Large-scale variable frequency air conditioner aggregation control method and device

Abstract

The invention belongs to the technical field of intelligent power grids, and particularly relates to a large-scale variable frequency air conditioner aggregation control method and device. The method comprises the steps that variable frequency air conditioners in a to-be-evaluated time period are clustered, and a plurality of target variable frequency air conditioner clusters are obtained; the temperature adjustable margin of the target variable frequency air conditioner cluster is determined; based on the temperature adjustable margin of the target variable frequency air conditioner cluster, determining the load aggregation power adjustable margin of the large-scale variable frequency air conditioner in the to-be-evaluated time period by using a pre-established large-scale variable frequency air conditioner aggregation response potential evaluation model; and based on the time sequence large-scale variable frequency air conditioner load aggregation power adjustable margin of the time period to be evaluated, the variable frequency air conditioner cluster control strategy is used for conducting temperature control on the target variable frequency air conditioner cluster. According to the technical scheme, the temperature of the large-scale variable frequency air conditioner is accurately controlled, the temperature control efficiency of the large-scale variable frequency air conditioner is improved, and the quality of demand response is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of smart grids, and in particular relates to a large-scale variable frequency air conditioner aggregation control method and device. Background Art

[0002] In recent years, the market share of variable-frequency air conditioners (VFAs) has steadily increased. VFAs are a typical flexible load, and large-scale VFA loads can serve as a flexible resource, providing ancillary services such as peak shaving and frequency regulation to the power grid through demand-side response. According to the peak-shaving rules of the electricity market, large-scale VFAs can participate in the peak-shaving market through aggregation, with load aggregators based on demand-side response. Furthermore, to meet the control needs of large-scale VFAs, load aggregators need to group VFAs into clusters for control. The aggregate response potential of large-scale VFAs is often characterized by the upper and lower limits of their aggregate power. Considering subjective factors, various factors influence the aggregate response potential, including outdoor temperature, VFA temperature setpoint, and VFA temperature adjustment margin. Therefore, research on methods to assess the aggregate response potential of large-scale VFAs is of great significance.

[0003] Currently, research on assessing the response potential of large-scale aggregated air conditioning loads is primarily divided into model-driven and data-driven approaches. The model-driven approach uses an equivalent thermodynamic model of individual air conditioning loads and comprehensively considers user subjective factors to derive parameter values ​​that characterize the response potential of air conditioning loads. The data-driven approach, based on actual operational data of air conditioning loads participating in demand response, explores the nonlinear mapping relationship between response potential parameters and their influencing factors. Compared to model-driven approaches, data-driven approaches have better generalization capabilities. However, data-driven approaches struggle to determine the upper and lower limits of the aggregated power of variable-frequency air conditioning loads and the indoor temperatures of a large number of individual variable-frequency air conditioners. Summary of the Invention

[0004] In order to overcome the problems existing in the above-mentioned related technologies, the present invention provides a large-scale variable frequency air conditioner aggregation control method and device.

[0005] According to a first aspect of an embodiment of the present invention, a method for controlling a large-scale variable-frequency air conditioner aggregation is provided, comprising:

[0006] Cluster the variable frequency air conditioners in the evaluation period to obtain several target variable frequency air conditioner clusters;

[0007] Determining a temperature adjustable margin of the target variable frequency air conditioning cluster;

[0008] Based on the temperature adjustable margin of the target variable frequency air conditioning cluster, a pre-established large-scale variable frequency air conditioning aggregate response potential assessment model is used to determine the large-scale variable frequency air conditioning load aggregate power adjustable margin during the assessment period;

[0009] Based on the adjustable margin of the aggregated power of the large-scale variable frequency air conditioner load in the time period to be evaluated, the temperature of the target variable frequency air conditioner cluster is controlled by using the variable frequency air conditioner cluster control strategy;

[0010] The large-scale variable-frequency air-conditioning aggregate response potential assessment model is constructed using the adjustable margin of the aggregate power of large-scale variable-frequency air-conditioning loads in historical periods.

[0011] Preferably, clustering the variable frequency air conditioners in the evaluation period to obtain several target variable frequency air conditioner clusters includes:

[0012] Calculate the steady-state parameters and temperature-varying parameters of the variable-frequency air conditioner during the evaluation period;

[0013] The steady-state parameters and temperature change parameters of the variable-frequency air conditioner in the evaluation period are used as characteristic attributes, and the K-means clustering method is used to cluster the variable-frequency air conditioners in the evaluation period to obtain several cluster groups and their cluster centers, wherein the cluster groups are the target variable-frequency air conditioner clusters, and the cluster centers are characteristic attributes.

[0014] Preferably, the calculation formula for the steady-state parameters of the variable-frequency air conditioner during the period to be evaluated includes:

[0015] W1=Rη

[0016] The calculation formula for the temperature change parameter of the variable frequency air conditioner during the evaluation period includes:

[0017] W2=RC

[0018] In the above formula, W1 is the steady-state parameter of the variable-frequency air conditioner during the evaluation period, W2 is the temperature change parameter of the variable-frequency air conditioner during the evaluation period, R is the equivalent thermal resistance of the air inside the building, η is the energy efficiency ratio of the variable-frequency air conditioner, and C is the equivalent heat capacity of the air inside the building.

[0019] Preferably, the determining of the temperature adjustable margin of the target variable frequency air conditioning cluster includes:

[0020] The temperature adjustable margin adjustment amount of the target variable frequency air conditioning cluster caused by user willingness is calculated using the initial temperature adjustable upper limit and the initial temperature adjustable lower limit of the target variable frequency air conditioning cluster.

[0021] Calculating the upper and lower limits of the temperature adjustment of the target variable frequency air conditioning cluster by using the temperature adjustment margin adjustment amount caused by the user's willingness;

[0022] The upper and lower limits of the temperature adjustment of the target variable-frequency air-conditioning cluster are the temperature adjustment margin of the target variable-frequency air-conditioning cluster.

[0023] Preferably, the calculation formula for the temperature adjustable margin adjustment amount of the target variable frequency air conditioning cluster caused by the user's willingness includes:

[0024] ΔT h,t =β h,t (T max0,h,t -T min0,h,t )

[0025] The calculation formula for the temperature adjustment margin of the target variable frequency air conditioning cluster includes:

[0026]

[0027] In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,T e ], T s is the starting time of the period to be evaluated, T e is the result time of the period to be evaluated; ΔT h,t is the temperature adjustable margin adjustment amount of the hth target variable frequency air conditioning cluster caused by user willingness at time t; β h,t is the user willingness influencing factor, that is, the comprehensive willingness of the variable frequency air conditioner users of the hth target variable frequency air conditioner cluster to participate in demand response at time t; T max0,h,t is the upper limit of the initial temperature adjustment of the hth target variable frequency air conditioning cluster at time t, T min0,h,t is the adjustable lower limit of the initial temperature of the hth target variable frequency air conditioning cluster at time t, T max,h,t is the upper limit of the temperature adjustment of the hth target variable frequency air conditioning cluster at time t, T min,h,t is the adjustable lower limit of the temperature of the hth target variable frequency air conditioning cluster at time t.

[0028] Preferably, the method of determining the aggregate power adjustable margin of large-scale variable-frequency air conditioner loads during the evaluation period based on the temperature adjustable margin of the target variable-frequency air conditioner cluster and using a pre-established large-scale variable-frequency air conditioner aggregate response potential evaluation model includes:

[0029] Collect the temporal outdoor temperature of the time period to be evaluated and the target aggregate power of large-scale variable frequency air conditioning loads issued by the power grid;

[0030] The large-scale variable-frequency air conditioning aggregate response potential evaluation model is input with the temporal outdoor temperature of the time period to be evaluated, the target aggregate power of the large-scale variable-frequency air conditioning load issued by the power grid during the time period to be evaluated, and the temperature adjustment margin of the target variable-frequency air conditioning cluster. The model outputs the upper and lower limits of the adjustable aggregate power of the large-scale variable-frequency air conditioning load during the time period to be evaluated.

[0031] The upper limit and lower limit of the adjustable aggregate power of the large-scale variable-frequency air-conditioning load in the period to be evaluated are the adjustable margin of the aggregate power of the large-scale variable-frequency air-conditioning load in the period to be evaluated.

[0032] Preferably, the adjustable margin of the large-scale variable frequency air conditioner load aggregate power based on the temporal sequence of the time period to be evaluated and the use of a variable frequency air conditioner cluster control strategy to perform temperature control on a target variable frequency air conditioner cluster include:

[0033] Determine the actual aggregate power constraint of the target variable frequency air conditioning cluster by utilizing the adjustable margin of the aggregate power of the large-scale variable frequency air conditioning load in the time period to be evaluated;

[0034] Based on the constraints of the actual aggregate power of the target variable frequency air conditioning cluster, the objective function of the variable frequency air conditioning cluster control strategy is solved with the goal of minimizing the difference between the target aggregate power of the large-scale variable frequency air conditioning load and the actual aggregate power of the target variable frequency air conditioning cluster, and the optimal value of the actual aggregate power of the target variable frequency air conditioning cluster is obtained;

[0035] Calculating an indoor temperature control set value of the target variable frequency air conditioning cluster based on the optimal actual aggregate power value of the target variable frequency air conditioning cluster;

[0036] The target variable frequency air conditioning cluster is temperature controlled by using the indoor temperature control set value of the target variable frequency air conditioning cluster. Preferably, the calculation formula of the constraint condition of the actual aggregate power of the target variable frequency air conditioning cluster includes:

[0037] P samin,t ≤P sa,h,t ≤P samax,t

[0038] The calculation formula of the objective function of the variable frequency air conditioning cluster control strategy includes:

[0039]

[0040] In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,T e ], T s is the starting time of the period to be evaluated, T e is the result time of the period to be evaluated; P sa,h,t is the actual aggregate power of the hth target variable frequency air conditioning cluster at time t, P samax,t is the upper limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t, P samin,t is the adjustable lower limit of the aggregated power of large-scale variable frequency air conditioner load at time t, S is the objective function of the variable frequency air conditioner cluster control strategy, abs(.) is the absolute value operation, P vfset,tis the target aggregate power of large-scale variable frequency air conditioning load issued by the power grid at time t.

[0041] Preferably, the calculation formula for the actual aggregate power of the target variable frequency air conditioning cluster includes:

[0042]

[0043] In the above formula, δ is the user controllability, N h is the number of variable frequency air conditioners in the hth target variable frequency air conditioner cluster, T out,t is the outdoor temperature at time t, T setc,h,t is the temperature setting value of the hth target variable frequency air conditioning cluster at time t, R c,h η c,h is the characteristic attribute value of the cluster center of the h-th target variable frequency air conditioner cluster, R c,h is the steady-state parameter corresponding to the cluster center of the h-th target variable frequency air conditioner cluster, η c,h is the temperature change parameter corresponding to the cluster center of the hth target variable frequency air conditioner cluster, T in,h,t is the indoor temperature of the hth target variable frequency air conditioning cluster at time t, and k is the power variation coefficient.

[0044] Preferably, the calculation formula for the indoor temperature control setting value of the target variable frequency air conditioning cluster includes:

[0045]

[0046] In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,T e ], T s is the starting time of the period to be evaluated, T e T is the result time of the period to be evaluated; in,h,t+1 is the indoor temperature of the hth target variable frequency air conditioning cluster at time t+1, that is, the indoor temperature control set value of the target variable frequency air conditioning cluster; T out,t+1 is the outdoor temperature at time t+1, η is the energy efficiency ratio of the variable frequency air conditioner, k is the power variation coefficient, P′ sa,h,t is the optimal value of the actual aggregate power of the hth target variable frequency air conditioning cluster at time t, R is the equivalent thermal resistance of the air inside the building, T out,t is the outdoor temperature at time t, T in,h,t is the indoor temperature of the hth target variable frequency air conditioning cluster at time t, C is the equivalent heat capacity of the air inside the building, and Δt is the duration of a period.

[0047] Preferably, the process of establishing the large-scale variable frequency air conditioner aggregate response potential assessment model includes:

[0048] Cluster the variable frequency air conditioners in the historical period and obtain several historical variable frequency air conditioner clusters and their cluster centers;

[0049] Determine the adjustable margin of aggregated power of large-scale variable frequency air conditioner loads during a historical period by using the cluster center of the historical variable frequency air conditioner cluster;

[0050] Collect the time-series outdoor temperature of historical periods and the target aggregate power of large-scale variable-frequency air-conditioning loads issued by the power grid during historical periods;

[0051] Determine the temperature adjustment margin of the historical variable frequency air conditioning cluster;

[0052] Constructing a data set using the temporal outdoor temperature of the historical period, the target aggregate power of the large-scale variable frequency air conditioning load issued by the power grid during the historical period, the adjustable margin of the variable frequency air conditioning cluster temperature during the historical period, and the adjustable margin of the large-scale variable frequency air conditioning load aggregate power during the historical period;

[0053] The data set is used to train and verify the spatiotemporal convolutional network model, and after successful verification, the large-scale variable-frequency air-conditioning aggregate response potential evaluation model is obtained.

[0054] Preferably, clustering the variable frequency air conditioners in the historical period to obtain several historical variable frequency air conditioner clusters and their cluster centers includes:

[0055] Calculate the steady-state parameters and temperature variation parameters of the variable frequency air conditioner during the historical period;

[0056] The steady-state parameters and temperature change parameters of the variable-frequency air conditioner in the historical period are used as characteristic attributes, and the K-means clustering method is used to cluster the variable-frequency air conditioners in the historical period to obtain several cluster groups and their cluster centers, wherein the cluster groups are the historical variable-frequency air conditioner clusters, and the cluster centers are characteristic attributes.

[0057] Preferably, the method of using the cluster center of the historical variable frequency air conditioner cluster to determine the adjustable margin of aggregated power of large-scale variable frequency air conditioner loads in the historical period includes:

[0058] Utilizing the cluster center of the historical variable frequency air conditioner cluster, the aggregated power of the large-scale variable frequency air conditioner load in the historical period is calculated;

[0059] Calculate the adjustable upper and lower limits of the aggregated power of the large-scale variable-frequency air-conditioning load during the historical period using the aggregated power of the large-scale variable-frequency air-conditioning load during the historical period;

[0060] The upper limit and lower limit of the adjustable aggregate power of the large-scale variable-frequency air-conditioning load in the historical period are the adjustable margin of the aggregate power of the large-scale variable-frequency air-conditioning load in the historical period.

[0061] Preferably, the calculation formula for the aggregate power of large-scale variable frequency air-conditioning loads in the historical period includes:

[0062]

[0063] The calculation formula for the adjustable margin of the aggregated power of the large-scale variable frequency air-conditioning load during the historical period includes:

[0064]

[0065] In the above formula, h′∈[1,H′], H′ is the total number of historical variable frequency air conditioner clusters; t′∈[1,T′], T′ is the total time of the historical period; P′ sa,t′ is the aggregate power of large-scale variable frequency air conditioning load at time t′, N′ h is the number of variable frequency air conditioners in the h′th historical variable frequency air conditioner cluster, T out is the outdoor temperature, T setc,h′,t′ is the temperature setting value of the h′th historical variable frequency air conditioning cluster at time t′; R′ c,h and η′ c,h is the characteristic attribute value of the cluster center of the h′th historical variable frequency air conditioner cluster, R′ c,h is the steady-state parameter corresponding to the cluster center of the h′th historical variable frequency air conditioner cluster, η′ c,h is the temperature change parameter corresponding to the cluster center of the h′th historical variable frequency air conditioning cluster; P′ samax,t′ is the upper limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t′, P′ samin,t′ is the lower limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t′, δ is the user controllability, T out,t′ is the outdoor temperature at time t′, T max,h′,t′ is the upper limit of the temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, T min,h′,t′ is the adjustable lower limit of the temperature of the h′th historical variable frequency air conditioning cluster at time t′.

