Modeling method of regulation characteristics and boundary of virtual power plant with space cooling load regulation

Abstract

The application provides a space refrigeration load regulation type virtual power plant regulation characteristic and boundary modeling method, comprising the following steps: constructing a space refrigeration load monomer discrete physical model; constructing a load aggregate model by comprehensively using a consistency control algorithm and the load monomer discrete physical model; using statistical quantities to represent load aggregate model parameters and curve parameters; calculating the aggregate model parameters based on the load equipment monomer parameter statistics; updating the multi-time scale aggregate model parameters based on the difference between the basic model time scale and the actual scheduling time scale; calculating the regulation boundary by comprehensively considering the power constraint and the energy constraint based on the updated multi-time scale aggregate model; and quantitatively evaluating the maximum up-regulation power and the maximum down-regulation power of the virtual power plant under the specified regulation time length. The application accurately depicts the regulation characteristic and the regulation boundary of the space refrigeration load virtual power plant, designs a parameter dynamic updating mechanism for different scheduling periods, and enhances the time scale adaptability of the model.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of new power system dispatching control technology, and particularly relates to a space refrigeration load regulation type virtual power plant regulation characteristic and boundary modeling method. BACKGROUND

[0002] A virtual power plant is an advanced technical system that aggregates and optimizes dispersed resources such as distributed power sources, adjustable loads, energy storage devices, etc. through information communication and intelligent control technology. Its core value lies in improving system flexibility, promoting efficient use of resources, and enhancing the ability of the power system to cope with the uncertainty of renewable energy fluctuations. According to the differences in resource composition and main functions, virtual power plants can be divided into power source type virtual power plants (mainly distributed photovoltaic, wind power and other renewable energy sources), full power type virtual power plants (integrated power, load and energy storage resources, pursuing comprehensive power optimization) and load regulation type virtual power plants (mainly demand-side adjustable load resources, providing flexible services such as peak shaving and frequency modulation). Among them, the load regulation type virtual power plant, due to its fast response speed, low investment cost and extensive resource distribution, has become an important support form for building a new power system and promoting the coordinated development of source, network and load. Among various adjustable load resources, space refrigeration load (such as air conditioners, cold storage, ice making systems, etc.) is the most representative flexible load resource as the main load type in peak periods, with wide distribution and great control potential at the user side. This type of load has certain heat storage capacity and operational adjustability, and can participate in demand response, dispatch optimization and ancillary services of the power system without significantly affecting user comfort or production conditions. With the improvement of load observability and controllability, space refrigeration load gradually shows its value in virtual power plants, and has broad development and application prospects.

[0003] In order to realize the effective participation of space refrigeration load regulation type virtual power plant in actual dispatching and market, its regulation characteristics and boundaries must be accurately described. The regulation characteristics refer to the use of the heat storage capacity of space refrigeration load to flexibly adjust the power of the cluster load within a certain range and time length under certain operating conditions and comfort constraints. The regulation boundary defines the upper and lower limits of the adjustable power of the space refrigeration load cluster, which is the key basis for realizing dispatch feasibility and safety. Precise regulation modeling is not only a prerequisite for virtual power plants to participate in demand response and ancillary services, but also an important support for source-load coordination and optimal operation decision-making. However, due to the characteristics of space refrigeration load resources such as large individual differences, time-varying response behavior, and significant influence of environmental and behavioral disturbances, the regulation capacity has high uncertainty and time variability.

[0004] Currently, the modeling methods of space cooling load aggregations mainly fall into two categories: the first category is a data-driven modeling method. This method analyzes the historical operation data of the load aggregation, uses machine learning techniques such as clustering, regression, and neural networks to construct an input-output mapping relationship model, and thus describes the response characteristics of the aggregated load. The significant advantage of this method is low dependence on single-body parameters, high modeling efficiency, and the need for only historical power curve data of the aggregated load, which saves the work of single-body modeling and parameter measurement and is suitable for large-scale deployment scenarios. However, this method has obvious shortcomings: on the one hand, the resulting model lacks physical mechanism constraints and is difficult to describe the unique thermal inertia and operating boundaries of space cooling loads; on the other hand, the aggregated data samples often cover limited operating conditions and adjustment scenarios, have low data dimensions and poor generalization ability, and are difficult to accurately reflect the response behavior under future atypical conditions, resulting in low model accuracy and robustness. The second category is a modeling method based on Minkowski summation. This method performs Minkowski summation on the adjustment capability set (such as the power-energy boundary) of each load single-body, obtains the overall adjustable capability set, and thus theoretically completely retains the boundary information of each single-body, achieving accurate modeling of the group adjustment boundary. This method performs well in modeling capability and can accurately describe the adjustment upper and lower limits, power trajectory space, and state transition boundary. However, this method also has obvious defects: first, it requires high accuracy of the input single-body model and needs to obtain the physical parameters and operating states of all resources, which is costly in modeling and maintenance; second, the aggregated model obtained by Minkowski summation is complex and has no intuitive physical meaning, which is not conducive to the formulation of scheduling instructions and model verification; in addition, this type of modeling method usually assumes that all single-body operating conditions are known and stable, ignoring the influence of environmental disturbances and user behavior, which limits its adaptability in dynamic environments. In summary, the two mainstream methods have advantages and disadvantages in aggregated energy modeling: the former is suitable for rapid deployment scenarios but has limited accuracy, and the latter has high modeling accuracy but poor scalability and adaptability. Therefore, it is urgent to propose a space cooling load aggregation model construction method that takes into account the physical mechanism accuracy and uncertainty adaptability to meet the actual regulation and control needs of virtual power plants in multiple scenarios. In addition, most existing space cooling load characteristic models are constructed for a single time scale, while the regulation and control time scales differ significantly in different interactive scenarios. Therefore, the space cooling load aggregation model should also have the ability to adapt to changes in time scales. SUMMARY

[0005] In order to solve the above technical problems, the application provides a space refrigeration load regulation type virtual power plant regulation characteristic and boundary modeling method.

