A temperature control load cluster control method and system for a microgrid and a medium

Abstract

The application discloses a kind of temperature-controlled load cluster control method, system and medium for micro-grid, the present application includes collecting the load power consumption of micro-grid at current time and photovoltaic output, all temperature-controlled load total scheduling power is calculated in combination with scheduling requirement;For the upper object containing multiple temperature-controlled loads, the scheduling power increment needed to be provided in the next control cycle is calculated based on the total scheduling power of temperature-controlled load, and the scheduling power of each temperature-controlled load is calculated based on this, and the consensus control of temperature-controlled load cluster in the same upper object is realized.The present application can respond to the step scheduling signal of main grid, suppress the power fluctuation of photovoltaic power generation and load at the same time, realize the consensus control of temperature-controlled load cluster in different upper objects (for example building), can relieve the communication pressure of micro-grid system, and can optimize the distribution of scheduling power between buildings by considering the difference of each building parameter, effectively avoid the overline of comfortable state.

Description

Technical Field

[0001] This invention belongs to the field of temperature-statically controlled loads (TCL) regulation, specifically relating to a method, system, and medium for temperature-statically controlled load cluster control in microgrids. Background Technology

[0002] Using renewable energy in microgrids can alleviate energy and environmental problems. However, the intermittency and uncertainty of renewable energy generation limit its absorption capacity, and if left uncontrolled, it can significantly impact the main grid. Using energy storage devices such as batteries and supercapacitors is an effective way to smooth power fluctuations, but the aging and wear of these devices can affect their dispatch performance and increase the operating costs of the microgrid. Air conditioning loads in a room constitute a thermostatically controlled load (TCL). As the number of TCLs increases, their proportion of the total load also increases. By adopting appropriate control strategies, power fluctuations in the microgrid can be effectively smoothed, meeting the dispatch requirements of the main grid. A single TCL has a relatively small impact on the system; achieving power regulation requires a large number of TCLs.

[0003] Existing TCL cluster control methods include temperature-based state queuing and intelligent algorithm-based control methods. Temperature-based state queuing prioritizes TCL control based on the distance between the current temperature and the upper temperature limit, ensuring user comfort. Dynamically adjusting the TCL temperature setpoints further addresses the frequent start-stop issues of some TCLs under state queuing algorithms, reducing the impact of scheduling on TCL lifespan. Intelligent algorithm-based control methods treat TCL cluster control as a nonlinear programming problem, using intelligent algorithms to regulate each TCL, offering high computational accuracy. However, these methods can only control TCLs with the same setpoint temperature, neglecting the parameter heterogeneity of TCLs and buildings within the microgrid. Furthermore, these methods place significant computational and communication burdens on large-scale systems with wide spatial distribution. Two-dimensional state sequence modeling is an effective method for addressing parameter heterogeneity, incorporating temperature change rate information to reflect heterogeneous parameters. While this method reduces computation, it does not accurately describe the TCL state through an equivalent thermal parameter model. Furthermore, with the increasing popularity of variable frequency air conditioners (VFAC), the control methods for VFAC to participate in scheduling need further research. Existing TCL cluster control methods mainly focus on the configuration of scheduling power among loads, and user comfort is usually used as a control constraint, without achieving fair allocation among various TCLs. Summary of the Invention

[0004] The technical problem to be solved by this invention is as follows: Addressing the aforementioned problems of existing technologies, this invention provides a method, system, and medium for temperature-controlled load cluster control in microgrids. This invention can respond to step dispatch signals from the main grid, smooth power fluctuations in photovoltaic power generation and loads, and simultaneously achieve consensus control of two parameters: TCL power offset and comfort level within buildings. Dispatch can still be completed under three conditions: partial building and TCL connection / output, different set temperatures for each air conditioning load, and different parameters between buildings. Furthermore, this invention can alleviate communication pressure on the microgrid system and optimize the allocation of dispatch power between buildings by considering the differences in building parameters, effectively preventing comfort level deviations.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0006] A temperature-controlled load cluster control method for microgrids includes:

[0007] S1 collects the load power consumption P of the microgrid at the current time t. load (t) and photovoltaic output P pv (t), initialization time Δt is 0;

[0008] S2, based on the load power consumption P load (t), Photovoltaic output P pv (t) and scheduling requirements to calculate the total scheduling power P of all temperature-controlled loads. req (t);