[0066] Preferably, the determining of the temperature adjustable margin of the historical variable frequency air conditioning cluster includes:

[0067] The temperature adjustment margin adjustment amount of the historical variable frequency air conditioning cluster caused by user willingness is calculated by using the initial temperature adjustable upper limit and the initial temperature adjustable lower limit of the historical variable frequency air conditioning cluster.

[0068] Calculate the upper and lower limits of the temperature adjustment margin of the historical variable frequency air conditioning cluster by using the temperature adjustment margin adjustment amount caused by the user's willingness;

[0069] The temperature adjustable upper limit and lower limit of the historical variable frequency air conditioning cluster are the temperature adjustable margin of the historical variable frequency air conditioning cluster.

[0070] Preferably, the calculation formula for the temperature adjustable margin adjustment amount of the historical variable frequency air conditioning cluster caused by user willingness includes:

[0071] ΔT′ h′,t′ =β′ h′,t′ (T′ max0,h′,t′ -T′ min0,h′,t′ )

[0072] The calculation formula of the temperature adjustable margin of the historical variable frequency air conditioning cluster includes:

[0073]

[0074] In the above formula, h′∈[1,H′], H′ is the total number of historical variable frequency air conditioner clusters; t′∈[1,T′], T′ is the total time of the historical period; ΔT′ h′,t′ is the temperature adjustable margin adjustment amount of the h′th historical variable frequency air conditioning cluster caused by user willingness at time t′; β′ h′,t′ is the user willingness influencing factor, that is, the comprehensive willingness of the variable frequency air conditioner users of the h′th historical variable frequency air conditioner cluster to participate in demand response at time t′; T′ max0,h′,t′ T′ is the upper limit of the initial temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, min0,h′,t′ is the lower limit of the initial temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, T′ max,h′,t′ T′ is the upper limit of the temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, min,h′,t′ is the adjustable lower limit of the temperature of the h′th historical variable frequency air conditioning cluster at time t′.

[0075] Preferably, the training and verification of the spatiotemporal convolutional network model using the dataset includes:

[0076] Normalizing the data set, and dividing the normalized data set into a training set and a test set;

[0077] The temporal outdoor temperature of the historical period in the training set, the target aggregate power of the large-scale variable-frequency air-conditioning load issued by the power grid in the historical period, and the adjustable margin of the variable-frequency air-conditioning cluster temperature in the historical period are used as input layer training samples of the spatiotemporal convolutional network model. The upper and lower limits of the adjustable aggregate power of the large-scale variable-frequency air-conditioning load in the historical period in the training set are used as output layer training samples of the spatiotemporal convolutional network model. The spatiotemporal convolutional network model is trained to obtain a trained spatiotemporal convolutional network model.

[0078] The trained spatiotemporal convolutional network model uses the temporal outdoor temperature of the historical period in the test set, the target aggregate power of the large-scale variable-frequency air-conditioning load issued by the power grid during the historical period, and the adjustable margin of the variable-frequency air-conditioning cluster temperature during the historical period as inputs, and outputs the predicted adjustable upper and lower limits of the large-scale variable-frequency air-conditioning load aggregate power;

[0079] Determining the prediction accuracy of the trained spatiotemporal convolutional network model based on the upper and lower adjustable limits of the aggregated power of the large-scale variable-frequency air-conditioning load in the historical period in the test set and the predicted upper and lower adjustable limits of the aggregated power of the large-scale variable-frequency air-conditioning load;

[0080] If the prediction accuracy is greater than or equal to the accuracy threshold, the verification is successful, and the trained spatiotemporal convolutional network model is the large-scale variable-frequency air-conditioning aggregation response potential evaluation model; otherwise, the verification fails, the parameters of the spatiotemporal convolutional network model are adjusted, and the spatiotemporal convolutional network model with adjusted parameters is retrained until the verification is successful.

[0081] According to a second aspect of an embodiment of the present invention, there is provided a large-scale variable frequency air conditioner aggregation control device, comprising:

[0082] A clustering unit is used to cluster the variable frequency air conditioners in the evaluation period to obtain several target variable frequency air conditioner clusters;

[0083] A first determining unit, configured to determine a temperature adjustable margin of the target variable frequency air conditioning cluster;

[0084] A second determining unit is configured to determine, based on the temperature adjustable margin of the target variable frequency air conditioning cluster, a large-scale variable frequency air conditioning aggregated power adjustable margin for the evaluation period using a pre-established large-scale variable frequency air conditioning aggregated response potential evaluation model;

[0085] A control unit, configured to control the temperature of a target variable frequency air conditioning cluster using a variable frequency air conditioning cluster control strategy based on the adjustable margin of the large-scale variable frequency air conditioning load aggregate power in the time period to be evaluated;

[0086] The large-scale variable-frequency air-conditioning aggregate response potential assessment model is constructed using the adjustable margin of the aggregate power of large-scale variable-frequency air-conditioning loads in historical periods.

[0087] Preferably, the clustering unit includes:

[0088] A first calculation subunit is used to calculate the steady-state parameters and temperature variation parameters of the variable frequency air conditioner during the period to be evaluated;

[0089] The first clustering subunit is used to cluster the variable-frequency air conditioners in the evaluation period using the steady-state parameters and temperature change parameters of the variable-frequency air conditioners in the evaluation period as characteristic attributes, and adopt the K-means clustering method to obtain a number of cluster groups and their cluster centers, wherein the cluster groups are the target variable-frequency air conditioner clusters, and the cluster centers are characteristic attributes.

[0090] Preferably, the calculation formula for the steady-state parameters of the variable-frequency air conditioner during the period to be evaluated includes:

[0091] W1=Rη

[0092] The calculation formula for the temperature change parameter of the variable frequency air conditioner during the evaluation period includes:

[0093] W2=RC

[0094] In the above formula, W1 is the steady-state parameter of the variable-frequency air conditioner during the evaluation period, W2 is the temperature change parameter of the variable-frequency air conditioner during the evaluation period, R is the equivalent thermal resistance of the air inside the building, η is the energy efficiency ratio of the variable-frequency air conditioner, and C is the equivalent heat capacity of the air inside the building.

[0095] Preferably, the first determining unit includes:

[0096] The second calculation subunit is configured to calculate the temperature adjustable margin adjustment amount of the target variable frequency air conditioning cluster caused by the user's willingness by using the initial temperature adjustable upper limit and the initial temperature adjustable lower limit of the target variable frequency air conditioning cluster;

[0097] a third calculation subunit, configured to calculate an upper limit and a lower limit of the temperature adjustment of the target variable frequency air conditioning cluster by using the temperature adjustment margin adjustment amount of the target variable frequency air conditioning cluster caused by the user's willingness;

[0098] The first determining subunit is configured to provide a temperature adjustable upper limit and a lower limit for the target variable frequency air conditioning cluster as a temperature adjustable margin for the target variable frequency air conditioning cluster.

[0099] Preferably, the calculation formula for the temperature adjustable margin adjustment amount of the target variable frequency air conditioning cluster caused by the user's willingness includes:

[0100] ΔT h,t =β h,t (T max0,h,t -T min0,h,t )

[0101] The calculation formula for the temperature adjustment margin of the target variable frequency air conditioning cluster includes:

[0102]

[0103] In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,T e ], T s is the starting time of the period to be evaluated, T e is the result time of the period to be evaluated; ΔT h,t is the temperature adjustable margin adjustment amount of the hth target variable frequency air conditioning cluster caused by user willingness at time t; β h,t is the user willingness influencing factor, that is, the comprehensive willingness of the variable frequency air conditioner users of the hth target variable frequency air conditioner cluster to participate in demand response at time t; T max0,h,t is the upper limit of the initial temperature adjustment of the hth target variable frequency air conditioning cluster at time t, T min0,h,t is the adjustable lower limit of the initial temperature of the hth target variable frequency air conditioning cluster at time t, T max,h,t is the upper limit of the temperature adjustment of the hth target variable frequency air conditioning cluster at time t, T min,h,t is the adjustable lower limit of the temperature of the hth target variable frequency air conditioning cluster at time t.

[0104] Preferably, the second determining unit includes:

[0105] The first acquisition subunit is used to collect the temporal outdoor temperature of the time period to be evaluated and the target aggregate power of the large-scale variable frequency air conditioning load issued by the power grid;

[0106] an output subunit, configured to use the temporal outdoor temperature of the time period to be evaluated, the target aggregate power of the large-scale variable-frequency air-conditioning load issued by the power grid during the time period to be evaluated, and the temperature adjustment margin of the target variable-frequency air-conditioning cluster as inputs to the large-scale variable-frequency air-conditioning aggregate response potential evaluation model, and output the adjustable upper and lower limits of the large-scale variable-frequency air-conditioning load aggregate power during the time period to be evaluated;

[0107] The second determining subunit is configured to provide an upper limit and a lower limit for adjusting the aggregate power of the large-scale variable-frequency air-conditioning load in the period to be evaluated as an adjustable margin for the aggregate power of the large-scale variable-frequency air-conditioning load in the period to be evaluated.

[0108] Preferably, the control unit includes:

[0109] A third determining subunit is configured to determine the actual aggregate power constraint of the target variable frequency air conditioning cluster by utilizing the adjustable margin of the aggregate power of the large-scale variable frequency air conditioning load in the time period to be evaluated;

[0110] The first acquisition subunit is configured to solve the objective function of the variable frequency air conditioning cluster control strategy based on the constraint condition of the actual aggregate power of the target variable frequency air conditioning cluster, with the goal of minimizing the difference between the target aggregate power of the large-scale variable frequency air conditioning load and the actual aggregate power of the target variable frequency air conditioning cluster, to obtain the optimal value of the actual aggregate power of the target variable frequency air conditioning cluster;

[0111] A fourth calculation subunit, configured to calculate an indoor temperature control set value of the target variable frequency air conditioning cluster according to the actual optimal aggregate power value of the target variable frequency air conditioning cluster;

[0112] The control subunit is configured to perform temperature control on the target variable frequency air conditioning cluster by using the indoor temperature control set value of the target variable frequency air conditioning cluster.

[0113] Preferably, the calculation formula for the constraint condition of the actual aggregate power of the target variable frequency air conditioning cluster includes:

[0114] P samin,t ≤P sa,h,t ≤P samax,t

[0115] The calculation formula of the objective function of the variable frequency air conditioning cluster control strategy includes:

[0116]

[0117] In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,T e ], T s is the starting time of the period to be evaluated, T e is the result time of the period to be evaluated; P sa,h,t is the actual aggregate power of the hth target variable frequency air conditioning cluster at time t, P samax,t is the upper limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t, P samin,t is the adjustable lower limit of the aggregated power of large-scale variable frequency air conditioner load at time t, S is the objective function of the variable frequency air conditioner cluster control strategy, abs(.) is the absolute value operation, P vfset,t is the target aggregate power of large-scale variable frequency air conditioning load issued by the power grid at time t.

[0118] Preferably, the calculation formula for the actual aggregate power of the target variable frequency air conditioning cluster includes:

[0119]

[0120] In the above formula, δ is the user controllability, N h is the number of variable frequency air conditioners in the hth target variable frequency air conditioner cluster, T out,t is the outdoor temperature at time t, Tsetc,h,t is the temperature setting value of the hth target variable frequency air conditioning cluster at time t, R c,h η c,h is the characteristic attribute value of the cluster center of the h-th target variable frequency air conditioner cluster, R c,h is the steady-state parameter corresponding to the cluster center of the h-th target variable frequency air conditioner cluster, η c,h is the temperature change parameter corresponding to the cluster center of the hth target variable frequency air conditioner cluster, T in,h,t is the indoor temperature of the hth target variable frequency air conditioning cluster at time t, and k is the power variation coefficient.

[0121] Preferably, the calculation formula for the indoor temperature control setting value of the target variable frequency air conditioning cluster includes:

[0122]

[0123] In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,T e ], T s is the starting time of the period to be evaluated, T e T is the result time of the period to be evaluated; in,h,t+1 is the indoor temperature of the hth target variable frequency air conditioning cluster at time t+1, that is, the indoor temperature control set value of the target variable frequency air conditioning cluster; T out,t+1 is the outdoor temperature at time t+1, η is the energy efficiency ratio of the variable frequency air conditioner, k is the power variation coefficient, P′ sa,h,t is the optimal value of the actual aggregate power of the hth target variable frequency air conditioning cluster at time t, R is the equivalent thermal resistance of the air inside the building, T out,t is the outdoor temperature at time t, T in,h,t is the indoor temperature of the hth target variable frequency air conditioning cluster at time t, C is the equivalent heat capacity of the air inside the building, and Δt is the duration of a period.

[0124] Preferably, the method further comprises: an establishing unit for establishing the large-scale variable frequency air conditioner aggregate response potential evaluation model; the establishing unit comprises:

[0125] The second clustering subunit is used to cluster the variable frequency air conditioners in the historical period to obtain several historical variable frequency air conditioner clusters and their cluster centers;

[0126] a fourth determining subunit, configured to determine an adjustable margin of aggregated power of large-scale variable frequency air conditioner loads in a historical period by using the cluster center of the historical variable frequency air conditioner cluster;

[0127] The second collection sub-unit is used to collect the time-series outdoor temperature of the historical period and the target aggregate power of the large-scale variable frequency air conditioning load issued by the power grid during the historical period;

[0128] a fifth determining subunit, configured to determine a temperature adjustable margin of a historical variable frequency air conditioning cluster;

[0129] A construction subunit is configured to construct a data set using the temporal outdoor temperature of the historical period, the target aggregate power of the large-scale variable frequency air conditioning load issued by the power grid during the historical period, the adjustable margin of the variable frequency air conditioning cluster temperature during the historical period, and the adjustable margin of the large-scale variable frequency air conditioning load aggregate power during the historical period;

[0130] The second acquisition subunit is used to train and verify the spatiotemporal convolutional network model using the data set, and obtain the large-scale variable frequency air conditioner aggregation response potential evaluation model after successful verification.

[0131] Preferably, the second clustering subunit includes:

[0132] A first calculation module is used to calculate the steady-state parameters and temperature variation parameters of the variable frequency air conditioner in a historical period;

[0133] A clustering module is used to cluster the variable-frequency air conditioners in the historical period using the steady-state parameters and temperature change parameters of the variable-frequency air conditioners in the historical period as characteristic attributes, and adopt a K-means clustering method to obtain a number of cluster groups and their cluster centers, wherein the cluster groups are the historical variable-frequency air conditioner clusters, and the cluster centers are characteristic attributes.

[0134] Preferably, the fourth determining subunit includes:

[0135] The second calculation module is used to calculate the aggregated power of large-scale variable frequency air conditioner loads in the historical period by using the cluster center of the historical variable frequency air conditioner cluster;

[0136] A third calculation module is used to calculate the adjustable upper and lower limits of the large-scale variable frequency air conditioning load aggregate power during the historical period by using the large-scale variable frequency air conditioning load aggregate power during the historical period;

[0137] The first determining module is used to determine that the upper and lower limits of the adjustable aggregate power of the large-scale variable-frequency air-conditioning load in the historical period are the adjustable margin of the aggregate power of the large-scale variable-frequency air-conditioning load in the historical period.