[0006] In order to achieve the above purpose, the application adopts the following technical scheme:

[0007] The space refrigeration load regulation type virtual power plant regulation characteristic and boundary modeling method comprises the following steps:

[0008] A discrete physical model of a space refrigeration load unit is constructed; a space refrigeration load aggregate model is constructed by comprehensively considering a SOC consistency control algorithm and the discrete physical model of the space refrigeration load unit; statistical quantities are used to represent parameters and curve parameters of the space refrigeration load aggregate model; and the aggregate model parameters based on the statistical quantities of the space refrigeration load unit parameters are calculated.

[0009] Based on the difference between the time scale of the basic model and the actual scheduling time scale, the multi-time scale aggregate model parameters are updated, and the dynamic adaptation of the multi-time scale aggregate model under the multi-stage scheduling cycle is realized.

[0010] Based on the updated multi-time scale aggregate model, the regulation boundary is calculated by comprehensively considering the power constraint and the energy constraint, and the maximum up-regulation power and the maximum down-regulation power of the virtual power plant under the specified regulation time length are quantitatively evaluated.

[0011] The effects provided in the summary of the application are only the effects of the embodiments, not all the effects of the application. One of the technical schemes in the above technical scheme has the following advantages or beneficial effects:

[0012] The application proposes a space refrigeration type load regulation virtual power plant regulation characteristic and boundary modeling method, including the following steps: constructing a space refrigeration type load monomer discrete physical model; constructing a space refrigeration type load aggregate model by synthesizing a SOC consistency control algorithm and the space refrigeration type load monomer discrete physical model; adopting a statistical quantity to represent space refrigeration type load aggregate model parameters and curve parameters; calculating the aggregate model parameters based on the statistical quantity of space refrigeration load equipment monomer parameters; updating the multi-time scale aggregate model parameters based on the difference between the basic model time scale and the actual scheduling time scale, realizing the dynamic adaptation of the multi-time scale aggregate model under the multi-stage scheduling period; based on the updated multi-time scale aggregate model, considering the power constraint and the energy constraint, calculating the regulation boundary, and quantitatively evaluating the maximum up-regulation power and the maximum down-regulation power of the virtual power plant under the specified regulation time length. The application can be applied to typical scenes such as virtual power plant participation in new power system scheduling and demand response management, accurately depicts the regulation characteristics and regulation boundary of the space refrigeration type load virtual power plant, designs a parameter dynamic updating mechanism for different scheduling periods, and enhances the time scale adaptability of the model. The method reduces the modeling workload while ensuring the modeling accuracy of the virtual power plant model, and provides theoretical support and engineering tools for constructing a high-precision, strong-adaptability regulation type virtual power plant modeling system. BRIEF DESCRIPTION OF DRAWINGS

[0013] Figure 1 A space refrigeration type load regulation virtual power plant regulation characteristic and boundary modeling method flowchart for the embodiment 1 of the application is proposed.

[0014] Figure 2 A space refrigeration load aggregate model construction flowchart for the embodiment 1 of the application is proposed, which comprehensively considers the monomer physical characteristics and the cluster SOC consistency control algorithm.

[0015] Figure 3 An aggregate model parameter adaptive updating mechanism flowchart for the embodiment 1 of the application is proposed based on time scale mapping.

[0016] Figure 4 A virtual power plant regulation capacity evaluation flowchart for the embodiment 1 of the application is proposed based on power-energy constraints.

[0017] Figure 5 A cluster operating environment parameter and grid interaction regulation instruction schematic diagram for the embodiment 1 of the application is proposed.

[0018] Figure 6 An air conditioner cluster power baseline schematic diagram for the embodiment 1 of the application is proposed.

[0019] Figure 7 An air conditioner cluster total power schematic diagram for the embodiment 1 of the application is proposed.

[0020] Figure 8 A schematic diagram of a basic aggregation model SOC proposed for embodiment 1 of the present application;

[0021] Figure 9 A schematic diagram of a multi-time scale aggregation model SOC proposed for embodiment 1 of the present application;

[0022] Figure 10 A power constraint diagram of an air conditioning cluster under different adjustment durations proposed for embodiment 1 of the present application;

[0023] Figure 11 An energy constraint diagram of an air conditioning cluster under different adjustment durations proposed for embodiment 1 of the present application;

[0024] Figure 12 A total adjustment capacity diagram of an air conditioning cluster under different time scales proposed for embodiment 1 of the present application. DETAILED DESCRIPTION

[0025] To clearly illustrate the technical features of the present scheme, the present application is described in detail below with reference to specific embodiments and the accompanying drawings. The following disclosure provides many different embodiments or examples to implement the different structures of the present application. In order to simplify the disclosure of the present application, the components and settings of specific examples are described below. In addition, the present application can repeatedly refer to numerals and / or letters in different examples. Such repetition is for the purpose of simplification and clarity, and does not in itself indicate the relationship between the various embodiments and / or settings being discussed. It should be noted that the components illustrated in the drawings are not necessarily drawn to scale. The present application omits the description of well-known components and processing techniques and processes to avoid unnecessary limitation of the present application.