[0009] S3, for each upper-layer object containing multiple temperature-controlled loads, is based on the total scheduling power P of the temperature-controlled loads. req (t) Calculate the scheduling power increment ΔP that the upper-level object L needs to provide in the next control cycle. L (t+Δt s );

[0010] S4, for the temperature control load in each upper-level object, calculate the scheduling power increment ΔP that the corresponding upper-level object needs to provide in the next control cycle. L (t+Δt s ) Calculate the scheduling power of each temperature-controlled load and achieve consensus control of the temperature-controlled load cluster within the same upper-level object;

[0011] S5, add the time interval Δt from the lower-level control to the time Δt. s If the new time Δt equals the preset load and photovoltaic data acquisition time interval Δt c If the condition is met, proceed to step S6; otherwise, proceed to step S2.

[0012] S6, determine whether the scheduling has ended. If the scheduling has ended, then end and exit; otherwise, jump to step S1.

[0013] Optionally, in step S2, the temperature-controlled load dispatching power P is calculated. req The function expression for (t) is:

[0014] P req (t)=P pv (t)-P dreq (t)-P load (t)-P TCL (t),

[0015] In the above formula, P dreq (t) represents the output power required for scheduling, P TCL (t) represents the total power consumed by the temperature-controlled load cluster.

[0016] Optionally, in step S3, the scheduling power increment ΔP that the upper-level object L needs to provide in the next control cycle is calculated. L (t+Δt s )include:

[0017] First, calculate the upper-level object L at t+t d Scheduling power P at time L (t+t d ):

[0018]

[0019] In the above formula, the subscripts L and M are the numbers of the upper-level objects, which are directly connected to the microgrid through the aggregator or indirectly connected to the microgrid through other upper-level objects. ΔP L (t+t d ) represents the upper-level object L at t+t d The scheduling power increment at time t d The time for a single update of the upper-level object is given by N, where w represents the coupling coefficient and N is the number of updates. bc c is the number of parent objects. LM The correlation coefficient c represents the correlation between whether there is communication between the parent object L and the parent object M. LM The value can be 1 or 0; P M (t) represents the scheduling power required by the upper-level object M at time t, P L (t) represents the scheduling power required by the upper-level object L at time t, q L The constraint coefficient q represents whether the upper-level object is directly connected to the aggregator. L The value can be 1 or 0; P req (t) represents the total dispatch power of all temperature-controlled loads at time t, P L (t+td ) represents the upper-level object L at t+t d The required scheduling power, P, at any given time L-max and P L-min β represents the maximum and minimum scheduling power that the upper-level object L can provide, respectively. L α represents the consensus-building coefficient. ML Let be the scheduling capacity proportional coefficient, and Δt s =2t d ;

[0020] Then, it will be used as t+t d As the new current time t, recalculate the upper-level object L at t+t. d Scheduling power P at time L (t+t d And as the scheduling power P that the upper-level object L needs to provide in the next control cycle. L (t+Δt s );

[0021] Finally, the scheduling power increment ΔP that the upper-level object L needs to provide in the next control cycle is calculated according to the following formula. L (t+Δt s ):

[0022] ΔP L (t+Δt s ) = P L (t+Δt s )-P L (t),

[0023] In the above formula, P L (t) represents the scheduling power required by the upper-level object L at time t.

[0024] Optionally, the upper-level object is a building, and one building corresponds to one upper-level object. The calculation function expressions for the constraint consensus coefficient and the scheduling capability ratio coefficient are as follows:

[0025]

[0026] In the above formula, β L α represents the consensus-building coefficient. ML N is the scheduling capacity ratio coefficient. L and N M This indicates the number of temperature-controlled loads that the upper-level objects L and M participate in scheduling. R represents the average high power of the temperature-controlled load. L and η L R represents the equivalent thermal resistance and thermal coefficient of the upper object L, respectively. M and η MN1 to N2 represent the equivalent thermal resistance and thermal coefficient of the upper object M, respectively. Nbc Representing the upper-level objects 1 to N respectively bc The number of temperature-controlled loads participating in the scheduling, R1 to R Nbc Representing the upper-level objects 1 to N respectively bc The equivalent thermal resistance, η1~η Nbc Representing the upper-level objects 1 to N respectively bc The coefficient of thermal conductivity.