[0138] Preferably, the calculation formula for the aggregate power of large-scale variable frequency air-conditioning loads in the historical period includes:

[0139]

[0140] The calculation formula for the adjustable margin of the aggregated power of the large-scale variable frequency air-conditioning load during the historical period includes:

[0141]

[0142] In the above formula, h′∈[1,H′], H′ is the total number of historical variable frequency air conditioner clusters; t′∈[1,T′], T′ is the total time of the historical period; P′ sa,t′ is the aggregate power of large-scale variable frequency air conditioning load at time t′, N′ h is the number of variable frequency air conditioners in the h′th historical variable frequency air conditioner cluster, T out is the outdoor temperature, T setc,h′,t′ is the temperature setting value of the h′th historical variable frequency air conditioning cluster at time t′; R′ c,h and η′ c,h is the characteristic attribute value of the cluster center of the h′th historical variable frequency air conditioner cluster, R′ c,h is the steady-state parameter corresponding to the cluster center of the h′th historical variable frequency air conditioner cluster, η′ c,h is the temperature change parameter corresponding to the cluster center of the h′th historical variable frequency air conditioning cluster; P′ samax,t′ is the upper limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t′, P′ samin,t′ is the lower limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t′, δ is the user controllability, T out,t′ is the outdoor temperature at time t′, T max,h′,t′ is the upper limit of the temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, T min,h′,t′ is the adjustable lower limit of the temperature of the h′th historical variable frequency air conditioning cluster at time t′.

[0143] Preferably, the fifth determining subunit includes:

[0144] a fourth calculation module, configured to calculate a temperature adjustable margin adjustment amount of the historical variable frequency air conditioning cluster caused by user willingness by using the initial temperature adjustable upper limit and the initial temperature adjustable lower limit of the historical variable frequency air conditioning cluster;

[0145] a fifth calculation module, configured to calculate an upper limit and a lower limit of the temperature adjustment of the historical variable frequency air conditioning cluster by using the temperature adjustment margin adjustment amount caused by the user's willingness of the historical variable frequency air conditioning cluster;

[0146] The second determining module is configured to determine that the upper and lower limits of the temperature adjustment of the historical variable frequency air-conditioning cluster are the temperature adjustment margin of the historical variable frequency air-conditioning cluster.

[0147] Preferably, the calculation formula for the temperature adjustable margin adjustment amount of the historical variable frequency air conditioning cluster caused by user willingness includes:

[0148] ΔT′ h′,t′ =β′ h′,t′ (T′ max0,h′,t′ -T′min0,h′,t′ )

[0149] The calculation formula of the temperature adjustable margin of the historical variable frequency air conditioning cluster includes:

[0150]

[0151] In the above formula, h′∈[1,H′], H′ is the total number of historical variable frequency air conditioner clusters; t′∈[1,T′], T′ is the total time of the historical period; ΔT′ h′,t′ is the temperature adjustable margin adjustment amount of the h′th historical variable frequency air conditioning cluster caused by user willingness at time t′; β′ h′,t′ is the user willingness influencing factor, that is, the comprehensive willingness of the variable frequency air conditioner users of the h′th historical variable frequency air conditioner cluster to participate in demand response at time t′; T′ max0,h′,t′ T′ is the upper limit of the initial temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, min0,h′,t′ is the lower limit of the initial temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, T′ max,h′,t′ T′ is the upper limit of the temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, min,h′,t′ is the adjustable lower limit of the temperature of the h′th historical variable frequency air conditioning cluster at time t′.

[0152] Preferably, the second acquiring subunit includes:

[0153] A processing module, configured to normalize the data set and divide the normalized data set into a training set and a test set;

[0154] A training module is configured to use the time-series outdoor temperature of the historical period in the training set, the target aggregate power of the large-scale variable-frequency air-conditioning load issued by the power grid in the historical period, and the adjustable margin of the variable-frequency air-conditioning cluster temperature in the historical period as input layer training samples of the spatiotemporal convolutional network model, and use the adjustable upper and lower limits of the aggregate power of the large-scale variable-frequency air-conditioning load in the historical period in the training set as output layer training samples of the spatiotemporal convolutional network model to train the spatiotemporal convolutional network model, thereby obtaining a trained spatiotemporal convolutional network model;

[0155] an output module, configured to use the time-series outdoor temperature of the historical period in the test set, the target aggregate power of the large-scale variable-frequency air-conditioning load issued by the power grid during the historical period, and the adjustable margin of the variable-frequency air-conditioning cluster temperature during the historical period as inputs to the trained spatiotemporal convolutional network model, and output the predicted upper and lower limits of the adjustable aggregate power of the large-scale variable-frequency air-conditioning load;

[0156] A third determination module is used to determine the prediction accuracy of the trained spatiotemporal convolutional network model based on the upper and lower adjustable limits of the aggregate power of the large-scale variable-frequency air-conditioning load in the historical period in the test set and the predicted upper and lower adjustable limits of the aggregate power of the large-scale variable-frequency air-conditioning load;

[0157] A verification module is used to verify that if the prediction accuracy is greater than or equal to the accuracy threshold, the verification is successful, and the trained spatiotemporal convolutional network model is the large-scale variable-frequency air-conditioning aggregation response potential evaluation model; otherwise, the verification fails, the parameters of the spatiotemporal convolutional network model are adjusted, and the spatiotemporal convolutional network model with adjusted parameters is retrained until the verification is successful.

[0158] According to a third aspect of an embodiment of the present invention, there is provided an electronic device, comprising: at least one processor and a memory; the memory and the processor are connected via a bus;

[0159] The memory is used to store one or more programs;

[0160] When the one or more programs are executed by the at least one processor, the large-scale variable-frequency air conditioner aggregation control method is implemented.

[0161] According to a fourth aspect of an embodiment of the present invention, a readable storage medium is provided, on which an execution program is stored. When the execution program is executed, the large-scale variable-frequency air conditioner aggregation control method is implemented.

[0162] The technical solution provided by the present invention has the following beneficial effects:

[0163] The present invention provides a large-scale variable-frequency air-conditioning aggregation control method and device, which clusters the variable-frequency air-conditioners in the evaluation period to obtain several target variable-frequency air-conditioning clusters, determines the temperature adjustment margin of the target variable-frequency air-conditioning cluster, and determines the large-scale variable-frequency air-conditioning load aggregation power adjustment margin in the evaluation period based on the temperature adjustment margin of the target variable-frequency air-conditioning cluster using a pre-established large-scale variable-frequency air-conditioning aggregation response potential evaluation model. Combined with the subjective influencing factors of user thermal comfort, willingness and controllability, the large-scale variable-frequency air-conditioning aggregation response potential is obtained, which can ensure the accuracy of subsequent data-driven evaluation and is suitable for model-driven scenarios for large-scale variable-frequency air-conditioning aggregation response potential evaluation. Based on the temporal large-scale variable-frequency air-conditioning load aggregation power adjustment margin in the evaluation period, the variable-frequency air-conditioning cluster control strategy is used to control the temperature of the target variable-frequency air-conditioning cluster, thereby achieving precise control of the temperature of the large-scale variable-frequency air-conditioning, improving the temperature control efficiency of the large-scale variable-frequency air-conditioning, and helping to improve the quality of demand response. BRIEF DESCRIPTION OF THE DRAWINGS

[0164] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. 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 paying any creative work.

[0165] Figure 1 This is a flow chart of a large-scale variable frequency air conditioner aggregation control method provided by an embodiment of the present invention;

[0166] Figure 2 This is a diagram of the model architecture for evaluating the aggregate response potential of large-scale variable-frequency air conditioners;

[0167] Figure 3 This is a structural block diagram of a large-scale variable frequency air conditioner aggregation control device provided by an embodiment of the present invention;

[0168] Figure 4 This is a structural block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0169] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the following embodiments are part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0170] Example 1

[0171] The present invention provides a large-scale variable frequency air conditioner aggregation control method, such as Figure 1 As shown, the following steps are included:

[0172] Step 11: Cluster the variable frequency air conditioners in the evaluation period to obtain several target variable frequency air conditioner clusters;

[0173] Step 12: Determine the temperature adjustment margin of the target variable frequency air conditioning cluster;

[0174] Step 13: Based on the temperature adjustable margin of the target variable frequency air conditioner cluster, the pre-established large-scale variable frequency air conditioner aggregate response potential assessment model is used to determine the large-scale variable frequency air conditioner load aggregate power adjustable margin during the assessment period.

[0175] Step 14: Based on the adjustable margin of the aggregated power of the large-scale variable frequency air conditioner load in the time period to be evaluated, the variable frequency air conditioner cluster control strategy is used to control the temperature of the target variable frequency air conditioner cluster;

[0176] Among them, the large-scale variable-frequency air-conditioning aggregate response potential assessment model is constructed using the large-scale variable-frequency air-conditioning load aggregate power adjustable margin in historical periods.

[0177] Furthermore, step 11 includes:

[0178] Step 111: Calculate the steady-state parameters and temperature-varying parameters of the variable-frequency air conditioner during the evaluation period;

[0179] Specifically, the calculation formula for the steady-state parameters of the variable-frequency air conditioner during the evaluation period includes:

[0180] W1=Rη

[0181] The calculation formula for the temperature change parameters of the variable frequency air conditioner during the evaluation period includes:

[0182] W2=RC

[0183] In the above formula, W1 is the steady-state parameter of the variable-frequency air conditioner during the evaluation period, W2 is the temperature change parameter of the variable-frequency air conditioner during the evaluation period, R is the equivalent thermal resistance of the air inside the building, η is the energy efficiency ratio of the variable-frequency air conditioner, and C is the equivalent heat capacity of the air inside the building;

[0184] Step 112: Using the steady-state parameters and temperature variation parameters of the variable-frequency air conditioner during the evaluation period as characteristic attributes, the variable-frequency air conditioner during the evaluation period is clustered using the K-means clustering method to obtain several cluster groups and their cluster centers, where the cluster groups are the target variable-frequency air conditioner clusters and the cluster centers are the characteristic attributes.

[0185] It should be noted that the “K-means clustering method” involved in the embodiments of the present invention is well known to those skilled in the art, and therefore, its specific implementation will not be described in detail.

[0186] Furthermore, step 12 includes:

[0187] Step 121: Calculate the temperature adjustable margin adjustment amount of the target variable frequency air conditioning cluster due to user willingness using the initial temperature adjustable upper limit and the initial temperature adjustable lower limit of the target variable frequency air conditioning cluster.

[0188] Specifically, the calculation formula for the temperature adjustment margin adjustment amount of the target variable frequency air conditioning cluster caused by user willingness includes:

[0189] ΔT h,t =β h,t (T max0,h,t -T min0,h,t )

[0190] In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,Te ], T s is the starting time of the period to be evaluated, T e is the result time of the period to be evaluated, ΔT h,t is the temperature adjustable margin adjustment amount of the hth target variable frequency air conditioning cluster caused by user willingness at time t; β h,t is the user willingness influencing factor, that is, the comprehensive willingness of the variable frequency air conditioner users of the hth target variable frequency air conditioner cluster to participate in demand response at time t; T max0,h,t is the upper limit of the initial temperature adjustment of the hth target variable frequency air conditioning cluster at time t, T min0,h,t is the adjustable lower limit of the initial temperature of the hth target variable frequency air conditioning cluster at time t;

[0191] Step 122: Calculate the upper and lower limits of the temperature adjustment of the target variable frequency air conditioning cluster using the temperature adjustment margin adjustment amount caused by the user's willingness;

[0192] Specifically, the calculation formula for the temperature adjustment margin of the target variable frequency air conditioning cluster includes:

[0193]

[0194] In the above formula, T max,h,t is the upper limit of the temperature adjustment of the hth target variable frequency air conditioning cluster at time t, T min,h,t is the adjustable lower limit of the temperature of the hth target variable frequency air conditioning cluster at time t.

[0195] Step 123: The temperature adjustable upper limit and lower limit of the target variable frequency air conditioning cluster are the temperature adjustable margin of the target variable frequency air conditioning cluster.

[0196] Furthermore, step 13 includes:

[0197] Step 131: Collecting the temporal outdoor temperature of the time period to be evaluated and the target aggregate power of the large-scale variable frequency air conditioning load issued by the power grid;

[0198] Step 132: Using the temporal outdoor temperature of the time period to be evaluated, the target aggregate power of the large-scale variable-frequency air-conditioning load issued by the power grid during the time period to be evaluated, and the temperature adjustment margin of the target variable-frequency air-conditioning cluster as inputs to a large-scale variable-frequency air-conditioning aggregate response potential evaluation model, the model outputs the upper and lower limits of the adjustable aggregate power of the large-scale variable-frequency air-conditioning load during the time period to be evaluated;

[0199] Step 133: The upper and lower limits of the adjustable aggregate power of the large-scale variable-frequency air-conditioning load in the evaluation period are the adjustable margin of the aggregate power of the large-scale variable-frequency air-conditioning load in the evaluation period.

[0200] The present invention uses a large-scale variable frequency air conditioner aggregate response potential assessment model to assess the aggregate response potential of large-scale variable frequency air conditioners, and obtains the upper and lower limits of the aggregate power that can be adjusted. The load aggregator then reports these limits to the power grid dispatching center to participate in power grid regulation and provide auxiliary services such as peak load regulation and frequency regulation.

[0201] Furthermore, step 14 includes:

[0202] Step 141: using the adjustable margin of the aggregate power of the large-scale variable frequency air conditioner load in the time period to be evaluated, determine the constraint conditions of the actual aggregate power of the target variable frequency air conditioner cluster;

[0203] Specifically, the calculation formula for the constraint conditions of the actual aggregate power of the target variable frequency air conditioning cluster includes:

[0204] P samin,t ≤P sa,h,t ≤P samax,t

[0205] In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,T e ], T s is the starting time of the period to be evaluated, T e is the result time of the period to be evaluated; P sa,h,t is the actual aggregate power of the hth target variable frequency air conditioning cluster at time t, P samax,t is the upper limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t, P samin,t is the adjustable lower limit of the aggregate power of large-scale variable frequency air-conditioning load at time t;

[0206] Step 142: Based on the actual aggregate power constraint of the target variable frequency air conditioning cluster, the objective function of the variable frequency air conditioning cluster control strategy is solved with the goal of minimizing the difference between the target aggregate power of the large-scale variable frequency air conditioning load and the actual aggregate power of the target variable frequency air conditioning cluster, thereby obtaining the optimal actual aggregate power of the target variable frequency air conditioning cluster.