[0026] Embodiment 1

[0027] The embodiment 1 of the present application proposes a space cooling load regulation type virtual power plant regulation characteristic and boundary modeling method, which is used to solve the problems of complex modeling process, poor multi-time scale adaptability and insufficient consideration of uncertain factors in the existing space cooling load regulation type virtual power plant modeling method.

[0028] Figure 1 A flowchart of a space cooling load regulation type virtual power plant regulation characteristic and boundary modeling method proposed for embodiment 1 of the present application;

[0029] In step 1, a discrete physical model of a space cooling load monomer is constructed; a space cooling load aggregate model is constructed by integrating a SOC consistency control algorithm and a discrete physical model of a space cooling load monomer; statistical quantities are used to represent the parameters and curve parameters of the space cooling load aggregate model; and the aggregate model parameters based on the statistical quantities of the space cooling load device monomer parameters are calculated;

[0030] Figure 2 This is a flowchart of the construction process of the space cooling load aggregate model proposed in Embodiment 1 of the present invention, which comprehensively considers the physical characteristics of individual units and the cluster SOC consistency control algorithm.

[0031] The process of constructing a discrete physical model of a single unit of space cooling load includes:

[0032] Based on the principle of thermal balance, a discrete physical model of a single space cooling load is constructed, specifically as follows:

[0033] (1)

[0034] in, for Real-time indoor temperature; for outdoor temperature at all times; For building thermal resistance, For the heat capacity of the house; For cooling efficiency; for The operating power of a single unit in a space-time cooling load; This refers to the number of individual units in a space cooling load category;

[0035] Analogous to the energy and power relationship of traditional energy storage systems, the relationship between indoor temperature and cooling power is described using normalized temperature as the state of charge, specifically:

[0036] (2)

[0037] (3)

[0038] in, for Normalized temperature status of individual spatial cooling load units at any given time; for Baseline power of the unit at any given time; for Real-time individual unit power adjustment; The upper limit of the permissible temperature for indoor use; The lower limit of the allowable temperature for indoor use; To set the temperature;

[0039] Based on the principle that the discretization time step is much smaller than the individual thermal inertia time constant, the discretization step size is determined. Discretize equation (2) as follows:

[0040] (4)

[0041] in, This represents the upper limit of the unit's operating power. This represents the lower limit of the unit's operating power. These are the parameters for the first individual model; These are the parameters for the second monomer model; For the parameters of the third monomer model; For the first Each space cooling load unit will be in the next time step SOC; Representing the Individual space cooling load units at time step SOC; Represents the time step; Representing the Individual space cooling load units at time step The reference power; Representing the Individual space cooling load units at time step The change in charging and discharging power; specifically expressed as:

[0042] (5)

[0043] in, The individual cell bias charging power; This is a correction factor; The corrected thermal resistance; The corrected heat capacity; is the thermal inertia time constant of a single unit.

[0044] A spatial cooling load aggregation model is constructed by combining the SOC consistency control algorithm and the normalized individual discrete model.

[0045] Based on the SOC consistency control algorithm, its mathematical rules can be written as follows:

[0046] ; (6)

[0047] Combining formula (6) and formula (4), we get:

[0048] ; (7)

[0049] in, Indicates the space cooling load cluster at time step The state of charge; Indicates the space cooling load cluster at time step Adjustable power; This indicates the upper limit of the adjustable power of a space-cooled load cluster; This indicates the lower limit of the adjustable power of a space-cooled load cluster; Indicates the first the internal temperature of a space refrigeration type load monomer at a time step ; is a first polymer model parameter; is a second polymer model parameter; is a third polymer model parameter;

[0050] ; (8)

[0051] wherein, is a power baseline of a space refrigeration type load cluster at a time step ; is a bias charging power of a space refrigeration type load cluster at a time step ; is a thermal resistance of a space refrigeration type load cluster; is a thermal capacity of a space refrigeration type load cluster; is a thermal inertia time constant of a space refrigeration type load cluster.

[0052] The process of representing the parameters and curve parameters of the space refrigeration type load polymer model by statistical quantities includes:

[0053] If , , , , , , meet the uniform distribution, then:

[0054] ; (9)

[0055] ; (10)

[0056] ; (11)

[0057] ; (12)

[0058] wherein, represents an equivalent saturated thermal inertia time constant of a space refrigeration type load cluster; is the total number of space refrigeration type loads in a space refrigeration type load cluster; is the average value of the thermal inertia time constant of a space refrigeration type load monomer in a space refrigeration type load cluster; , are respectively the maximum value and the minimum value of the thermal resistance of a space refrigeration type load monomer in a space refrigeration type load cluster; , are respectively the maximum value and the minimum value of the refrigeration efficiency of a space refrigeration type load monomer in a space refrigeration type load cluster, is the average value of the thermal capacity of a space refrigeration type load monomer in a space refrigeration type load cluster; , the average value of the upper and lower limits of the indoor temperature allowed by the space refrigeration type load cluster unit respectively; the average value of the temperature set value of the space refrigeration type load cluster unit; , , the cluster thermal resistance in the cluster statistical quantity parameter of the space refrigeration type load; the heat capacity in the cluster statistical quantity parameter of the space refrigeration type load; the bias charging power in the cluster statistical quantity parameter of the space refrigeration type load;

[0059] Substitute formula (9), (11) and (12) into formula (8) respectively to form a method for calculating the aggregation model parameter based on the statistical quantity as follows:

[0060] ; (13)

[0061] ; (14)

[0062] ; (15)

[0063] wherein, the first aggregation model parameter calculated by the statistical quantity; the second aggregation model parameter calculated by the statistical quantity; the third aggregation model parameter calculated by the statistical quantity;

[0064] The set temperature and the outdoor temperature curve are normalized respectively as follows:

[0065] ; (16)

[0066] ; (17)

[0067] wherein, the set temperature state; the outdoor temperature state;

[0068] The cluster power baseline is calculated as follows:

[0069] ; (18)

[0070] wherein, the cluster power baseline calculated by the statistical quantity; the average value of the set temperature state; the average value of the outdoor temperature state.