[0027] Optionally, before step S1, a state-space model for a single temperature-controlled load, as shown in the following equation, may be established:

[0028]

[0029] In the above formula, Let x be the state variable of the i-th temperature-controlled load. i (t)=[γ i (t); β i The first derivative of γ(t)], i (t) represents the power of the temperature-controlled load for the i-th temperature-controlled load, β i (t) represents the user comfort level for the i-th temperature-controlled load, A and B are the system state matrices, W is the constant matrix, and u i (t) represents the control variable for the i-th temperature-controlled load, where the subscript i indicates the i-th temperature-controlled load, and the control variable u i (t) Using the percentage change in rated power, and having:

[0030] γ i (t)=α i (t)-α s ,

[0031]

[0032]

[0033] In the above formula, α i (t) represents the percentage of the power of the i-th temperature-controlled load at time t relative to its rated power, α s The temperature is indicated by the user-defined value T. s The corresponding percentage of rated power, T i (t) represents the temperature at time t when the i-th temperature-controlled load is operating stably, ΔT represents the temperature deviation tolerance, and η represents the thermal coefficient of the variable frequency air conditioner. For rated power, C th and R th Representing the building's heat capacity and thermal resistance respectively; Step S4 includes:

[0034] S4.1, construct the internal degree matrix D = diag{d1,d2,…,d...} based on the communication topology of the N temperature-controlled loads in the upper-level object. N}∈R N×N Where d1, d2, ..., d N R is the internal temperature of the 1st to Nth temperature-controlled loads. N×N Given a dimension, the expression for calculating the intrinsic degree is:

[0035]

[0036] In the above formula, d i Let a be the internal degree of the i-th temperature-controlled load. ij This indicates whether the i-th temperature-controlled load and the j-th temperature-controlled load have a communication connection, with the value being 1 or 0. Based on the communication topology of the N temperature-controlled loads in the upper-level object, an adjacency matrix M = [a...]. ij ]∈R N×N And construct the Laplace matrix L = {l} according to the following formula. ij}∈R N×N :

[0037] L = DM,

[0038] In the above formula, D is the interior degree matrix, M is the adjacency matrix, and l ij Let be the Laplace operator values ​​of the i-th temperature-controlled load and the j-th temperature-controlled load;

[0039] S4.2, based on the scheduling power increment ΔP required by the upper-layer object in the next control cycle. L (t+Δt s The scheduling information received by the controller for each temperature-controlled load is determined according to the following formula;

[0040]

[0041] In the above formula, Δx i (t) represents the scheduling information received by the controller from the i-th temperature-controlled load at time t, and k1 represents the number of temperature-controlled loads directly connected to the controller. The average power of the temperature-controlled load, Δt s q represents the time interval for lower-level control. i Let q be the control coefficient for the i-th temperature-controlled load. If the i-th temperature-controlled load and the controller have a direct communication connection, then q = ... i =1, otherwise q i =0;

[0042] S4.3, it is determined that the states of temperature-controlled loads with communication connections are relatively available, and the function expression controlled by the distributed static consensus protocol is as follows:

[0043]

[0044] In the above formula, u i (t) represents the control variable at time t, c is the coupling coefficient greater than 0, and L′∈R 1×2 R represents the feedback gain matrix. 1×2 x represents the dimension. i (t) represents the state variable of the i-th temperature-controlled load, a ij x represents the communication connection status between the i-th temperature-controlled load and the j-th temperature-controlled load. j (t) represents the state variable of the j-th temperature-controlled load, Δx i (t) represents the scheduling information received by the controller from the i-th temperature-controlled load; substituting the function expression of the distributed static consensus protocol control into the state-space model of a single temperature-controlled load yields the control model of the i-th temperature-controlled load:

[0045]

[0046] In the above formula, l ij Let Δx be the Laplace operator value of the i-th temperature-controlled load and the j-th temperature-controlled load. j (t) represents the scheduling information received by the controller for the j-th temperature-controlled load;

[0047] S4.4, determine the feedback gain matrix L′ and coupling coefficient c, and substitute the determined feedback gain matrix L′ and coupling coefficient c into the control model of each temperature control load to control the state of each temperature control load.

[0048] Optionally, determining the feedback gain matrix L′ in step S4.4 includes:

[0049] S101, find the solution to the linear matrix inequality shown in the following equation when P>0:

[0050] AP+PA T -2BB T <0,

[0051] In the above formula, A and B are the system state matrices, and P is the solution to the linear matrix inequality;

[0052] S102, calculate the feedback gain matrix L′ according to the following formula:

[0053] L′=-B T P -1 ,

[0054] In the above equation, P is a solution to the linear matrix inequality.