[0207] It should be noted that the method of "solving the objective function of the variable frequency air conditioning cluster control strategy" involved in the embodiment of the present invention is well known to those skilled in the art, and therefore, its specific implementation method will not be described in detail;

[0208] Specifically, the calculation formula of the objective function of the variable frequency air conditioning cluster control strategy includes:

[0209]

[0210] In the above formula, S is the objective function of the variable frequency air conditioning cluster control strategy, abs(.) is the absolute value operation, P vfset,tis the target aggregate power of large-scale variable frequency air conditioning load issued by the power grid at time t;

[0211] The calculation formula for the actual aggregate power of the target variable-frequency air conditioning cluster includes:

[0212]

[0213] In the above formula, δ is the user controllability, N h is the number of variable frequency air conditioners in the hth target variable frequency air conditioner cluster, T out,t is the outdoor temperature at time t, T setc,h,t is the temperature setting value of the hth target variable frequency air conditioning cluster at time t, R c,h η c,h is the characteristic attribute value of the cluster center of the h-th target variable frequency air conditioner cluster, R c,h is the steady-state parameter corresponding to the cluster center of the h-th target variable frequency air conditioner cluster, η c,h is the temperature change parameter corresponding to the cluster center of the hth target variable frequency air conditioner cluster, T in,h,t is the indoor temperature of the hth target variable frequency air conditioning cluster at time t, and k is the power variation coefficient;

[0214] Step 143: Calculate the indoor temperature control set value of the target variable frequency air conditioning cluster based on the actual optimal aggregate power value of the target variable frequency air conditioning cluster;

[0215] Specifically, the calculation formula for the indoor temperature control set value of the target variable frequency air conditioning cluster includes:

[0216]

[0217] In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,T e ], T s is the starting time of the period to be evaluated, T e T is the result time of the period to be evaluated; in,h,t+1 is the indoor temperature of the hth target variable frequency air conditioning cluster at time t+1, that is, the indoor temperature control set value of the target variable frequency air conditioning cluster; T out,t+1 is the outdoor temperature at time t+1, η is the energy efficiency ratio of the variable frequency air conditioner, k is the power variation coefficient, P′ sa,h,t is the optimal value of the actual aggregate power of the hth target variable frequency air conditioning cluster at time t, R is the equivalent thermal resistance of the air inside the building, T out,t is the outdoor temperature at time t, T in,h,t is the indoor temperature of the hth target variable frequency air conditioning cluster at time t, C is the equivalent heat capacity of the air inside the building, and Δt is the duration of a period;

[0218] Step 144: Temperature control is performed on the target variable frequency air conditioning cluster using the indoor temperature control set value of the target variable frequency air conditioning cluster.

[0219] The present invention constructs a spatiotemporal convolutional network model for evaluating the aggregated response potential of large-scale variable-frequency air conditioners, using load aggregators to participate in demand response. Based on the "offline modeling-online application" model, the model-driven results are used to train a temporal convolutional network, and the trained temporal convolutional network model is used to evaluate the aggregated response potential of large-scale variable-frequency air conditioners online. Furthermore, the method further includes: Step 10: Establishing a model for evaluating the aggregated response potential of large-scale variable-frequency air conditioners; specifically, Step 10 includes:

[0220] Step 101: clustering the variable frequency air conditioners in the historical period to obtain several historical variable frequency air conditioner clusters and their cluster centers;

[0221] Step 102: using the cluster center of the historical variable frequency air conditioner cluster, determine the adjustable margin of the aggregated power of the large-scale variable frequency air conditioner load in the historical period;

[0222] Step 103: Collecting the temporal outdoor temperature of the historical period and the target aggregate power of the large-scale variable frequency air conditioning load issued by the power grid during the historical period;

[0223] Step 104: Determine the temperature adjustable margin of the historical variable frequency air conditioning cluster;

[0224] Step 105: Construct a data set using the temporal outdoor temperature of the historical period, the target aggregate power of the large-scale variable frequency air conditioner load issued by the power grid during the historical period, the adjustable margin of the variable frequency air conditioner cluster temperature during the historical period, and the adjustable margin of the large-scale variable frequency air conditioner load aggregate power during the historical period;

[0225] Step 106: Use the dataset to train and verify the spatiotemporal convolutional network model, and obtain a large-scale variable frequency air conditioner aggregate response potential assessment model after successful verification.

[0226] This paper combines model-driven and data-driven approaches to investigate the potential for aggregated response of large-scale variable-frequency air conditioners, further implementing this method through a time-series convolutional network (TCN). TCNs are deep learning algorithms with strong processing and analysis capabilities for time series data. The nonlinear mapping relationship between the potential for aggregated response of large-scale air conditioner loads and its influencing factors exhibits a certain temporal nature, making TCNs suitable for evaluation.

[0227] Through deep learning, the present invention adopts the "offline modeling-online application" model to realize the online evaluation of the aggregate response potential of large-scale variable-frequency air conditioners. While ensuring the accuracy of the evaluation results, it improves the evaluation speed and generalization ability, which is conducive to improving the quality of demand response.

[0228] Furthermore, step 101 includes:

[0229] Step 1011: Calculate the steady-state parameters and temperature variation parameters of the variable frequency air conditioner during the historical period;

[0230] Step 1012: Using the steady-state parameters and temperature variation parameters of the variable-frequency air conditioner in the historical period as characteristic attributes, the variable-frequency air conditioner in the historical period is clustered using the K-means clustering method to obtain several cluster groups and their cluster centers, where the cluster groups are historical variable-frequency air conditioner clusters and the cluster centers are characteristic attributes.

[0231] Furthermore, step 102 includes:

[0232] Step 1021: Calculate the aggregated power of large-scale variable frequency air conditioner loads during the historical period using the cluster center of the historical variable frequency air conditioner cluster;

[0233] Specifically, the calculation formula for the aggregate power of large-scale variable-frequency air-conditioning loads during the historical period includes:

[0234]

[0235] In the above formula, h′∈[1,H′], H′ is the total number of historical variable frequency air conditioner clusters; t′∈[1,T′], T′ is the total time of the historical period; P′ sa,t′ is the aggregate power of large-scale variable frequency air conditioning load at time t′, N′ h is the number of variable frequency air conditioners in the h′th historical variable frequency air conditioner cluster, T out is the outdoor temperature, T setc,h′,t′ is the temperature setting value of the h′th historical variable frequency air conditioning cluster at time t′; R′ c,h and η′ c,h is the characteristic attribute value of the cluster center of the h′th historical variable frequency air conditioner cluster, R′ c,h is the steady-state parameter corresponding to the cluster center of the h′th historical variable frequency air conditioner cluster, η′ c,h is the temperature change parameter corresponding to the cluster center of the h′th historical variable frequency air conditioning cluster;

[0236] Step 1022: Using the aggregate power of the large-scale variable frequency air conditioning load in the historical period, calculate the adjustable upper and lower limits of the aggregate power of the large-scale variable frequency air conditioning load in the historical period;

[0237] The calculation formula for the adjustable margin of aggregated power of large-scale variable-frequency air-conditioning loads during the historical period includes:

[0238]

[0239] In the above formula, P′ samax,t′is the upper limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t′, P′ samin,t′ is the lower limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t′, δ is the user controllability, T out,t′ is the outdoor temperature at time t′, T max,h′,t′ is the upper limit of the temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, T min,h′,t′ is the adjustable lower limit of the temperature of the h′th historical variable frequency air conditioning cluster at time t′;

[0240] Step 1023: The upper and lower limits of the adjustable aggregate power of the large-scale variable-frequency air-conditioning load in the historical period are the adjustable margin of the large-scale variable-frequency air-conditioning load aggregate power in the historical period.

[0241] Furthermore, step 104 includes:

[0242] Step 1041: Calculate the temperature adjustable margin adjustment amount of the historical variable frequency air conditioning cluster caused by user willingness using the historical initial temperature adjustable upper limit and the historical initial temperature adjustable lower limit of the variable frequency air conditioning cluster;

[0243] Specifically, the calculation formula for the temperature adjustment margin adjustment amount caused by user willingness of the historical variable frequency air conditioning cluster includes:

[0244] ΔT′ h′,t′ =β′ h′,t′ (T′ max0,h′,t′ -T′ min0,h′,t′ )

[0245] In the above formula, h′∈[1,H′], H′ is the total number of historical variable frequency air conditioner clusters; t′∈[1,T′], T′ is the total time of the historical period; ΔT′ h′,t′ is the temperature adjustable margin adjustment amount of the h′th historical variable frequency air conditioning cluster caused by user willingness at time t′; β′ h′,t′ is the user willingness influencing factor, that is, the comprehensive willingness of the variable frequency air conditioner users of the h′th historical variable frequency air conditioner cluster to participate in demand response at time t′; T′ max0,h′,t′ T′ is the upper limit of the initial temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, min0,h′,t′ is the adjustable lower limit of the initial temperature of the h′th historical variable frequency air conditioning cluster at time t′;

[0246] Step 1042: Step 104 uses the temperature adjustable margin adjustment amount caused by the user's willingness of the historical variable frequency air conditioning cluster to calculate the temperature adjustable upper limit and lower limit of the historical variable frequency air conditioning cluster;

[0247] Specifically, the calculation formula for the temperature adjustment margin of the historical variable frequency air conditioning cluster includes:

[0248]

[0249] In the above formula, T′ max,h′,t′ T′ is the upper limit of the temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, min,h′,t′ is the adjustable lower limit of the temperature of the h′th historical variable frequency air conditioning cluster at time t′;

[0250] Step 1043: The temperature adjustable upper limit and lower limit of the historical variable frequency air conditioning cluster are the temperature adjustable margin of the historical variable frequency air conditioning cluster.

[0251] Furthermore, step 106 includes:

[0252] Step 1061: normalize the data set and divide the normalized data set into a training set and a test set;

[0253] It should be noted that the “normalization processing” method involved in the embodiments of the present invention is well known to those skilled in the art, and therefore, its specific implementation will not be described in detail.

[0254] Step 1062: Using the time-series outdoor temperature of the historical time period in the training set, the target aggregate power of the large-scale variable-frequency air-conditioning load issued by the power grid during the historical time period, and the adjustable margin of the variable-frequency air-conditioning cluster temperature during the historical time period as input layer training samples of the spatiotemporal convolutional network model, and using the adjustable upper and lower limits of the aggregate power of the large-scale variable-frequency air-conditioning load during the historical time period in the training set as output layer training samples of the spatiotemporal convolutional network model, the spatiotemporal convolutional network model is trained to obtain a trained spatiotemporal convolutional network model.

[0255] Step 1063: Using the time-series outdoor temperature of the historical period in the test set, the target aggregate power of the large-scale variable-frequency air-conditioning load issued by the power grid during the historical period, and the adjustable margin of the variable-frequency air-conditioning cluster temperature during the historical period as inputs to the trained spatiotemporal convolutional network model, the model outputs the predicted upper and lower limits of the adjustable aggregate power of the large-scale variable-frequency air-conditioning load.

[0256] Step 1064: Determine the prediction accuracy of the trained spatiotemporal convolutional network model based on the upper and lower adjustable limits of the aggregated power of the large-scale variable-frequency air-conditioning load in the historical period in the test set and the predicted upper and lower adjustable limits of the aggregated power of the large-scale variable-frequency air-conditioning load;

[0257] It should be noted that the method of "determining the prediction accuracy of the trained spatiotemporal convolutional network model" involved in the embodiments of the present invention is well known to those skilled in the art, and therefore, its specific implementation will not be described in detail;

[0258] Step 1065: If the prediction accuracy is greater than or equal to the accuracy threshold, the verification is successful, and the trained spatiotemporal convolutional network model is a large-scale variable-frequency air conditioner aggregate response potential assessment model; otherwise, the verification fails, and the parameters of the spatiotemporal convolutional network model are adjusted, and the spatiotemporal convolutional network model with the adjusted parameters is retrained until the verification is successful;

[0259] It should be noted that the method of "adjusting the parameters of the spatiotemporal convolutional network model" involved in the embodiment of the present invention is well known to those skilled in the art, and therefore, its specific implementation method will not be described in detail.

[0260] To further illustrate the above-mentioned large-scale variable frequency air conditioner aggregate control method, the present invention provides a specific example, including the following steps:

[0261] Step 21: Based on the clustering of large-scale variable-frequency air conditioner loads, cluster the variable-frequency air conditioner loads managed by the aggregator and establish a model-driven method for evaluating the response potential of large-scale variable-frequency air conditioner aggregation that considers subjective influencing factors;

[0262] Step 22: Construct a spatiotemporal convolutional network model for evaluating the aggregated response potential of large-scale variable-frequency air conditioners. Using the evaluation period as the time interval, the temporal outdoor temperature, the target aggregated power of large-scale variable-frequency air conditioner loads issued by the power grid, and the upper and lower limits of the adjustable temperature of the variable-frequency air conditioner cluster are used as input data samples. Using the established large-scale variable-frequency air conditioner aggregated response potential evaluation model-driven method, the temporal upper and lower limits of the adjustable power of large-scale variable-frequency air conditioner loads are obtained as output data samples for offline training of the temporal convolutional network.

[0263] Step 23: The outdoor temperature during the evaluation period and the K periods preceding it, the target aggregate power of the large-scale variable-frequency air conditioner load issued by the power grid, and the upper and lower limits of the adjustable temperature of the variable-frequency air conditioner cluster are used as inputs to the trained temporal convolutional network model to obtain the upper and lower limits of the adjustable aggregate power online.

[0264] Step 24: The load aggregator reports the adjustable upper and lower limits of the large-scale variable-frequency air conditioner load aggregate power to the power grid dispatching center. The power grid dispatching center obtains the large-scale variable-frequency air conditioner load aggregate power demand, i.e., the target aggregate power, based on the optimized dispatching model for variable-frequency air conditioners participating in demand response. The target aggregate power is then sent to the load aggregator. The load aggregator allocates power to multiple variable-frequency air conditioner clusters within its jurisdiction, with the goal of minimizing the difference between the target aggregate power and the actual aggregate power of the large-scale variable-frequency air conditioner load. Finally, the variable-frequency air conditioner cluster directly controls the temperature of the variable-frequency air conditioner units within it to achieve the control target. The resulting variable-frequency air conditioner cluster control strategy is:

[0265]

[0266] In the above formula, S is the objective function of the variable frequency air conditioning cluster control strategy, abs(.) is the absolute value operation, P vfset,t is the target aggregate power of large-scale variable frequency air conditioning load issued by the power grid during period t, P sa,h,t is the actual aggregate power of the hth variable frequency air conditioning cluster in period t, T s 、T e are the start time and the end time of the control period, T in,h,t+1 is the indoor temperature of the hth variable frequency air conditioning cluster at time t+1, that is, the indoor temperature control set value of the variable frequency air conditioning cluster.

[0267] Furthermore, step 21 includes:

[0268] Step 211: The variable frequency air conditioner is always in the on state when working, and the room temperature changes little. It can be approximately considered that: in the steady state, the indoor temperature remains unchanged and is equal to the variable frequency air conditioner temperature setting value, that is, T in,t+1 =T in,t =T set , where T in,t+1 、T in,t are the indoor temperatures at time t+1 and time t, respectively, T set is the set value of the variable frequency air conditioner temperature, and the steady-state power of a single variable frequency air conditioner is:

[0269] P s =(T out -T set ) / Rη (1)

[0270] In the above formula, P s is the steady-state power of the variable-frequency air conditioner, T out is the outdoor temperature, η is the energy efficiency ratio of the variable frequency air conditioner, and R is the equivalent thermal resistance of the air inside the building;

[0271] The power and indoor temperature changes of dynamic variable frequency air conditioner are:

[0272]

[0273] In the above formula, P d is the dynamic power of the variable frequency air conditioner; k is the power variation coefficient, which is proportional to the change in the adjusted temperature. When the set temperature rises, k < 0, and k is simplified to a constant (-0.15), otherwise k > 0; T out,t+1 、T out,t are the outdoor temperatures at time t+1 and time t respectively, C is the equivalent heat capacity of the air inside the building, and Δt is the duration of a moment;

[0274] Using the steady-state parameter Rη and the temperature variation parameter RC of variable-frequency air conditioners as characteristic attributes, the K-means clustering method was used to cluster the large-scale variable-frequency air conditioner loads managed by the load aggregator. The characteristic attribute values ​​of each cluster group and its cluster center were obtained. Each cluster group is a variable-frequency air conditioner cluster, and the aggregated power of the variable-frequency air conditioner cluster is the sum of the steady-state power of all individual variable-frequency air conditioners within it.