[0071] The application calculates the aggregation model parameter based on the statistical quantity of the space refrigeration load equipment unit, which comprises:

[0072] Calculating cluster statistical quantity parameters based on sampling of spatial refrigeration load equipment monomer parameters;

[0073] Calculating aggregate model parameters based on sampling sample statistical quantity;

[0074] Calculating cluster power regulation boundaries based on research parameters;

[0075] Calculating cluster power baselines based on multi-dimensional outdoor temperature curves.

[0076] The process of calculating cluster statistical quantity parameters based on sampling of spatial refrigeration load equipment monomer parameters includes:

[0077] Determine the total number of spatial refrigeration load equipment monomers to be aggregated for modeling , Usually thousands or more;

[0078] Number the equipment monomers as , determine the sampling ratio , and calculate the sample capacity as On this basis, the non-replacement random sampling method is adopted, that is, randomly sampling equipment samples from the population, and each sample is not returned to the sample pool after being drawn, ensuring that there is no repetition in the sampling process;

[0079] For each sampling object, research its running power upper and lower limits and , refrigeration efficiency , allowable indoor temperature upper and lower limits and , and calculate the correction coefficient . At the same time, set the discretization step size , record the indoor temperature , outdoor temperature , and running power of the monomer within time steps, and construct the matrix:

[0080] ; (19)

[0081] wherein is the temperature difference vector; is the indoor temperature of the th equipment at the th time step;

[0082] ; (20)

[0083] wherein is the feature matrix;​ the outdoor temperature of the first device at the first time step; the outdoor temperature of the first device at the first time step; the outdoor temperature of the first device at the first time step; the outdoor temperature of the first device at the first time step; the outdoor temperature of the first device at the first time step; the outdoor temperature of the first device at the first time step; the outdoor temperature of the first device at the first time step; the outdoor temperature of the first device at the first time step; the outdoor temperature of the first device at the first time step;

[0084] the outdoor temperature of the first device at the first time step; the outdoor temperature of the first device at the first time step; the outdoor temperature of the first device at the first time step;

[0085] ; (21)

[0086] wherein, the outdoor temperature of the first device at the first time step; the outdoor temperature of the first device at the first time step; the outdoor temperature of the first device at the first time step;

[0087] the outdoor temperature of the first device at the first time step; the outdoor temperature of the first device at the first time step; the outdoor temperature of the first device at the first time step;

[0088] investigating the user set temperature, uniformly setting the temperature state value as , calculating the sampling user temperature set value ;

[0089] based on the sampling sample parameters, calculating , and the statistical maximum and minimum values of the adjustment boundary factor :

[0090] ; (22)

[0091] wherein, is the maximum value of the thermal resistance in the cluster; is the minimum value of the thermal resistance in the cluster; is the maximum value of the refrigeration efficiency in the cluster; is the minimum value of the refrigeration efficiency in the cluster; is the minimum value of the power up-regulation ability factor; is the maximum value of the power down-regulation ability factor;

[0092] calculating , , , ,​ , the statistical mean of the amount is:

[0093] ; (23)

[0094] wherein, is the cluster average thermal inertia time constant; is the cluster average thermal capacity; is the cluster average allowable temperature upper limit; is the cluster average allowable temperature lower limit; is the cluster average set temperature; is the cluster average set SOC;

[0095] The process of calculating the aggregate model parameters based on the statistical amount of the sampling sample includes:

[0096] Substituting , , , , , into equation (10), we get ;

[0097] Substituting into equation (9), we get , and further substituting into equation (13), we get the aggregate model parameters ;

[0098] Substituting , , , , into equation (11), we get , and further substituting into equation (14), we get the aggregate model parameters ;

[0099] Substituting , , , , , into equation (12), we get , and further substituting , into equation (15), we get the aggregate model parameters .

[0100] The process of calculating the cluster power regulation boundary based on the investigation parameters includes: based on the statistical maximum and minimum of the regulation boundary factor Computing the cluster power regulation boundary:

[0101] ; (24)

[0102] wherein, is the upper limit of the cluster power regulation boundary; is the lower limit of the cluster power regulation boundary.

[0103] The process of computing the cluster power baseline based on multi-dimensional outdoor temperature curves includes:

[0104] Assuming that the aggregated space cooling loads are located in similar regions and have the same outdoor temperature curve, based on historical meteorological data, short-term weather forecasts are made or directly based on the meteorological bureau public database, the outdoor temperature curve of the region where the space cooling load is located is obtained, including the highest temperature curve , the lowest temperature curve and the average temperature curve ; the normalized SOC corresponding to each outdoor temperature curve is calculated by substituting formula (17):

[0105] ; (25)

[0106] wherein, is the outdoor temperature state under the average value model of the outdoor temperature curve, is the outdoor temperature state under the maximum value model of the outdoor temperature curve; is the outdoor temperature state under the minimum value model of the outdoor temperature curve;

[0107] Substituting , , into formula (18) respectively to calculate the cluster power baseline corresponding to each temperature curve:

[0108] ; (26)

[0109] wherein, the power baseline and , , , , the aggregation parameters are the average value model describing the space cooling load cluster, the power baseline and , , , , the aggregation parameters are the maximum value model describing the space cooling load cluster, the power baseline and , , , , The aggregated parameters are the minimum model describing the space cooling load cluster. Combined, these three parameters form a set of models that consider the uncertainty of operating temperature; the appropriate model can be selected for operation scheduling and regulation capability assessment as needed.

