air conditioning load data cluster subset time series pattern feature vectors and the query feature vector to obtain a sequence of semantic associations; Normalizing each semantic relevance in the sequence of semantic relevance to obtain a sequence of information interaction matching query factors; The sequence of the information interaction matching query factors is used as a weight, and the position-weighted sum of the set of time series pattern feature vectors of the cluster subset of the air-conditioning load data is calculated to obtain the semantic coding vector of the adjustable potential of the air-conditioning load.

6. The method for evaluating and calculating the adjustable potential of air conditioning load in a substation considering the end-user usage behavior according to claim 5 is characterized in that: Calculating the semantic association between each air conditioning load data cluster subset time series pattern feature vector in the set of the air conditioning load data cluster subset time series pattern feature vectors and the query feature vector to obtain a sequence of semantic associations, including: After calculating the position-by-element difference between the query feature vector and each air conditioning load data cluster subset time series pattern feature vector in the set of the air conditioning load data cluster subset time series pattern feature vectors, the norm of each obtained feature vector is calculated to obtain the sequence of the semantic association degree.

7. The method for evaluating and calculating the adjustable potential of air conditioning load in a substation considering the end-user usage behavior according to claim 6 is characterized in that: Normalizing each semantic relevance in the sequence of semantic relevance to obtain a sequence of information interaction matching query factors includes: A soft maximum normalization process based on a softmax activation function is performed on each semantic relevance in the sequence of semantic relevance to obtain a sequence of information interaction matching query factors.

8. A computer-readable medium having computer program instructions stored thereon, wherein when the computer program instructions are executed in a computer, the computer is caused to perform the method according to any one of claims 1 to 7.

Images