[0275] Step 212: After clustering into H groups, the aggregate power of the large-scale variable frequency air conditioning load at time t is:

[0276]

[0277] In the above formula, P sa,t is the aggregate power of large-scale variable frequency air conditioning load at time t; T setc,h,t is the temperature setting value of the h-th cluster group, i.e., the h-th variable frequency air conditioning cluster, at time t, which can be obtained by the variable frequency air conditioning cluster control strategy at time t-1; R c,h η c,h is the characteristic attribute value of the Rη cluster center of the hth cluster group, R c,h is the steady-state parameter corresponding to the cluster center of the h-th target variable frequency air conditioner cluster, η c,h is the temperature change parameter corresponding to the cluster center of the hth target variable frequency air conditioner cluster, N h is the number of variable frequency air conditioners in the hth variable frequency air conditioner cluster;

[0278] The aggregated response potential of large-scale variable-frequency air conditioners is characterized by the upper and lower limits of the adjustable aggregated power of large-scale variable-frequency air conditioners. Based on the aggregated power of large-scale variable-frequency air conditioners, the aggregated response potential of large-scale variable-frequency air conditioners is obtained by combining the subjective influencing factors of user thermal comfort, willingness, and controllability. First, the impact of user thermal comfort on the response potential of variable-frequency air conditioners is considered. When the user is in the optimal comfort state, the corresponding indoor temperature is 24.8℃~27.3℃, and this is set as the initial adjustable margin of the variable-frequency air conditioner temperature. Then, the user willingness model is introduced to analyze the impact of electricity price sensitivity on the user's willingness to participate in demand response. If the user's willingness to participate is high, it can be considered that he is willing to sacrifice some comfort to obtain subsidies, and the user's temperature adjustable margin increases. The variable-frequency air conditioner temperature adjustable margin changes on the initial temperature adjustable margin that meets the user's thermal comfort. The upper and lower limits of the variable-frequency air conditioner cluster temperature are expressed as:

[0279]

[0280] In the above formula, T max,h,t 、T min,h,t are the upper and lower limits of the adjustable temperature of the hth variable frequency air conditioning cluster at time t, T max0,h,t 、T min0,h,tare the upper and lower limits of the initial adjustable temperature of the h-th variable frequency air conditioning cluster at time t, β h,t is the influence factor of user willingness, that is, the comprehensive willingness of the variable frequency air conditioner users of the hth variable frequency air conditioner cluster to participate in demand response at time t, ΔT h,t is the temperature adjustable margin adjustment amount of the hth variable frequency air conditioning cluster caused by user willingness at time t;

[0281] Finally, considering the user controllability δ, combined with the large-scale variable frequency air conditioning load aggregate power and the upper and lower limits of the adjustable temperature of the air conditioning cluster, the upper and lower limits of the adjustable power of the large-scale variable frequency air conditioning load aggregate power are obtained as follows:

[0282]

[0283] In the above formula, P samax,t 、P samin,t The upper and lower limits of the aggregate power of large-scale variable frequency air-conditioning load can be adjusted at time t respectively.

[0284] Furthermore, step 22 includes:

[0285] Step 221: Construct a spatiotemporal convolutional network model for evaluating the aggregated response potential of large-scale variable-frequency air conditioners. Using outdoor temperature, the target aggregated power of large-scale variable-frequency air conditioner loads issued by the power grid, and the upper and lower limits of adjustable temperature for the variable-frequency air conditioner cluster as characteristic variables, establish a characteristic variable matrix X, and output the upper and lower limits Y for the adjustable aggregated power of large-scale variable-frequency air conditioner loads. The dilated causal convolution of the spatiotemporal convolutional network can fully exploit the mapping relationship between the upper and lower limits of the adjustable aggregated power of large-scale variable-frequency air conditioner loads and the temporal characteristic variables. The residual connection structure of the spatiotemporal convolutional network can effectively avoid the gradient vanishing and gradient exploding phenomena that occur during the evaluation of the aggregated response potential of large-scale variable-frequency air conditioners. Input the characteristic variable matrix X into the spatiotemporal convolutional network layer, integrate the upper layer information through the fully connected layer, and output Y.

[0286] Step 222: With the evaluation period scale as the time interval, the time series outdoor temperature, the large-scale variable frequency air conditioner load target aggregate power issued by the power grid, and the variable frequency air conditioner cluster temperature adjustable upper and lower limits are used as input data samples. The variable frequency air conditioner cluster temperature adjustable upper and lower limits are obtained by formula (4). The established response potential evaluation model-driven method is used to obtain the time series large-scale variable frequency air conditioner load aggregate power adjustable upper and lower limits as output data samples. The large-scale variable frequency air conditioner load target aggregate power value issued by the power grid for the first time is obtained by setting the variable frequency air conditioner cluster temperature setting value to T0℃ through the power grid dispatching center. When there is no large-scale variable frequency air conditioner load target aggregate power issued by the power grid, it means that the variable frequency air conditioner is in a non-demand response period. In this scenario, the large-scale variable frequency air conditioner load target aggregate power value is taken as the large-scale variable frequency air conditioner load aggregate power, and Y is obtained by setting the variable frequency air conditioner cluster temperature setting value to T0℃.

[0287] like Figure 2 As shown, step 223: construct a training sample set {(X i , Y i ), i=1,2,…,n}, n is the number of samples, Y i is the corresponding i-th output data sample:

[0288]

[0289] In the above formula, X i is the i-th input data sample, i.e. the i-th feature variable matrix, the column vector [x 1k ,x 2k ,...,x jk ,...,x Mk ] T is the input data sample at time k, x 11 is the first characteristic variable value at the first moment, x 12 is the first characteristic variable value at the second moment, x 1K is the first characteristic variable value at the Kth moment, x 21 is the value of the second characteristic variable at the first moment, x 22 is the second characteristic variable value at the second moment, x 2K is the first characteristic variable value at the Kth moment, x jk is the jth characteristic variable value at time k, x M1 is the Mth characteristic variable value at the first moment, x M2 is the Mth characteristic variable value at the second moment, x MKis the Mth characteristic variable value at the Kth time, j = 1, 2, ..., M, k = 1, 2, ..., K, M is the number of characteristic variables, M = 4, is the outdoor temperature, the target aggregate power of the large-scale variable-frequency air-conditioning load issued by the power grid, the upper and lower limits of the adjustable temperature of the variable-frequency air-conditioning cluster, and K is the total number of time periods;

[0290] Step 224: normalize the training samples, train the time series convolutional network offline, and obtain a trained large-scale variable frequency air conditioner aggregate response potential assessment model.

[0291] To further validate the aforementioned large-scale variable-frequency air conditioner aggregate response potential assessment model, the present invention conducted an experiment using a distribution network system containing 10,000 variable-frequency air conditioner loads. For 120 days of summer in a particular region, the variable-frequency air conditioner loads during all periods of the 120-day period were configured to participate in distribution network regulation through a load aggregator. Using the large-scale variable-frequency air conditioner aggregate response control method based on a time series convolutional network proposed in this invention, the intraday response potential of the 10,000 variable-frequency air conditioner loads in the example was evaluated. The evaluation period was set to 15 minutes, K was set to 23, and T0 was set to 25°C. The results are shown below:

[0292] 1) Based on the clustering of large-scale variable-frequency air conditioner loads, the variable-frequency air conditioner loads managed by the aggregator are clustered and classified. 10,000 variable-frequency air conditioners are clustered into 30 cluster groups, i.e., 30 variable-frequency air conditioner clusters.

[0293] 2) Obtain data for 120 days at 15-minute intervals for outdoor temperature, the target aggregate power of large-scale variable-frequency air-conditioning loads issued by the power grid, and the upper and lower limits of the adjustable temperature of the variable-frequency air-conditioning cluster, totaling 11,520 time periods. Use data from 24 consecutive time periods as a set of data samples to construct input data sample X. The upper and lower limits of the adjustable aggregate power of large-scale variable-frequency air-conditioning loads obtained by the established large-scale variable-frequency air-conditioning aggregate response potential assessment model drive method for the corresponding time period are used as output data sample Y. The input data samples are then updated in a rolling manner, with 24 time periods as the time scale unit for a set of input data and 15-minute time intervals as the time unit. After normalization, the input and output data are divided into training set samples and test set samples in a ratio of 7:3.

[0294] 3) Using the training samples, the time series convolutional network model is trained offline to obtain a trained large-scale variable frequency air conditioner aggregate response potential assessment model;

[0295] 4) Using the trained large-scale variable-frequency air conditioner aggregate response potential assessment model, the response potential of the test set sample data is evaluated. The online assessment every 15 minutes includes: the outdoor temperature during the evaluation period and the 23 periods preceding it, the target aggregate power of the large-scale variable-frequency air conditioner load issued by the power grid, and the upper and lower limits of the variable-frequency air conditioner cluster temperature adjustment. These are used as inputs to the spatiotemporal convolutional network model, and the upper and lower limits of the adjustable aggregate power of the large-scale variable-frequency air conditioner are obtained online as output by the spatiotemporal convolutional network model.

[0296] For the test set samples, the average relative error of the adjustable upper and lower limits of the aggregate power of large-scale variable-frequency air conditioners obtained by the present invention method was 6.57%, and the maximum relative error was 8.69%, compared to the model-driven method. A comparison of the response potentials obtained by the model-driven method and the present invention method for a single day of test set samples is shown in Tables 1 and 2. The evaluation results of the test set samples and Tables 1 and 2 show that, compared to the model-driven method, the average and maximum relative errors of the aggregate response potentials of large-scale variable-frequency air conditioners obtained by the present invention method are both within 10%, validating the effectiveness of the method of the present invention.

[0297] Table 1 Comparison of adjustable lower limits of variable frequency air conditioning load cluster aggregate power

[0298]

[0299] Table 2 Comparison of adjustable upper limits of variable frequency air conditioning load cluster aggregate power

[0300]

[0301] The present invention provides a large-scale variable frequency air conditioner aggregate control method and device with the following advantages:

[0302] 1) The model-driven method for evaluating the aggregated response potential of large-scale variable-frequency air conditioners, which considers subjective factors, is proposed in this paper. Based on the aggregated power of large-scale variable-frequency air conditioner loads, it combines the subjective factors of user thermal comfort, willingness, and controllability to derive the aggregated response potential of large-scale variable-frequency air conditioners. This method ensures the accuracy of subsequent data-driven evaluations and is suitable for model-driven scenarios for evaluating the aggregated response potential of large-scale variable-frequency air conditioners.

[0303] 2) The time series convolutional network method for the aggregated response potential of large-scale variable-frequency air conditioners proposed in this invention combines model-driven and data-driven methods. It adopts the "offline modeling-online application" mode, uses the model-driven results to train the time series convolutional network, and uses the trained time series convolutional network model to evaluate the aggregated response potential of large-scale variable-frequency air conditioners online. While ensuring the accuracy of the evaluation results, it improves the evaluation speed and generalization ability, which is conducive to improving the quality of demand response. It can also effectively improve the evaluation speed while ensuring the accuracy of the response potential evaluation results.

[0304] Example 2

[0305] The present invention also provides a large-scale variable frequency air conditioner aggregation control device, such as Figure 3 Shown, including:

[0306] A clustering unit is used to cluster the variable frequency air conditioners in the evaluation period to obtain several target variable frequency air conditioner clusters;

[0307] A first determining unit, configured to determine a temperature adjustable margin of a target variable frequency air conditioning cluster;

[0308] The second determining unit is configured to determine the power adjustment margin of the aggregated load of the large-scale variable-frequency air conditioner in the evaluation period based on the temperature adjustment margin of the target variable-frequency air conditioner cluster and using a pre-established large-scale variable-frequency air conditioner aggregate response potential evaluation model;

[0309] A control unit is configured to control the temperature of a target variable frequency air conditioning cluster using a variable frequency air conditioning cluster control strategy based on the adjustable margin of power aggregation of large-scale variable frequency air conditioning loads in a time period to be evaluated;

[0310] Among them, the large-scale variable-frequency air-conditioning aggregate response potential assessment model is constructed using the large-scale variable-frequency air-conditioning load aggregate power adjustable margin in historical periods.

[0311] Furthermore, the clustering unit includes:

[0312] A first calculation subunit is used to calculate the steady-state parameters and temperature variation parameters of the variable frequency air conditioner during the period to be evaluated;

[0313] The first clustering subunit is used to cluster the variable-frequency air conditioners in the evaluation period using the steady-state parameters and temperature change parameters of the variable-frequency air conditioners in the evaluation period as characteristic attributes, and adopt the K-means clustering method to obtain several cluster groups and their cluster centers, wherein the cluster group is the target variable-frequency air conditioner cluster and the cluster center is the characteristic attribute.

[0314] Furthermore, the calculation formula for the steady-state parameters of the variable-frequency air conditioner during the evaluation period includes:

[0315] W1=Rη

[0316] The calculation formula for the temperature change parameters of the variable frequency air conditioner during the evaluation period includes:

[0317] W2=RC

[0318] In the above formula, W1 is the steady-state parameter of the variable-frequency air conditioner during the evaluation period, W2 is the temperature change parameter of the variable-frequency air conditioner during the evaluation period, R is the equivalent thermal resistance of the air inside the building, η is the energy efficiency ratio of the variable-frequency air conditioner, and C is the equivalent heat capacity of the air inside the building.

[0319] Furthermore, the first determining unit includes:

[0320] The second calculation subunit is configured to calculate the temperature adjustable margin adjustment amount of the target variable frequency air conditioning cluster caused by the user's willingness by using the initial temperature adjustable upper limit and the initial temperature adjustable lower limit of the target variable frequency air conditioning cluster;

[0321] The third calculation subunit is configured to calculate the upper and lower limits of the temperature adjustment of the target variable frequency air conditioning cluster by using the temperature adjustment margin adjustment amount caused by the user's willingness.

[0322] The first determining subunit is configured to provide a temperature adjustable upper limit and a lower limit for the target variable frequency air-conditioning cluster as a temperature adjustable margin for the target variable frequency air-conditioning cluster.

[0323] Furthermore, the calculation formula for the temperature adjustable margin adjustment amount of the target variable frequency air conditioning cluster caused by the user's willingness includes:

[0324] ΔT h,t =β h,t (T max0,h,t -T min0,h,t )

[0325] The calculation formula for the temperature adjustment margin of the target variable frequency air conditioning cluster includes:

[0326]

[0327] In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,T e ], T s is the starting time of the period to be evaluated, T e is the result time of the period to be evaluated; ΔT h,t is the temperature adjustable margin adjustment amount of the hth target variable frequency air conditioning cluster caused by user willingness at time t; β h,t is the user willingness influencing factor, that is, the comprehensive willingness of the variable frequency air conditioner users of the hth target variable frequency air conditioner cluster to participate in demand response at time t; T max0,h,t is the upper limit of the initial temperature adjustment of the hth target variable frequency air conditioning cluster at time t, T min0,h,t is the adjustable lower limit of the initial temperature of the hth target variable frequency air conditioning cluster at time t, T max,h,t is the upper limit of the temperature adjustment of the hth target variable frequency air conditioning cluster at time t, T min,h,t is the adjustable lower limit of the temperature of the hth target variable frequency air conditioning cluster at time t.