[0110] In step 2, based on the difference between the time scale of the basic model and the actual scheduling time scale, the parameters of the multi-time scale aggregation model are updated to achieve dynamic adaptation of the multi-time scale aggregation model under multi-level scheduling cycles.

[0111] Based on the difference between the actual scheduling time scale and the discretization step size of the basic model, this section establishes a time scale mapping relationship for the model parameters, designs a parameter update formula based on recursive summation, and realizes the adaptive adjustment of aggregated model parameters for different scheduling cycles. Figure 3 The flowchart is shown in Embodiment 1 of this invention for the adaptive update mechanism of aggregation model parameters based on time scale mapping.

[0112] Obtain the actual scheduling time step Discretization step size of the basic model Calculate their ratio That is:

[0113] ;

[0114] like Then use directly replace The parameters of the aggregate model remain unchanged, that is:

[0115] (27)

[0116] but These are the updated parameters for the first aggregation model; These are the updated parameters for the second aggregation model; Updated parameters for the third aggregation model;

[0117] The updated state transition equation is:

[0118] (28)

[0119] Adjusting the upper and lower limit curves of power , Compared with power baseline The numerical values ​​remain unchanged; the discretization step size of the original curve needs to be adjusted accordingly. Every Linear interpolation is performed to obtain updated upper and lower limit curves and power baseline;

[0120] like , then the aggregate model parameters need to be updated according to The specific updating method is as follows:

[0121] ; (29)

[0122] The updated state transition equation is:

[0123] ; (30)

[0124] Adjusting the upper and lower limit curves of power , The numerical value of the power baseline is unchanged, and the original curve is sampled every to obtain the updated upper and lower limit curves of power and the power baseline.

[0125] In step 3, based on the updated multi-time scale aggregate model, the power constraint and the energy constraint are comprehensively considered to calculate the adjustment boundary and quantitatively evaluate the maximum up-regulated power and the maximum down-regulated power of the virtual power plant under the specified adjustment time length. Figure 4 The flow chart of the virtual power plant adjustment capacity evaluation process based on power-energy constraint for embodiment 1 of the present application is as follows:

[0126] The space cooling type load is subject to power regulation constraint during operation, and the total power regulation constraint of the cluster is:

[0127] ; (31)

[0128] Among them, is the upper limit of the cluster power regulation capacity under the power regulation constraint; is the lower limit of the cluster power regulation capacity under the power regulation constraint;

[0129] During the power regulation of the space cooling type load, the normalized SOC constraint should be met:

[0130] ; (32)

[0131] Determine the length of the adjustment capacity duration ; and determine the upper limit and the lower limit of SOC that needs to be met;

[0132] Calculate the upper limit and the lower limit of the adjustment capacity under the length of the adjustment capacity duration corresponding to each time point:

[0133] ; (33)

[0134] Within the constraints of energy and power, the power regulation boundary of the virtual power plant for a specified regulation duration is:

[0135] (34)

[0136] in, The upper limit of the total power regulation capacity of the virtual power plant when the regulation duration is specified; The lower limit of the total power regulation capacity of the virtual power plant when the specified regulation duration is specified.

[0137] This invention mainly comprises three parts: aggregate modeling that comprehensively considers the physical characteristics of individual units and the cluster SOC consistency control algorithm; adaptive parameter updating based on time-scale mapping; and regulation capability calculation considering power-energy constraints. The first part adopts a bottom-up modeling approach, combining the thermal physical characteristics of individual spatial cooling loads with the group SOC consistency control strategy to construct an aggregate state model that uniformly describes the response behavior of the load cluster. By statistically processing parameters such as individual unit heat capacity, thermal resistance, and rated power, the state transition parameters of the aggregate model are calculated. Simultaneously, a method for calculating aggregate model parameters based on changes in operating conditions such as outdoor temperature is designed, forming a model set characterized by average and extreme value models. The second part addresses the practical need for power dispatch across multiple time scales by proposing a mapping relationship between model parameters and the dispatch cycle. Based on the difference between the actual dispatch time scale and the basic model time scale, a time resolution adaptive mechanism for the aggregate model is constructed. The third part, based on the constructed aggregate model, comprehensively considers power and energy constraints, derives the calculation expressions for the maximum adjustable power and maximum adjustable power of the virtual power plant under a given regulation duration, obtaining a precise and executable regulation capability boundary.

[0138] To verify the regulation characteristics and boundary modeling method of the virtual power plant for spatial cooling load regulation proposed in this invention, taking the interaction process of an air conditioning cluster containing 1000 variable frequency air conditioners with the power grid as an example, the operating data curves and regulation boundaries of the air conditioning cluster are simulated based on this method. The specific process is as follows:

[0139] The total number of individual space cooling load devices to be aggregated and modeled is 1000. The sampling ratio is determined to be 0.2. The random sampling method without replacement is adopted, that is, 200 devices are randomly selected from the population. Each sample is not returned to the sample pool after being selected, ensuring that there is no duplication in the sampling process.