[0055] Optionally, determining the coupling coefficient c in step S4.4 includes:

[0056] S201, the condition for reaching a consensus on the state variables of each temperature-controlled load in the upper-level object is matrix A+cλ. i BL′ is the Herwitz matrix, and the matrix A+cλ i The characteristic polynomial det(sI-(A+σBL′)) of BL′ is stable, where λ i s is the non-zero eigenvalue of the Laplacian matrix L, and s is the matrix A+cλ i The eigenvalues ​​of BL′ are: I is the second-order identity matrix, A and B are the system state matrices, σ is the coupling coefficient c, and λ is the eigenvalue. i The product of , where L′ is the feedback gain matrix, and σ=cλ i = x + jy, where x and y are the coordinates of the real axis and the imaginary axis, respectively, and j is the imaginary unit;

[0057] S202, based on matrix A+cλ i The eigenvalues ​​s of BL′ establish the complex coefficient polynomial p(s) as shown in the following equation:

[0058] p(s) = s 2 +(a+jb)s+e+jd,

[0059] In the above formula, a, b, d, and e are all polynomial coefficients, j is the imaginary unit, and a, b, e ∈ R, where R is a real number;

[0060] S203, according to the necessary and sufficient condition for the stability of a complex coefficient polynomial p(s), a>0 and abd+a 2 ed 2 If the value is greater than 0, the range of values ​​for the coupling coefficient c is determined, and the value of the coupling coefficient c is determined within this range.

[0061] Optionally, after substituting the determined feedback gain matrix L′ and coupling coefficient c into the control model of each temperature-controlled load in step S4.4 to control the state of each temperature-controlled load, the state variable of any i-th temperature-controlled load after each control is:

[0062]

[0063] In the above formula, x i (t+Δt s Let x be the state variable after regulation. i (t) represents the state variable of the i-th temperature-controlled load. Let Δt be the first derivative of the state variable of the i-th temperature-controlled load. s This is the time interval for lower-level control.

[0064] Furthermore, the present invention also provides a temperature-controlled load clustering control system for microgrids, including a microprocessor and a memory interconnected thereto, wherein the microprocessor is programmed or configured to execute the temperature-controlled load clustering control method for microgrids.

[0065] Furthermore, the present invention also provides a computer-readable storage medium storing a computer program for being programmed or configured by a microprocessor to execute the temperature-controlled load cluster control method for microgrids.

[0066] Compared with existing technologies, the present invention has the following main advantages: The present invention includes collecting the current load power consumption and photovoltaic output of the microgrid, and calculating the total dispatch power of all temperature-controlled loads based on dispatch requirements; for upper-level objects containing multiple temperature-controlled loads, it calculates the dispatch power increment required for the next control cycle based on the total dispatch power of the temperature-controlled loads, and calculates the dispatch power of each temperature-controlled load accordingly, achieving consensus control of temperature-controlled load clusters within the same upper-level object. The present invention can respond to step dispatch signals from the main grid, smooth out power fluctuations in photovoltaic power generation and loads, and achieve consensus control of temperature-controlled load clusters within different upper-level objects (e.g., buildings). This alleviates the communication pressure on the microgrid system and optimizes the allocation of dispatch power between buildings by considering the differences in building parameters, effectively avoiding exceeding comfort state limits. Attached Figure Description

[0067] Figure 1 This is a schematic diagram of the basic process of the method in Embodiment 1 of the present invention.

[0068] Figure 2 This is a schematic diagram of the hierarchical control principle in Embodiment 1 of the present invention.

[0069] Figure 3 This is a schematic diagram of the topological structure between buildings in Embodiment 1 of the present invention.

[0070] Figure 4 This is a schematic diagram of the topology of the TCL inside the building in Embodiment 1 of the present invention.

[0071] Figure 5 This is a comparison chart of the interaction power curves between the microgrid and the main grid using and not using the method of Embodiment 1 of the present invention.

[0072] Figure 6 This is a comparison chart of the interaction power curves between the microgrid and the main grid using and not using the method of Embodiment 2 of the present invention.

[0073] Figure 7 The comfort index β for TCL users inside buildings 1-5 i (t) Change graph.

[0074] Figure 8 The comfort index β for TCL users inside buildings 6-10 i (t) Change graph.

[0075] Figure 9 The comfort index β for TCL users inside buildings 11-15 i (t) Change graph.