[0328] Furthermore, the second determining unit includes:

[0329] The first acquisition subunit is used to collect the temporal outdoor temperature of the time period to be evaluated and the target aggregate power of the large-scale variable frequency air conditioning load issued by the power grid;

[0330] An output subunit is configured to use the temporal outdoor temperature of the time period to be evaluated, the target aggregate power of the large-scale variable-frequency air-conditioning load issued by the power grid during the time period to be evaluated, and the temperature adjustment margin of the target variable-frequency air-conditioning cluster as inputs to a large-scale variable-frequency air-conditioning aggregate response potential assessment model, and output the upper and lower limits of the adjustable aggregate power of the large-scale variable-frequency air-conditioning load during the time period to be evaluated;

[0331] The second determining subunit is configured to set the upper and lower adjustable limits of the aggregated power of the large-scale variable-frequency air-conditioning load in the evaluation period as the adjustable margin of the aggregated power of the large-scale variable-frequency air-conditioning load in the evaluation period.

[0332] Furthermore, the control unit includes:

[0333] The third determining subunit is configured to determine the actual aggregate power constraint of the target variable frequency air conditioning cluster by utilizing the adjustable margin of the aggregate power of the large-scale variable frequency air conditioning load in the time period to be evaluated;

[0334] The first acquisition subunit is used to solve the objective function of the variable frequency air conditioning cluster control strategy based on the constraint condition of the actual aggregate power of the target variable frequency air conditioning cluster, with the goal of minimizing the difference between the target aggregate power of the large-scale variable frequency air conditioning load and the actual aggregate power of the target variable frequency air conditioning cluster, and obtain the optimal value of the actual aggregate power of the target variable frequency air conditioning cluster;

[0335] A fourth calculation subunit, configured to calculate an indoor temperature control set value of the target variable frequency air conditioning cluster according to the actual optimal aggregate power value of the target variable frequency air conditioning cluster;

[0336] The control subunit is used to control the temperature of the target variable frequency air conditioning cluster by using the indoor temperature control set value of the target variable frequency air conditioning cluster.

[0337] Furthermore, the calculation formula for the constraint conditions of the actual aggregate power of the target variable frequency air conditioning cluster includes:

[0338] P samin,t ≤P sa,h,t ≤P samax,t

[0339] The calculation formula of the objective function of the variable frequency air conditioning cluster control strategy includes:

[0340]

[0341] In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,Te ], T s is the starting time of the period to be evaluated, T e is the result time of the period to be evaluated; P sa,h,t is the actual aggregate power of the hth target variable frequency air conditioning cluster at time t, P samax,t is the upper limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t, P samin,t is the adjustable lower limit of the aggregated power of large-scale variable frequency air conditioner load at time t, S is the objective function of the variable frequency air conditioner cluster control strategy, abs(.) is the absolute value operation, P vfset,t is the target aggregate power of large-scale variable frequency air conditioning load issued by the power grid at time t.

[0342] Furthermore, the calculation formula for the actual aggregate power of the target variable frequency air conditioning cluster includes:

[0343]

[0344] In the above formula, δ is the user controllability, N h is the number of variable frequency air conditioners in the hth target variable frequency air conditioner cluster, T out,t is the outdoor temperature at time t, T setc,h,t is the temperature setting value of the hth target variable frequency air conditioning cluster at time t, R c,h η c,h is the characteristic attribute value of the cluster center of the h-th target variable frequency air conditioner cluster, R c,h is the steady-state parameter corresponding to the cluster center of the h-th target variable frequency air conditioner cluster, η c,h is the temperature change parameter corresponding to the cluster center of the hth target variable frequency air conditioner cluster, T in,h,t is the indoor temperature of the hth target variable frequency air conditioning cluster at time t, and k is the power variation coefficient.

[0345] Furthermore, the calculation formula for the indoor temperature control set value of the target variable frequency air conditioning cluster includes:

[0346]

[0347] In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,T e ], T s is the starting time of the period to be evaluated, T e T is the result time of the period to be evaluated; in,h,t+1 is the indoor temperature of the hth target variable frequency air conditioning cluster at time t+1, that is, the indoor temperature control set value of the target variable frequency air conditioning cluster; T out,t+1 is the outdoor temperature at time t+1, η is the energy efficiency ratio of the variable frequency air conditioner, k is the power variation coefficient, P′ sa,h,tis the optimal value of the actual aggregate power of the hth target variable frequency air conditioning cluster at time t, R is the equivalent thermal resistance of the air inside the building, T out,t is the outdoor temperature at time t, T in,h,t is the indoor temperature of the hth target variable frequency air conditioning cluster at time t, C is the equivalent heat capacity of the air inside the building, and Δt is the duration of a period.

[0348] Furthermore, the method further includes: establishing a unit for establishing a large-scale variable frequency air conditioner aggregate response potential assessment model; the establishing unit includes:

[0349] The second clustering subunit is used to cluster the variable frequency air conditioners in the historical period to obtain several historical variable frequency air conditioner clusters and their cluster centers;

[0350] The fourth determining subunit is used to determine the adjustable margin of aggregated power of large-scale variable frequency air conditioner loads in a historical period by using the cluster center of the historical variable frequency air conditioner cluster;

[0351] The second collection sub-unit is used to collect the time-series outdoor temperature of the historical period and the target aggregate power of the large-scale variable frequency air conditioning load issued by the power grid during the historical period;

[0352] a fifth determining subunit, configured to determine a temperature adjustable margin of a historical variable frequency air conditioning cluster;

[0353] Constructing a subunit for constructing a data set using the temporal outdoor temperature of the historical period, the target aggregate power of the large-scale variable frequency air conditioning load issued by the power grid during the historical period, the adjustable margin of the variable frequency air conditioning cluster temperature during the historical period, and the adjustable margin of the large-scale variable frequency air conditioning load aggregate power during the historical period;

[0354] The second acquisition subunit is used to train and verify the spatiotemporal convolutional network model using the data set, and obtain a large-scale variable-frequency air-conditioning aggregation response potential evaluation model after successful verification.

[0355] Furthermore, the second clustering subunit includes:

[0356] A first calculation module is used to calculate the steady-state parameters and temperature variation parameters of the variable frequency air conditioner in a historical period;

[0357] The clustering module is used to cluster the variable-frequency air conditioners in the historical period with the steady-state parameters and temperature change parameters of the variable-frequency air conditioners as characteristic attributes, using the K-means clustering method to obtain several cluster groups and their cluster centers, among which the cluster groups are historical variable-frequency air conditioner clusters and the cluster centers are characteristic attributes.

[0358] Furthermore, the fourth determining subunit includes:

[0359] The second calculation module is used to calculate the aggregated power of large-scale variable frequency air conditioner loads in the historical period by using the cluster center of the historical variable frequency air conditioner cluster;

[0360] The third calculation module is used to calculate the adjustable upper and lower limits of the large-scale variable frequency air conditioning load aggregate power during the historical period by using the large-scale variable frequency air conditioning load aggregate power during the historical period;

[0361] The first determining module is used for the upper and lower limits of the adjustable aggregate power of the large-scale variable frequency air-conditioning load in the historical period as the adjustable margin of the large-scale variable frequency air-conditioning load aggregate power in the historical period.

[0362] Furthermore, the calculation formula for the aggregate power of large-scale variable frequency air conditioning loads in the historical period includes:

[0363]

[0364] The calculation formula for the adjustable margin of aggregated power of large-scale variable-frequency air-conditioning loads during the historical period includes:

[0365]

[0366] In the above formula, h′∈[1,H′], H′ is the total number of historical variable frequency air conditioner clusters; t′∈[1,T′], T′ is the total time of the historical period; P′ sa,t′ is the aggregate power of large-scale variable frequency air conditioning load at time t′, N′ h is the number of variable frequency air conditioners in the h′th historical variable frequency air conditioner cluster, T out is the outdoor temperature, T setc,h′,t′ is the temperature setting value of the h′th historical variable frequency air conditioning cluster at time t′; R′ c,h and η′ c,h is the characteristic attribute value of the cluster center of the h′th historical variable frequency air conditioner cluster, R′ c,h is the steady-state parameter corresponding to the cluster center of the h′th historical variable frequency air conditioner cluster, η′ c,h is the temperature change parameter corresponding to the cluster center of the h′th historical variable frequency air conditioning cluster; P′ samax,t′ is the upper limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t′, P′ samin,t′ is the lower limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t′, δ is the user controllability, T out,t′ is the outdoor temperature at time t′, T max,h′,t′ is the upper limit of the temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, T min,h′,t′ is the adjustable lower limit of the temperature of the h′th historical variable frequency air conditioning cluster at time t′.

[0367] Furthermore, the fifth determining subunit includes:

[0368] a fourth calculation module, configured to calculate a temperature adjustable margin adjustment amount of the historical variable frequency air conditioning cluster caused by user willingness by using the initial temperature adjustable upper limit and the initial temperature adjustable lower limit of the historical variable frequency air conditioning cluster;

[0369] A fifth calculation module is used to calculate the upper and lower limits of the temperature adjustment margin of the historical variable frequency air conditioning cluster by using the temperature adjustment margin adjustment amount caused by the user's willingness;

[0370] The second determining module is used for the temperature adjustable upper limit and lower limit of the historical variable frequency air-conditioning cluster as the temperature adjustable margin of the historical variable frequency air-conditioning cluster.

[0371] Furthermore, the calculation formula for the temperature adjustable margin adjustment amount caused by the user's willingness of the historical variable frequency air conditioning cluster includes:

[0372] ΔT′ h′,t′ =β′ h′,t′ (T′ max0,h′,t′ -T′ min0,h′,t′ )

[0373] The calculation formula for the temperature adjustment margin of the historical variable frequency air conditioning cluster includes:

[0374]

[0375] In the above formula, h′∈[1,H′], H′ is the total number of historical variable frequency air conditioner clusters; t′∈[1,T′], T′ is the total time of the historical period; ΔT′ h′,t′ is the temperature adjustable margin adjustment amount of the h′th historical variable frequency air conditioning cluster caused by user willingness at time t′; β′ h′,t′ is the user willingness influencing factor, that is, the comprehensive willingness of the variable frequency air conditioner users of the h′th historical variable frequency air conditioner cluster to participate in demand response at time t′; T′ max0,h′,t′ T′ is the upper limit of the initial temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, min0,h′,t′ is the lower limit of the initial temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, T′ max,h′,t′ T′ is the upper limit of the temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, min,h′,t′ is the adjustable lower limit of the temperature of the h′th historical variable frequency air conditioning cluster at time t′.

[0376] Furthermore, the second acquisition subunit includes:

[0377] A processing module is used to normalize the data set and divide the normalized data set into a training set and a test set;

[0378] A training module is used to train the spatiotemporal convolutional network model using the temporal outdoor temperature of the historical time period in the training set, the target aggregate power of the large-scale variable-frequency air-conditioning load issued by the power grid during the historical time period, and the adjustable margin of the variable-frequency air-conditioning cluster temperature during the historical time period as input layer training samples of the spatiotemporal convolutional network model, and using the adjustable upper and lower limits of the aggregate power of the large-scale variable-frequency air-conditioning load during the historical time period in the training set as output layer training samples of the spatiotemporal convolutional network model to obtain a trained spatiotemporal convolutional network model;

[0379] The output module is used to use the time-series outdoor temperature of the historical period in the test set, the target aggregate power of the large-scale variable-frequency air-conditioning load issued by the power grid during the historical period, and the adjustable margin of the variable-frequency air-conditioning cluster temperature during the historical period as inputs to the trained spatiotemporal convolutional network model, and output the predicted upper and lower limits of the adjustable aggregate power of the large-scale variable-frequency air-conditioning load;

[0380] A third determination module is used to determine the prediction accuracy of the trained spatiotemporal convolutional network model based on the upper and lower adjustable limits of the aggregate power of the large-scale variable-frequency air-conditioning load in the historical period in the test set and the predicted upper and lower adjustable limits of the aggregate power of the large-scale variable-frequency air-conditioning load;

[0381] The verification module is used to verify that if the prediction accuracy is greater than or equal to the accuracy threshold, the verification is successful and the trained spatiotemporal convolutional network model is a large-scale variable-frequency air-conditioning aggregation response potential evaluation model; otherwise, the verification fails, the parameters of the spatiotemporal convolutional network model are adjusted, and the spatiotemporal convolutional network model with adjusted parameters is retrained until the verification is successful.

[0382] It can be understood that the device embodiment provided above corresponds to the method embodiment above, and the corresponding specific contents can be referenced to each other and will not be repeated here.

[0383] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0384] Example 3

[0385] like Figure 4 As shown, the present invention also provides an electronic device, which may be a computer, a single-chip microcomputer, a smart mobile device, or the like. The electronic device in this embodiment may include a processor, a memory, a transceiver component, and the like. The memory, processor, and transceiver component are connected via a bus; the memory may be used to store an execution program, which may include instructions; and the processor may be used to execute the instructions stored in the memory. The memory may also be used to store data, which may be accessed and / or modified during the execution of the instructions.

[0386] The processor may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the storage medium to implement the corresponding method flow or corresponding function, so as to implement the steps of a large-scale variable-frequency air-conditioning aggregation control method in the above embodiment.

[0387] Example 4

[0388] Based on the same inventive concept, the present invention also provides a readable storage medium, specifically an electronic device readable storage medium (Memory), which is a memory device in an electronic device for storing programs and data. It can be understood that the storage medium here can include both built-in storage media in the electronic device and, of course, extended storage media supported by the electronic device. The storage medium provides a storage space that stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space. These instructions can be one or more execution programs (including program codes). It should be noted that the storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor loads and executes one or more instructions stored in the storage medium, which can implement the steps of a large-scale variable frequency air conditioning aggregation control method in the above embodiment.

[0389] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0390] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0391] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0392] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0393] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A large-scale variable frequency air conditioner aggregation control method, characterized in that: include: Cluster the variable frequency air conditioners in the evaluation period to obtain several target variable frequency air conditioner clusters; Determining a temperature adjustable margin of the target variable frequency air conditioning cluster; Based on the temperature adjustable margin of the target variable frequency air conditioning cluster, a pre-established large-scale variable frequency air conditioning aggregate response potential assessment model is used to determine the large-scale variable frequency air conditioning load aggregate power adjustable margin during the assessment period; Based on the adjustable margin of the aggregated power of the large-scale variable frequency air conditioner load in the time period to be evaluated, the temperature of the target variable frequency air conditioner cluster is controlled by using the variable frequency air conditioner cluster control strategy; The large-scale variable-frequency air-conditioning aggregate response potential assessment model is constructed using the adjustable margin of the aggregate power of large-scale variable-frequency air-conditioning loads in historical periods.

2. The method according to claim 1, characterized in that The variable frequency air conditioners in the evaluation period are clustered to obtain several target variable frequency air conditioner clusters, including: Calculate the steady-state parameters and temperature-varying parameters of the variable-frequency air conditioner during the evaluation period; The steady-state parameters and temperature change parameters of the variable-frequency air conditioner in the evaluation period are used as characteristic attributes, and the K-means clustering method is used to cluster the variable-frequency air conditioners in the evaluation period to obtain several cluster groups and their cluster centers, wherein the cluster groups are the target variable-frequency air conditioner clusters, and the cluster centers are characteristic attributes.

3. The method according to claim 2, characterized in that The calculation formula for the steady-state parameters of the variable-frequency air conditioner during the evaluation period includes: W1=Rη The calculation formula for the temperature change parameter of the variable frequency air conditioner during the evaluation period includes: W2=RC In the above formula, W1 is the steady-state parameter of the variable-frequency air conditioner during the evaluation period, W2 is the temperature change parameter of the variable-frequency air conditioner during the evaluation period, R is the equivalent thermal resistance of the air inside the building, η is the energy efficiency ratio of the variable-frequency air conditioner, and C is the equivalent heat capacity of the air inside the building.