[0140] For each sampled object, investigate its upper and lower limits of operating power. and Refrigeration efficiency Permissible upper and lower limits of indoor temperature and Calculate the thermal resistance of a single unit Heat capacity Thermal inertia time constant According to equations (22)-(23), the statistics can be calculated to form a table of statistical parameters. Some statistical parameters of this cluster are shown in Table 1;

[0141] Table 1: Statistical parameters of air conditioning clusters

[0142]

[0143] The survey investigated user-set temperatures, assuming that air conditioning users maintain their setpoints at the optimal room temperature when not participating in grid response. The temperature setpoints for this cluster are shown in Table 2.

[0144] Table 2: Temperature Setpoints for Partial Air Conditioning Clusters

[0145]

[0146] The outdoor temperature survey directly obtained the outdoor temperature curves for the area where the space cooling load is located from the meteorological bureau's publicly available database, generating three curves: average, maximum, and minimum values. Figure 5 (a) indicates that the air conditioning cluster is designed to interact with the power grid, and the interaction process is achieved through the air conditioning cluster's response, such as... Figure 5 (b) shows the step-type power grid regulation command to simulate it. Figure 5 This is a schematic diagram of the interaction and adjustment commands between the cluster operating environment parameters and the power grid proposed in Embodiment 1 of the present invention.

[0147] The basic discretization time step is set to 10s. The aggregate model parameters A, B, and C are calculated using the individual parameter statistics to construct the state transition equation in equation (7). The aggregate model parameters are shown in Table 3.

[0148] Table 3: Basic Aggregation Model Parameters

[0149]

[0150] Considering the uncertainty of outdoor temperature parameters, the cluster power baseline under three different outdoor temperature curves is calculated, such as... Figure 6 As shown, Figure 6 This is a schematic diagram of the power baseline of the air conditioning cluster proposed in Embodiment 1 of the present invention.

[0151] Using the state transition equations obtained above, a cluster power baseline under the average outdoor temperature curve is selected. Considering the grid response, the actual grid-connected power of the air conditioning cluster and the state of charge (SOC) of the basic aggregation model are calculated; the results are as follows. Figure 7 and Figure 8 As shown. Figure 7 This is a schematic diagram of the total power of the air conditioning cluster proposed in Embodiment 1 of the present invention; Figure 8The basic aggregation model soc proposed for the embodiment 1 of the present application is shown in the schematic diagram. As shown in the diagram, the aggregation model can accurately describe the power curve and soc of the air conditioner cluster in the process of participating in the grid response, has good aggregation effect, and the method of using parameter statistics to obtain the state transition model parameters significantly reduces the difficulty and amount of model acquisition.

[0152] Further change the scheduling time scale, update the state transition parameters of the air conditioner aggregation model under the 5min, 30min, 1h, 2h time scale respectively based on step 2, and the calculation results are shown in Table 4.

[0153] Table 4: Parameters of multi-time scale aggregation model

[0154]

[0155] Using the updated aggregation parameters, the cluster power baseline under the average outdoor temperature curve is selected, and the soc of the air conditioner cluster is calculated considering the grid response, and the results are shown in Figure 9 , Figure 9 The multi-time scale aggregation model soc proposed for the embodiment 1 of the present application is shown in the schematic diagram. As shown in the diagram, the multi-time scale extension does not reduce the modeling accuracy of the aggregation model, and the proposed aggregation model can be used for short-term frequency modulation application analysis, and can also be used for minute-level to hour-level long-term economic dispatch scenarios.

[0156] Further using the aggregation model parameters in Table 4, the time-varying curves of adjustable capacity considering power constraint and energy constraint respectively and comprehensively considering power constraint and energy constraint are calculated based on step 3, and the calculation results are shown in Figure 10 , Figure 11 and Figure 12 . Figure 10 The air conditioner cluster power constraint diagram under different regulation time lengths proposed for the embodiment 1 of the present application is shown in Figure 11 The air conditioner cluster energy constraint diagram under different regulation time lengths proposed for the embodiment 1 of the present application is shown in Figure 12 The air conditioner cluster power total regulation capacity diagram under different time scales proposed for the embodiment 1 of the present application is shown in. As shown in the diagram, the adjustable capacity of the air conditioner cluster presents time-varying and asymmetric characteristics, basically presents the rule that the up-regulation capacity is greater than the down-regulation capacity, and the up-regulation capacity peak mainly appears in the late night and early morning when the temperature is low, while the down-regulation capacity peak appears in the period when the temperature is high; the regulation time length will affect the adjustable capacity considering energy constraint, and the adjustable capacity presents a downward trend with the increase of the regulation time length, and the time-varying and the asymmetry of the up-regulation and down-regulation capacity are weakened.

[0157] The above describes the specific embodiments of the present application in combination with the drawings, but is not a limitation on the protection scope of the present application. Based on the above description, other different forms of modifications or changes can be made by those skilled in the art. Here, all the embodiments need not and cannot be exhausted. Various modifications or changes made by those skilled in the art on the basis of the technical solutions of the present application without creative labor are still within the protection scope of the present application.