4. The method according to claim 1, wherein Determining the temperature adjustable margin of the target variable frequency air conditioning cluster includes: The temperature adjustable margin adjustment amount of the target variable frequency air conditioning cluster caused by user willingness is calculated using the initial temperature adjustable upper limit and the initial temperature adjustable lower limit of the target variable frequency air conditioning cluster. Calculating the upper and lower limits of the temperature adjustment of the target variable frequency air conditioning cluster by using the temperature adjustment margin adjustment amount caused by the user's willingness; The upper and lower limits of the temperature adjustment of the target variable-frequency air-conditioning cluster are the temperature adjustment margin of the target variable-frequency air-conditioning cluster.

5. The method according to claim 4, characterized in that The calculation formula for the temperature adjustable margin adjustment amount of the target variable frequency air conditioning cluster caused by the user's willingness includes: ΔT h,t =β h,t (T max0,h,t -T min0,h,t ) The calculation formula for the temperature adjustment margin of the target variable frequency air conditioning cluster includes: In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,T e ], T s is the starting time of the period to be evaluated, T e is the result time of the period to be evaluated; ΔT h,t is the temperature adjustable margin adjustment amount of the hth target variable frequency air conditioning cluster caused by user willingness at time t; β h,t is the user willingness influencing factor, that is, the comprehensive willingness of the variable frequency air conditioner users of the hth target variable frequency air conditioner cluster to participate in demand response at time t; T max0,h,t is the upper limit of the initial temperature adjustment of the hth target variable frequency air conditioning cluster at time t, T min0,h,t is the adjustable lower limit of the initial temperature of the hth target variable frequency air conditioning cluster at time t, T max,h,t is the upper limit of the temperature adjustment of the hth target variable frequency air conditioning cluster at time t, T min,h,t is the adjustable lower limit of the temperature of the hth target variable frequency air conditioning cluster at time t.

6. The method according to claim 1, wherein The method of determining the adjustable power margin of the aggregated load of large-scale variable-frequency air conditioners in the evaluation period based on the temperature adjustable margin of the target variable-frequency air conditioner cluster and using a pre-established large-scale variable-frequency air conditioner aggregate response potential evaluation model includes: Collect the temporal outdoor temperature of the time period to be evaluated and the target aggregate power of large-scale variable frequency air conditioning loads issued by the power grid; The large-scale variable-frequency air conditioning aggregate response potential evaluation model is input with the temporal outdoor temperature of the time period to be evaluated, the target aggregate power of the large-scale variable-frequency air conditioning load issued by the power grid during the time period to be evaluated, and the temperature adjustment margin of the target variable-frequency air conditioning cluster. The model outputs the upper and lower limits of the adjustable aggregate power of the large-scale variable-frequency air conditioning load during the time period to be evaluated. The upper limit and lower limit of the adjustable aggregate power of the large-scale variable-frequency air-conditioning load in the period to be evaluated are the adjustable margin of the aggregate power of the large-scale variable-frequency air-conditioning load in the period to be evaluated.

7. The method according to claim 1, characterized in that The large-scale variable frequency air conditioning load aggregate power adjustable margin based on the temporal nature of the time period to be evaluated, and the variable frequency air conditioning cluster control strategy are used to perform temperature control on the target variable frequency air conditioning cluster, including: Determine the actual aggregate power constraint of the target variable frequency air conditioning cluster by utilizing the adjustable margin of the aggregate power of the large-scale variable frequency air conditioning load in the time period to be evaluated; Based on the constraints of the actual aggregate power of the target variable frequency air conditioning cluster, the objective function of the variable frequency air conditioning cluster control strategy is solved with the goal of minimizing the difference between the target aggregate power of the large-scale variable frequency air conditioning load and the actual aggregate power of the target variable frequency air conditioning cluster, and the optimal value of the actual aggregate power of the target variable frequency air conditioning cluster is obtained; Calculating an indoor temperature control set value of the target variable frequency air conditioning cluster based on the optimal actual aggregate power value of the target variable frequency air conditioning cluster; The target variable frequency air conditioning cluster is temperature controlled by using the indoor temperature control set value of the target variable frequency air conditioning cluster.

8. The method according to claim 7, characterized in that The calculation formula for the constraint condition of the actual aggregate power of the target variable frequency air conditioning cluster includes: P samin,t ≤P sa,h,t ≤P samax,t The calculation formula of the objective function of the variable frequency air conditioning cluster control strategy includes: In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,T e ], T s is the starting time of the period to be evaluated, T e is the result time of the period to be evaluated; P sa,h,t is the actual aggregate power of the hth target variable frequency air conditioning cluster at time t, P samax,t is the upper limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t, P samin,t is the adjustable lower limit of the aggregated power of large-scale variable frequency air conditioner load at time t, S is the objective function of the variable frequency air conditioner cluster control strategy, abs(.) is the absolute value operation, P vfset,t is the target aggregate power of large-scale variable frequency air conditioning load issued by the power grid at time t.

9. The method according to claim 8, characterized in that The calculation formula for the actual aggregate power of the target variable frequency air conditioning cluster includes: In the above formula, δ is the user controllability, N h is the number of variable frequency air conditioners in the hth target variable frequency air conditioner cluster, T out,t is the outdoor temperature at time t, T setc,h,t is the temperature setting value of the hth target variable frequency air conditioning cluster at time t, R c,h η c,h is the characteristic attribute value of the cluster center of the h-th target variable frequency air conditioner cluster, R c,h is the steady-state parameter corresponding to the cluster center of the h-th target variable frequency air conditioner cluster, η c,h is the temperature change parameter corresponding to the cluster center of the hth target variable frequency air conditioner cluster, T in,h,t is the indoor temperature of the hth target variable frequency air conditioning cluster at time t, and k is the power variation coefficient.

10. The method according to claim 7, characterized in that The calculation formula for the indoor temperature control set value of the target variable frequency air conditioning cluster includes: In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,T e ], T s is the starting time of the period to be evaluated, T e T is the result time of the period to be evaluated; in,h,t+1 is the indoor temperature control set value of the hth target variable frequency air conditioning cluster at time t+1, that is, the indoor temperature control set value of the target variable frequency air conditioning cluster; T out,t+1 is the outdoor temperature at time t+1, η is the energy efficiency ratio of the variable frequency air conditioner, k is the power variation coefficient, P′ sa,h,t is the optimal value of the actual aggregate power of the hth target variable frequency air conditioning cluster at time t, R is the equivalent thermal resistance of the air inside the building, T out,t is the outdoor temperature at time t, T in,h,t is the indoor temperature of the hth target variable frequency air conditioning cluster at time t, C is the equivalent heat capacity of the air inside the building, and Δt is the duration of a period.

11. The method according to claim 1, wherein The process of establishing the large-scale variable frequency air conditioner aggregate response potential assessment model includes: Cluster the variable frequency air conditioners in the historical period and obtain several historical variable frequency air conditioner clusters and their cluster centers; Determine the adjustable margin of aggregated power of large-scale variable frequency air conditioner loads during a historical period by using the cluster center of the historical variable frequency air conditioner cluster; Collect the time-series outdoor temperature of historical periods and the target aggregate power of large-scale variable-frequency air-conditioning loads issued by the power grid during historical periods; Determine the temperature adjustment margin of the historical variable frequency air conditioning cluster; Constructing a data set using the temporal outdoor temperature of the historical period, the target aggregate power of the large-scale variable frequency air conditioning load issued by the power grid during the historical period, the adjustable margin of the variable frequency air conditioning cluster temperature during the historical period, and the adjustable margin of the large-scale variable frequency air conditioning load aggregate power during the historical period; The data set is used to train and verify the spatiotemporal convolutional network model, and after successful verification, the large-scale variable-frequency air-conditioning aggregate response potential evaluation model is obtained.

12. The method according to claim 11, characterized in that The variable frequency air conditioners in the historical period are clustered to obtain several historical variable frequency air conditioner clusters and their cluster centers, including: Calculate the steady-state parameters and temperature variation parameters of the variable frequency air conditioner during the historical period; The steady-state parameters and temperature change parameters of the variable-frequency air conditioner in the historical period are used as characteristic attributes, and the K-means clustering method is used to cluster the variable-frequency air conditioners in the historical period to obtain several cluster groups and their cluster centers, wherein the cluster groups are the historical variable-frequency air conditioner clusters, and the cluster centers are characteristic attributes.

13. The method according to claim 11, characterized in that The method of using the cluster center of the historical variable frequency air conditioner cluster to determine the adjustable margin of aggregated power of large-scale variable frequency air conditioner loads in the historical period includes: Utilizing the cluster center of the historical variable frequency air conditioner cluster, the aggregated power of the large-scale variable frequency air conditioner load in the historical period is calculated; Calculate the adjustable upper and lower limits of the aggregated power of the large-scale variable-frequency air-conditioning load during the historical period using the aggregated power of the large-scale variable-frequency air-conditioning load during the historical period; The upper limit and lower limit of the adjustable aggregate power of the large-scale variable-frequency air-conditioning load in the historical period are the adjustable margin of the aggregate power of the large-scale variable-frequency air-conditioning load in the historical period.

14. The method according to claim 13, characterized in that The calculation formula for the aggregate power of large-scale variable frequency air conditioning loads during the historical period includes: The calculation formula for the adjustable margin of the aggregated power of the large-scale variable frequency air-conditioning load during the historical period includes: In the above formula, h′∈[1,H′], H′ is the total number of historical variable frequency air conditioner clusters; t′∈[1,T′], T′ is the total time of the historical period; P′ sa,t′ is the aggregate power of large-scale variable frequency air conditioning load at time t′, N′ h is the number of variable frequency air conditioners in the h′th historical variable frequency air conditioner cluster, T out is the outdoor temperature, T setc,h′,t′ is the temperature setting value of the h′th historical variable frequency air conditioning cluster at time t′; R′ c,h and η′ c,h is the characteristic attribute value of the cluster center of the h′th historical variable frequency air conditioner cluster, R′ c,h is the steady-state parameter corresponding to the cluster center of the h′th historical variable frequency air conditioner cluster, η′ c,h is the temperature change parameter corresponding to the cluster center of the h′th historical variable frequency air conditioning cluster; P′ samax,t′ is the upper limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t′, P′ samin,t′ is the lower limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t′, δ is the user controllability, T out,t′ is the outdoor temperature at time t′, T max,h′,t′ is the upper limit of the temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, T min,h′,t′ is the adjustable lower limit of the temperature of the h′th historical variable frequency air conditioning cluster at time t′.

15. The method according to claim 11, characterized in that The determining of the temperature adjustable margin of the historical variable frequency air conditioning cluster includes: The temperature adjustment margin adjustment amount of the historical variable frequency air conditioning cluster caused by user willingness is calculated by using the initial temperature adjustable upper limit and the initial temperature adjustable lower limit of the historical variable frequency air conditioning cluster. Calculate the upper and lower limits of the temperature adjustment margin of the historical variable frequency air conditioning cluster by using the temperature adjustment margin adjustment amount caused by the user's willingness; The temperature adjustable upper limit and lower limit of the historical variable frequency air conditioning cluster are the temperature adjustable margin of the historical variable frequency air conditioning cluster.

16. The method according to claim 15, characterized in that The calculation formula for the temperature adjustable margin adjustment amount of the historical variable frequency air conditioning cluster caused by user willingness includes: ΔT′ h′,t′ =β′ h′,t′ (T′ max0,h′,t′ -T′ min0,h′,t′ ) The calculation formula of the temperature adjustable margin of the historical variable frequency air conditioning cluster includes: In the above formula, h′∈[1,H′], H′ is the total number of historical variable frequency air conditioner clusters; t′∈[1,T′], T′ is the total time of the historical period; ΔT′ h′,t′ is the temperature adjustable margin adjustment amount of the h′th historical variable frequency air conditioning cluster caused by user willingness at time t′; β′ h′,t′ is the user willingness influencing factor, that is, the comprehensive willingness of the variable frequency air conditioner users of the h′th historical variable frequency air conditioner cluster to participate in demand response at time t′; T′ max0,h′,t′ T′ is the upper limit of the initial temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, min0,h′,t′ is the lower limit of the initial temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, T′ max,h′,t′ T′ is the upper limit of the temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, min,h′,t′ is the adjustable lower limit of the temperature of the h′th historical variable frequency air conditioning cluster at time t′.

17. The method according to claim 11, characterized in that The method of using the data set to train and verify the spatiotemporal convolutional network model includes: Normalizing the data set, and dividing the normalized data set into a training set and a test set; The temporal outdoor temperature of the historical period in the training set, the target aggregate power of the large-scale variable-frequency air-conditioning load issued by the power grid in the historical period, and the adjustable margin of the variable-frequency air-conditioning cluster temperature in the historical period are used as input layer training samples of the spatiotemporal convolutional network model. The upper and lower limits of the adjustable aggregate power of the large-scale variable-frequency air-conditioning load in the historical period in the training set are used as output layer training samples of the spatiotemporal convolutional network model. The spatiotemporal convolutional network model is trained to obtain a trained spatiotemporal convolutional network model. The trained spatiotemporal convolutional network model uses the temporal outdoor temperature of the historical period in the test set, the target aggregate power of the large-scale variable-frequency air-conditioning load issued by the power grid during the historical period, and the adjustable margin of the variable-frequency air-conditioning cluster temperature during the historical period as inputs, and outputs the predicted adjustable upper and lower limits of the large-scale variable-frequency air-conditioning load aggregate power; Determining the prediction accuracy of the trained spatiotemporal convolutional network model based on the upper and lower adjustable limits of the aggregated power of the large-scale variable-frequency air-conditioning load in the historical period in the test set and the predicted upper and lower adjustable limits of the aggregated power of the large-scale variable-frequency air-conditioning load; If the prediction accuracy is greater than or equal to the accuracy threshold, the verification is successful, and the trained spatiotemporal convolutional network model is the large-scale variable-frequency air-conditioning aggregation response potential evaluation model; otherwise, the verification fails, the parameters of the spatiotemporal convolutional network model are adjusted, and the spatiotemporal convolutional network model with adjusted parameters is retrained until the verification is successful.

18. A large-scale variable frequency air conditioner aggregation control device, characterized in that: include: A clustering unit is used to cluster the variable frequency air conditioners in the evaluation period to obtain several target variable frequency air conditioner clusters; A first determining unit, configured to determine a temperature adjustable margin of the target variable frequency air conditioning cluster; A second determining unit is configured to determine, based on the temperature adjustable margin of the target variable frequency air conditioning cluster, a large-scale variable frequency air conditioning aggregated power adjustable margin for the evaluation period using a pre-established large-scale variable frequency air conditioning aggregated response potential evaluation model; A control unit, configured to control the temperature of a target variable frequency air conditioning cluster using a variable frequency air conditioning cluster control strategy based on the adjustable margin of the large-scale variable frequency air conditioning load aggregate power in the time period to be evaluated; The large-scale variable-frequency air-conditioning aggregate response potential assessment model is constructed using the adjustable margin of the aggregate power of large-scale variable-frequency air-conditioning loads in historical periods.

19. The device according to claim 18, characterized in that The clustering unit comprises: A first calculation subunit is used to calculate the steady-state parameters and temperature variation parameters of the variable frequency air conditioner during the period to be evaluated; The first clustering subunit is used to cluster the variable-frequency air conditioners in the evaluation period using the steady-state parameters and temperature change parameters of the variable-frequency air conditioners in the evaluation period as characteristic attributes, and adopt the K-means clustering method to obtain a number of cluster groups and their cluster centers, wherein the cluster groups are the target variable-frequency air conditioner clusters, and the cluster centers are characteristic attributes.

20. The device according to claim 19, characterized in that The calculation formula for the steady-state parameters of the variable-frequency air conditioner during the evaluation period includes: W1=Rη The calculation formula for the temperature change parameter of the variable frequency air conditioner during the evaluation period includes: W2=RC In the above formula, W1 is the steady-state parameter of the variable-frequency air conditioner during the evaluation period, W2 is the temperature change parameter of the variable-frequency air conditioner during the evaluation period, R is the equivalent thermal resistance of the air inside the building, η is the energy efficiency ratio of the variable-frequency air conditioner, and C is the equivalent heat capacity of the air inside the building.

21. The device according to claim 18, characterized in that The first determining unit includes: The second calculation subunit is configured to calculate the temperature adjustable margin adjustment amount of the target variable frequency air conditioning cluster caused by the user's willingness by using the initial temperature adjustable upper limit and the initial temperature adjustable lower limit of the target variable frequency air conditioning cluster; a third calculation subunit, configured to calculate an upper limit and a lower limit of the temperature adjustment of the target variable frequency air conditioning cluster by using the temperature adjustment margin adjustment amount of the target variable frequency air conditioning cluster caused by the user's willingness; The first determining subunit is configured to provide a temperature adjustable upper limit and a lower limit for the target variable frequency air conditioning cluster as a temperature adjustable margin for the target variable frequency air conditioning cluster.

22. The device according to claim 21, characterized in that The calculation formula for the temperature adjustable margin adjustment amount of the target variable frequency air conditioning cluster caused by the user's willingness includes: ΔT h,t =β h,t (T max0,h,t -T min0,h,t ) The calculation formula for the temperature adjustment margin of the target variable frequency air conditioning cluster includes: In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,T e ], T s is the starting time of the period to be evaluated, T e is the result time of the period to be evaluated; ΔT h,t is the temperature adjustable margin adjustment amount of the hth target variable frequency air conditioning cluster caused by user willingness at time t; β h,t is the user willingness influencing factor, that is, the comprehensive willingness of the variable frequency air conditioner users of the hth target variable frequency air conditioner cluster to participate in demand response at time t; T max0,h,t is the upper limit of the initial temperature adjustment of the hth target variable frequency air conditioning cluster at time t, T min0,h,t is the adjustable lower limit of the initial temperature of the hth target variable frequency air conditioning cluster at time t, T max,h,t is the upper limit of the temperature adjustment of the hth target variable frequency air conditioning cluster at time t, T min,h,t is the adjustable lower limit of the temperature of the hth target variable frequency air conditioning cluster at time t.

23. The device according to claim 1, characterized in that The second determining unit includes: The first acquisition subunit is used to collect the temporal outdoor temperature of the time period to be evaluated and the target aggregate power of the large-scale variable frequency air conditioning load issued by the power grid; an output subunit, configured to use the temporal outdoor temperature of the time period to be evaluated, the target aggregate power of the large-scale variable-frequency air-conditioning load issued by the power grid during the time period to be evaluated, and the temperature adjustment margin of the target variable-frequency air-conditioning cluster as inputs to the large-scale variable-frequency air-conditioning aggregate response potential evaluation model, and output the adjustable upper and lower limits of the large-scale variable-frequency air-conditioning load aggregate power during the time period to be evaluated; The second determining subunit is configured to provide an upper limit and a lower limit for adjusting the aggregate power of the large-scale variable-frequency air-conditioning load in the period to be evaluated as an adjustable margin for the aggregate power of the large-scale variable-frequency air-conditioning load in the period to be evaluated.

24. The device according to claim 18, wherein The control unit comprises: A third determining subunit is configured to determine the actual aggregate power constraint of the target variable frequency air conditioning cluster by utilizing the adjustable margin of the aggregate power of the large-scale variable frequency air conditioning load in the time period to be evaluated; The first acquisition subunit is configured to solve the objective function of the variable frequency air conditioning cluster control strategy based on the constraint condition of the actual aggregate power of the target variable frequency air conditioning cluster, with the goal of minimizing the difference between the target aggregate power of the large-scale variable frequency air conditioning load and the actual aggregate power of the target variable frequency air conditioning cluster, to obtain the optimal value of the actual aggregate power of the target variable frequency air conditioning cluster; A fourth calculation subunit, configured to calculate an indoor temperature control set value of the target variable frequency air conditioning cluster according to the actual optimal aggregate power value of the target variable frequency air conditioning cluster; The control subunit is configured to perform temperature control on the target variable frequency air conditioning cluster by using the indoor temperature control set value of the target variable frequency air conditioning cluster.

25. The device according to claim 24, characterized in that The calculation formula for the constraint condition of the actual aggregate power of the target variable frequency air conditioning cluster includes: P samin,t ≤P sa,h,t ≤P samax,t The calculation formula of the objective function of the variable frequency air conditioning cluster control strategy includes: In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,T e ], T s is the starting time of the period to be evaluated, T e is the result time of the period to be evaluated; P sa,h,t is the actual aggregate power of the hth target variable frequency air conditioning cluster at time t, P samax,t is the upper limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t, P samin,t is the adjustable lower limit of the aggregated power of large-scale variable frequency air conditioner load at time t, S is the objective function of the variable frequency air conditioner cluster control strategy, abs(.) is the absolute value operation, P vfset,t is the target aggregate power of large-scale variable frequency air conditioning load issued by the power grid at time t.

26. The device according to claim 25, characterized in that The calculation formula for the actual aggregate power of the target variable frequency air conditioning cluster includes: In the above formula, δ is the user controllability, N h is the number of variable frequency air conditioners in the hth target variable frequency air conditioner cluster, T out,t is the outdoor temperature at time t, T setc,h,t is the temperature setting value of the hth target variable frequency air conditioning cluster at time t, R c,h η c,h is the characteristic attribute value of the cluster center of the h-th target variable frequency air conditioner cluster, R c,h is the steady-state parameter corresponding to the cluster center of the h-th target variable frequency air conditioner cluster, η c,h is the temperature change parameter corresponding to the cluster center of the hth target variable frequency air conditioner cluster, T in,h,t is the indoor temperature of the hth target variable frequency air conditioning cluster at time t, and k is the power variation coefficient.

27. The device according to claim 24, characterized in that The calculation formula for the indoor temperature control set value of the target variable frequency air conditioning cluster includes: In the above formula, h∈[1,H], H is the total number of target variable frequency air conditioner clusters; t∈[T s ,T e ], T s is the starting time of the period to be evaluated, T e T is the result time of the period to be evaluated; in,h,t+1 is the indoor temperature of the hth target variable frequency air conditioning cluster at time t+1, that is, the indoor temperature control set value of the target variable frequency air conditioning cluster; T out,t+1 is the outdoor temperature at time t+1, η is the energy efficiency ratio of the variable frequency air conditioner, k is the power variation coefficient, P′ sa,h,t is the optimal value of the actual aggregate power of the hth target variable frequency air conditioning cluster at time t, R is the equivalent thermal resistance of the air inside the building, T out,t is the outdoor temperature at time t, T in,h,t is the indoor temperature of the hth target variable frequency air conditioning cluster at time t, C is the equivalent heat capacity of the air inside the building, and Δt is the duration of a period.

28. The device according to claim 18, wherein The system further comprises: an establishing unit for establishing a large-scale variable frequency air conditioner aggregate response potential evaluation model; the establishing unit comprises: The second clustering subunit is used to cluster the variable frequency air conditioners in the historical period to obtain several historical variable frequency air conditioner clusters and their cluster centers; a fourth determining subunit, configured to determine an adjustable margin of aggregated power of large-scale variable frequency air conditioner loads in a historical period by using the cluster center of the historical variable frequency air conditioner cluster; The second collection sub-unit is used to collect the time-series outdoor temperature of the historical period and the target aggregate power of the large-scale variable frequency air conditioning load issued by the power grid during the historical period; a fifth determining subunit, configured to determine a temperature adjustable margin of a historical variable frequency air conditioning cluster; A construction subunit is configured to construct a data set using the temporal outdoor temperature of the historical period, the target aggregate power of the large-scale variable frequency air conditioning load issued by the power grid during the historical period, the adjustable margin of the variable frequency air conditioning cluster temperature during the historical period, and the adjustable margin of the large-scale variable frequency air conditioning load aggregate power during the historical period; The second acquisition subunit is used to train and verify the spatiotemporal convolutional network model using the data set, and obtain the large-scale variable frequency air conditioner aggregation response potential evaluation model after successful verification.

29. The device according to claim 28, characterized in that The second clustering subunit includes: A first calculation module is used to calculate the steady-state parameters and temperature variation parameters of the variable frequency air conditioner in a historical period; A clustering module is used to cluster the variable-frequency air conditioners in the historical period using the steady-state parameters and temperature change parameters of the variable-frequency air conditioners in the historical period as characteristic attributes, and adopt a K-means clustering method to obtain a number of cluster groups and their cluster centers, wherein the cluster groups are the historical variable-frequency air conditioner clusters, and the cluster centers are characteristic attributes.

30. The device according to claim 28, wherein The fourth determining subunit includes: The second calculation module is used to calculate the aggregated power of large-scale variable frequency air conditioner loads in the historical period by using the cluster center of the historical variable frequency air conditioner cluster; A third calculation module is used to calculate the adjustable upper and lower limits of the large-scale variable frequency air conditioning load aggregate power during the historical period by using the large-scale variable frequency air conditioning load aggregate power during the historical period; The first determining module is used to determine that the upper and lower limits of the adjustable aggregate power of the large-scale variable-frequency air-conditioning load in the historical period are the adjustable margin of the aggregate power of the large-scale variable-frequency air-conditioning load in the historical period.

31. The device according to claim 30, characterized in that The calculation formula for the aggregate power of large-scale variable frequency air conditioning loads during the historical period includes: The calculation formula for the adjustable margin of the aggregated power of the large-scale variable frequency air-conditioning load during the historical period includes: In the above formula, h′∈[1,H′], H′ is the total number of historical variable frequency air conditioner clusters; t′∈[1,T′], T′ is the total time of the historical period; P′ sa,t′ is the aggregate power of large-scale variable frequency air conditioning load at time t′, N′ h is the number of variable frequency air conditioners in the h′th historical variable frequency air conditioner cluster, T out is the outdoor temperature, T setc,h′,t′ is the temperature setting value of the h′th historical variable frequency air conditioning cluster at time t′; R′ c,h and η′ c,h is the characteristic attribute value of the cluster center of the h′th historical variable frequency air conditioner cluster, R′ c,h is the steady-state parameter corresponding to the cluster center of the h′th historical variable frequency air conditioner cluster, η′ c,h is the temperature change parameter corresponding to the cluster center of the h′th historical variable frequency air conditioning cluster; P′ samax,t′ is the upper limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t′, P′ samin,t′ is the lower limit of the adjustable power of large-scale variable frequency air conditioning load aggregation at time t′, δ is the user controllability, T out,t′ is the outdoor temperature at time t′, T max,h′,t′ is the upper limit of the temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, T min,h′,t′ is the adjustable lower limit of the temperature of the h′th historical variable frequency air conditioning cluster at time t′.

32. The device according to claim 28, characterized in that The fifth determining subunit includes: a fourth calculation module, configured to calculate a temperature adjustable margin adjustment amount of the historical variable frequency air conditioning cluster caused by user willingness by using the initial temperature adjustable upper limit and the initial temperature adjustable lower limit of the historical variable frequency air conditioning cluster; a fifth calculation module, configured to calculate an upper limit and a lower limit of the temperature adjustment of the historical variable frequency air conditioning cluster by using the temperature adjustment margin adjustment amount caused by the user's willingness of the historical variable frequency air conditioning cluster; The second determining module is configured to determine that the upper and lower limits of the temperature adjustment of the historical variable frequency air-conditioning cluster are the temperature adjustment margin of the historical variable frequency air-conditioning cluster.

33. The device according to claim 32, characterized in that The calculation formula for the temperature adjustable margin adjustment amount of the historical variable frequency air conditioning cluster caused by user willingness includes: ΔT′ h′,t′ =β′ h′,t′ (T′ max0,h′,t′ -T′ min0,h′,t′ ) The calculation formula of the temperature adjustable margin of the historical variable frequency air conditioning cluster includes: In the above formula, h′∈[1,H′], H′ is the total number of historical variable frequency air conditioner clusters; t′∈[1,T′], T′ is the total time of the historical period; ΔT′ h′,t′ is the temperature adjustable margin adjustment amount of the h′th historical variable frequency air conditioning cluster caused by user willingness at time t′; β′ h′,t′ is the user willingness influencing factor, that is, the comprehensive willingness of the variable frequency air conditioner users of the h′th historical variable frequency air conditioner cluster to participate in demand response at time t′; T′ max0,h′,t′ T′ is the upper limit of the initial temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, min0,h′,t′ is the lower limit of the initial temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, T′ max,h′,t′ T′ is the upper limit of the temperature adjustment of the h′th historical variable frequency air conditioning cluster at time t′, min,h′,t′ is the adjustable lower limit of the temperature of the h′th historical variable frequency air conditioning cluster at time t′.

34. The device according to claim 28, wherein The second acquiring subunit includes: A processing module, configured to normalize the data set and divide the normalized data set into a training set and a test set; A training module is configured to use the time-series outdoor temperature of the historical period in the training set, the target aggregate power of the large-scale variable-frequency air-conditioning load issued by the power grid in the historical period, and the adjustable margin of the variable-frequency air-conditioning cluster temperature in the historical period as input layer training samples of the spatiotemporal convolutional network model, and use the adjustable upper and lower limits of the aggregate power of the large-scale variable-frequency air-conditioning load in the historical period in the training set as output layer training samples of the spatiotemporal convolutional network model to train the spatiotemporal convolutional network model, thereby obtaining a trained spatiotemporal convolutional network model; an output module, configured to use the time-series outdoor temperature of the historical period in the test set, the target aggregate power of the large-scale variable-frequency air-conditioning load issued by the power grid during the historical period, and the adjustable margin of the variable-frequency air-conditioning cluster temperature during the historical period as inputs to the trained spatiotemporal convolutional network model, and output the predicted upper and lower limits of the adjustable aggregate power of the large-scale variable-frequency air-conditioning load; A third determination module is used to determine the prediction accuracy of the trained spatiotemporal convolutional network model based on the upper and lower adjustable limits of the aggregate power of the large-scale variable-frequency air-conditioning load in the historical period in the test set and the predicted upper and lower adjustable limits of the aggregate power of the large-scale variable-frequency air-conditioning load; A verification module is used to verify that if the prediction accuracy is greater than or equal to the accuracy threshold, the verification is successful, and the trained spatiotemporal convolutional network model is the large-scale variable-frequency air-conditioning aggregation response potential evaluation model; otherwise, the verification fails, the parameters of the spatiotemporal convolutional network model are adjusted, and the spatiotemporal convolutional network model with adjusted parameters is retrained until the verification is successful.

35. An electronic device, characterized in that: include: at least one processor and memory; The memory and the processor are connected via a bus; The memory is used to store one or more programs; When the one or more programs are executed by the at least one processor, the large-scale variable-frequency air conditioner aggregation control method according to any one of claims 1 to 17 is implemented.

36. A readable storage medium, characterized in that An execution program is stored thereon, and when the execution program is executed, the large-scale variable-frequency air conditioner aggregation control method according to any one of claims 1 to 17 is implemented.

Images