Claims

1. A method for modeling the regulating characteristics and boundaries of a virtual power plant for regulating space refrigeration type loads, characterized by, Includes the following steps: A discrete physical model of a single space cooling load is constructed; a composite model of space cooling loads is constructed by integrating the SOC consistency control algorithm with the discrete physical model of the single space cooling load; the parameters and curve parameters of the composite model of space cooling loads are represented by statistical quantities. Calculate the aggregate model parameters based on the individual parameter statistics of space cooling loads; The process of constructing a discrete physical model of a single unit of space cooling load includes: Based on the principle of thermal balance, a discrete physical model of a single space cooling load is constructed, specifically as follows: ；（1） wherein, is the indoor temperature at the moment; is the outdoor temperature at the moment; is the monomer thermal resistance, is the monomer heat capacity; is the refrigeration efficiency; is the monomer operating power of the space refrigeration load at the moment; is the total number of monomers of the space refrigeration load in the space refrigeration load cluster; The relationship between indoor temperature and cooling power is described using normalized temperature as the state of charge, specifically: ；（2） ；（3） wherein, is the normalized temperature state of the space cooling load at the instant; is the baseline power of the unit at the instant; is the unit regulation power at the instant; is the room content temperature upper limit; is the room content temperature lower limit; is the set temperature; The discretization time step is determined according to the principle that the discretization time step is much smaller than the monomer thermal inertia time constant Equation (2) is discretized as: ；（4） in, This represents the upper limit of the unit's operating power. This represents the lower limit of the unit's operating power. These are the parameters for the first individual model; These are the parameters for the second monomer model; For the parameters of the third monomer model; For the first Each space cooling load unit will be in the next time step Temperature state; Representing the Individual space cooling load units at time step Temperature state; Represents the time step; Representing the Individual space cooling load units at time step Baseline power; Representing the Individual space cooling load units at time step Adjustable power; ；（5） wherein, is a monolithic bias charging power; is a correction factor; is a corrected thermal resistance; is a corrected thermal capacity; is a monolithic thermal inertia time constant; is the outdoor temperature at the th time step for the th space refrigeration type load monolith. The process of constructing a composite model of space cooling loads by integrating the SOC consistency control algorithm with the discrete physical model of individual space cooling loads is as follows: ；（6） Combining formula (6) and formula (4), we get: ；（7） wherein, represents the temperature state of the space refrigeration type load cluster at time step ; represents the regulation power of the space refrigeration type load cluster at time step ; represents the upper limit of the adjustable power of the space refrigeration type load cluster; represents the lower limit of the adjustable power of the space refrigeration type load cluster; represents the indoor temperature of the th space refrigeration type load unit at time step ; is a first aggregate model parameter; is a second aggregate model parameter; is a third aggregate model parameter; ；（8） wherein, is the baseline power for the space refrigeration type load cluster at the time step ; is the space refrigeration type load cluster bias charge power; is the space refrigeration type load cluster thermal resistance; is the space refrigeration type load cluster thermal capacitance; is the space refrigeration type load cluster thermal inertia time constant; Based on the difference between the time scale of the basic model and the actual scheduling time scale, the parameters of the multi-time scale aggregation model are updated to achieve dynamic adaptation of the multi-time scale aggregation model under multi-level scheduling cycles. Based on the updated multi-timescale aggregation model, considering both power and energy constraints, the regulation boundary is calculated, and the maximum adjustable power and maximum adjustable power of the virtual power plant under a specified regulation duration are quantitatively evaluated. 2.The space refrigeration load regulating type virtual power plant regulating characteristic and boundary modeling method according to claim 1, characterized in that, The process of using statistical measures to represent the parameters and curve parameters of the space cooling load aggregate model includes: If , , , , , , , the uniform distribution is met: ；（9） ；（10） ；（11） ；（12） wherein, represents the equivalent saturation thermal inertia time constant of the space refrigeration type load cluster; represents the total number of space refrigeration type load units in the space refrigeration type load cluster; represents the average value of the thermal inertia time constant of the space refrigeration type load units in the space refrigeration type load cluster; , respectively represent the maximum and minimum values of the thermal resistance of the space refrigeration type load units in the space refrigeration type load cluster; , respectively represent the maximum and minimum values of the refrigeration efficiency of the space refrigeration type load units in the space refrigeration type load cluster, represents the average value of the thermal capacity of the space refrigeration type load units in the space refrigeration type load cluster; , respectively represent the average values of the upper and lower limits of the allowable indoor temperature of the space refrigeration type load units in the space refrigeration type load cluster; represents the average value of the temperature set value of the space refrigeration type load units in the space refrigeration type load cluster; represents the cluster thermal resistance in the statistical quantity parameters of the space refrigeration type load cluster; represents the thermal capacity in the statistical quantity parameters of the space refrigeration type load cluster; represents the bias charging power in the statistical quantity parameters of the space refrigeration type load cluster; Substituting equations (9), (11), and (12) into equation (8), the method for calculating aggregate model parameters based on statistics is as follows: ；（13） ；（14） ；（15） wherein a first aggregate model parameter calculated for the statistical quantity; a second aggregate model parameter calculated for the statistical quantity; a third aggregate model parameter calculated for the statistical quantity; Normalize the set temperature and outdoor temperature curves separately: ；（16） ；（17） wherein, is a set temperature state; is an outdoor temperature state; Calculate cluster baseline power: ；（18） wherein, Cluster baseline power calculated for the statistical quantity; Mean value of the temperature state set for the cluster; Mean value of the outdoor temperature state. 3.The space refrigeration load regulating type virtual power plant regulating characteristic and boundary modeling method according to claim 2, characterized in that, The parameters of the aggregate model calculated based on the individual parameter statistics of space cooling loads include: Calculate cluster statistics parameters based on sampled individual parameters of space cooling loads; Calculate aggregate model parameters based on sample statistics; Calculate cluster power regulation boundary based on survey parameters; Cluster baseline power is calculated based on multidimensional outdoor temperature curves. 4.The method of claim 3, wherein, The process of calculating cluster statistics parameters based on sampled individual parameters of space-cooled loads includes: Determine the sampling ratio The sample size is calculated as follows: A random sampling method without replacement is used, and for each sampled object, the upper and lower limits of the individual unit's operating power are determined. and Refrigeration efficiency The upper and lower limits of the room temperature and Correction coefficients can be calculated. Set the discretization time step. Record individual operation indivual Indoor temperature Outdoor temperature Operating power Construct the matrix: ；（19） wherein is the temperature difference vector; ；（20） wherein, is a feature matrix; is a first is an outdoor temperature at a first is an indoor temperature at a first is a first is an operating power of the first is an operating power of the first represents a contribution of the indoor temperature to the temperature change; represents a contribution of the outdoor temperature to the temperature change; represents a contribution of the refrigeration power to the temperature change; Computing the thermal resistance of a cell , heat capacity : ；（21） wherein, represents an influence coefficient of the indoor temperature on the temperature change; represents an influence coefficient of the outdoor temperature on the temperature change; represents an influence coefficient of the refrigeration power on the temperature change; Further, a monomer thermal inertia time constant is calculated , the corrected heat capacity , the baseline power ; The user set temperature is investigated, and the uniform set temperature state value is calculated as , and the sampling user temperature set value is calculated. Statistical maximum and minimum values of the sampling sample parameter calculation , and the adjustment boundary factor are: ；（22） wherein, , are respectively the maximum and minimum values of the individual thermal resistances of the cluster of space refrigeration type loads; , are respectively the maximum and minimum values of the individual refrigeration efficiencies of the cluster of space refrigeration type loads; is the minimum value of the power up ability factor statistic; is the maximum value of the power down ability factor statistic; The statistical mean of the values of the calculation , , , , , ​ ；（23） The process of calculating aggregate model parameters based on sample statistics includes: Substituting equation (10) into equation (9) gives , , , , , Substituting equation (10) into equation (9) gives ; Will Substituting into equation (9), we get Then Substituting further into equation (13), we obtain the aggregate model parameters. ; Will , , , , Substituting into equation (11), we get Then Substituting further into equation (14), we obtain the aggregate model parameters. ; Will , , , , , Substituting into equation (12), we get Then , Substituting into equation (15), we obtain the aggregate model parameters. . 5.The space refrigeration load regulating type virtual power plant regulating characteristic and boundary modeling method according to claim 4, characterized in that, The process of calculating the cluster power adjustment boundary based on the investigation parameters includes: calculating the cluster power adjustment boundary based on the maximum value of the statistical quantity of the adjustment boundary factor and the minimum value ​ ；（24）。 6.The method of claim 5, wherein, The process of calculating cluster baseline power based on multidimensional outdoor temperature curves includes: Obtaining the outdoor temperature curve of the region where the space cooling type load is located, including the highest temperature curve , the lowest temperature curve , and the average temperature curve ; substituting formula (17) to calculate the normalized SOC corresponding to each outdoor temperature curve: ；（25） wherein, is the outdoor temperature state under the model of taking the average value of the outdoor temperature curve, is the outdoor temperature state under the model of taking the maximum value of the outdoor temperature curve; is the outdoor temperature state under the model of taking the minimum value of the outdoor temperature curve. Substitute , , into equation (18) to calculate the cluster baseline power corresponding to each temperature curve, respectively: ；（26） wherein, Baseline power and , , , , The aggregated parameter is a mean value model describing a cluster of space cooling type loads, Baseline power and , , , , The aggregated parameter is a maximum value model describing a cluster of space cooling type loads, Baseline power and , , , , The aggregated parameter is a minimum value model describing a cluster of space cooling type loads. 7.The method of claim 6, wherein, Based on the difference between the time scale of the base model and the actual scheduling time scale, the process of updating the parameters of the multi-time-scale aggregation model to achieve dynamic adaptation of the multi-time-scale aggregation model under multi-level scheduling cycles includes: Acquiring actual scheduling time steps Discretizing time steps of the base model , calculating the ratio of them That is, ； If , then directly replace with , and the polymer model parameters remain the same, i.e. ；（27） then the updated first aggregated model parameters; the updated second aggregated model parameters; the updated third aggregated model parameters; The updated state transition equation is: ；（28） , Compared with baseline power The numerical values ​​remain unchanged; the discretization time step of the original curve needs to be adjusted accordingly. Every Linear interpolation is performed to obtain the updated upper and lower limit curves and baseline power; If , then the aggregate model parameters need to be updated according to , the specific updating method is: ；（29） The updated state transition equation is: ；（30） , With baseline power The numerical value is constant, and the original curve needs to be sampled every to obtain the updated upper and lower limit curves and the baseline power. 8.The method of claim 7, wherein, Based on the updated multi-timescale aggregation model, and considering both power and energy constraints, the process of calculating the regulation boundary and quantitatively evaluating the maximum adjustable power and maximum adjustable power of the virtual power plant under a specified regulation duration includes: Space cooling loads are subject to power regulation constraints during operation. The total power regulation constraint of the cluster is: ；（31） wherein, is an upper limit of the cluster power regulation capability under the power regulation constraint; is a lower limit of the cluster power regulation capability under the power regulation constraint. During power regulation, space cooling loads should meet the normalized SOC constraint: ；（32） Determining the length of the duration of the regulation capability ; and determining an upper limit that the SOC needs to meet and a lower limit ; calculating an upper limit of the regulation capacity at each time point corresponding to the duration length of the regulation capacity and a lower limit of the regulation capacity : ；（33） Within the constraints of energy and power, the power regulation boundary of the virtual power plant for a specified regulation duration is: ；（34） wherein, is the upper limit of the total power regulation capability of the virtual power plant for a specified regulation duration; is the lower limit of the total power regulation capability of the virtual power plant for a specified regulation duration.

Images