[0076] Figure 10 The comfort index β for TCL users inside buildings 16-20 i (t) Change graph. Detailed Implementation

[0077] Example 1:

[0078] like Figure 1 As shown, the temperature-controlled load cluster control method for microgrids in this embodiment includes:

[0079] S1 collects the load power consumption P of the microgrid at the current time t. load (t) and photovoltaic output P pv (t), initialization time Δt is 0;

[0080] S2, based on the load power consumption P load (t), Photovoltaic output P pv (t) and scheduling requirements to calculate the total scheduling power P of all temperature-controlled loads. req (t);

[0081] S3, for each upper-layer object containing multiple temperature-controlled loads, is based on the total scheduling power P of the temperature-controlled loads. req (t) Calculate the scheduling power increment ΔP that the upper-level object L needs to provide in the next control cycle. L (t+Δt s );

[0082] S4, for the temperature-controlled loads (TCL) in each upper-level object, calculate the scheduling power increment ΔP that the corresponding upper-level object needs to provide in the next control cycle. L (t+Δt s ) Calculate the scheduling power of each temperature-controlled load and achieve consensus control of the temperature-controlled load cluster within the same upper-level object;

[0083] S5, add the time interval Δt from the lower-level control to the time Δt. s If the new time Δt equals the preset load and photovoltaic data acquisition time interval Δt c If the condition is met, proceed to step S6; otherwise, proceed to step S2.

[0084] S6, determine whether the scheduling has ended. If the scheduling has ended, then end and exit; otherwise, jump to step S1.

[0085] like Figure 2 As shown, the temperature-controlled load cluster control method for microgrids in this embodiment includes two levels: upper-level control and lower-level control. The control object of the upper-level control is the upper-level object, and the control means is step S3; the control object of the lower-level control is the temperature-controlled load of the upper-level object, and the control means is step S4. This temperature-controlled load cluster control method for microgrids, which focuses on the cluster control of temperature-controlled loads (TCLs) in microgrids, needs to be implemented through aggregators and controllers in the microgrid composed of upper-level objects (a collection of temperature-controlled loads, which can be a building, a floor, a unit, etc.). Specifically, the aggregator establishes communication connections with the main grid, the photovoltaic power generation system, and some upper-level objects, while the controller establishes communication connections with some TCLs within the upper-level objects.

[0086] In this embodiment, step S2 calculates the temperature-controlled load scheduling power P. req The function expression for (t) is:

[0087] P req (t)=P pv (t)-P dreq (t)-P load (t)-P TCL (t), (1)

[0088] In the above formula, P dreq (t) represents the output power required for scheduling, P TCL (t) represents the total power consumed by the temperature-controlled load cluster.

[0089] In this embodiment, the data update time of the upper-layer control is t. d , there is t d =0.5Δt s Microgrids have N bc Several upper-level objects participate in the scheduling, with i = 1, 2, ..., N. bc This indicates that step S3 calculates the scheduling power increment ΔP that the upper-layer object L needs to provide in the next control cycle. L (t+Δt s This includes: First, calculating the upper-level object L at t+t d Scheduling power P at time L (t+t d ):

[0090]

[0091] In the above formula, the subscripts L and M are the numbers of the upper-level objects, which are directly connected to the microgrid through the aggregator or indirectly connected to the microgrid through other upper-level objects. ΔP L (t+t d ) represents the upper-level object L at t+t d The scheduling power increment at time t d The time for a single update of the upper-level object is given by N, where w represents the coupling coefficient and N is the number of updates. bc c is the number of parent objects. LM The correlation coefficient c represents the correlation between whether there is communication between the parent object L and the parent object M. LM The value can be 1 or 0 (c) LM =1 indicates that there is a communication link between the two buildings, c LM =0 indicates no connection); P M (t) represents the scheduling power required by the upper-level object M at time t, P L (t) represents the scheduling power required by the upper-level object L at time t, q L The constraint coefficient q represents whether the upper-level object is directly connected to the aggregator. L The value is either 1 or 0 (q) L =1 indicates that the building is directly connected to the aggregator and can receive adjustment information from the aggregator, q L =0 indicates that the building and the aggregator are not directly connected; P req (t) represents the total dispatch power of all temperature-controlled loads at time t, P L (t+t d ) represents the upper-level object L at t+t d The required scheduling power, P, at any given time L-max and P L-min β represents the maximum and minimum scheduling power that the upper-level object L can provide, respectively. L α represents the consensus-building coefficient. ML Let be the scheduling capacity proportional coefficient, and Δt s =2t d Substituting the constants, we get:

[0092]

[0093] Then, it will be used as t+t d As the new current time t, recalculate the upper-level object L at t+t. d Scheduling power P at time L (t+t d And as the scheduling power P that the upper-level object L needs to provide in the next control cycle. L (t+Δt s );

[0094] Finally, the scheduling power increment ΔP that the upper-level object L needs to provide in the next control cycle is calculated according to the following formula. L (t+Δt s ):

[0095] ΔP L (t+Δt s ) = P L (t+Δt s )-P L (t), (3)

[0096] In the above formula, P L (t) represents the scheduling power required by the upper-level object L at time t.

[0097] like Figure 3 As shown, in this embodiment, the upper-level object is a building, and one building corresponds to one upper-level object. See also... Figure 3 As can be seen, in this specific implementation, a total of 20 buildings are included, denoted as Building 1 to Building 20, of which only Buildings 6 to 10 are directly connected to the microgrid via a collector. Figure 4 As shown, a building contains 20 temperature-controlled loads, denoted as temperature-controlled load 1 to temperature-controlled load 20, of which only temperature-controlled loads 1 to 8 are directly connected to the controller. In this embodiment, the calculation function expressions for the constraint consensus coefficient and the scheduling capability ratio coefficient are:

[0098]

[0099]

[0100] In the above formula, β L α represents the consensus-building coefficient. ML N is the scheduling capacity ratio coefficient. L and N M This indicates the number of temperature-controlled loads that the upper-level objects L and M participate in scheduling. R represents the average high power of the temperature-controlled load. L and η L R represents the equivalent thermal resistance and thermal coefficient of the upper object L, respectively. M and η M N1 to N2 represent the equivalent thermal resistance and thermal coefficient of the upper object M, respectively. Nbc Representing the upper-level objects 1 to N respectively bc The number of temperature-controlled loads participating in the scheduling, R1 to R Nbc Representing the upper-level objects 1 to N respectively bc The equivalent thermal resistance, η1~η Nbc Representing the upper-level objects 1 to N respectively bcThe heat coefficient. The data update time for the aforementioned upper-level control is t. d To prevent the scheduling power of each building from diverging in the upper-level control, it is necessary to prevent the coupling coefficient w from being too large. However, an excessively small w will slow down the convergence speed of the upper-level control. Therefore, the upper-level control speed is accelerated by reducing the update time. (Set t) d =0.5Δt s That is, the lower-level control updates data every twice the time interval t. d Receive one scheduling message from the upper layer. The aforementioned β L and α ML Equations (4) and (5) fully consider the parameters of each upper-level object, namely, the number of TCLs that can participate in scheduling within the upper-level object is different, the rated power is different, the heat capacity, thermal resistance and TCL thermal coefficient of the building are different, and the influence of the heterogeneity of these parameters on the scheduling capability of the upper-level object is taken into account.

[0101] In this embodiment, before step S1, a state-space model as shown in the following formula is established for a single temperature-controlled load:

[0102]

[0103] In the above formula, Let x be the state variable of the i-th temperature-controlled load. i (t)=[γ i (t); β i The first derivative of γ(t)], i (t) represents the power of the temperature-controlled load for the i-th temperature-controlled load, β i (t) represents the user comfort level for the i-th temperature-controlled load, A and B are the system state matrices, W is the constant matrix, and u i (t) represents the control variable for the i-th temperature-controlled load, where the subscript i indicates the i-th temperature-controlled load, and the control variable u i (t) Using the percentage change in rated power, and having:

[0104] γ i (t)=α i (t)-α s (7)

[0105]

[0106]

[0107]

[0108]

[0109] In the above formula, α i(t) represents the percentage of the power of the i-th temperature-controlled load at time t relative to its rated power, α s The temperature is indicated by the user-defined value T. s The corresponding percentage of rated power, T i (t) represents the temperature at time t when the i-th temperature-controlled load is operating stably, ΔT represents the temperature deviation tolerance, and η represents the thermal coefficient of the variable frequency air conditioner. For rated power, C th and R th These represent the building's heat capacity and thermal resistance, respectively. The process of establishing a state-space model for a single temperature-controlled load (TLC) is as follows:

[0110] S101, Establish the equivalent thermal parameter model for a single TCL:

[0111]

[0112] In the above formula, i = 1, 2, ..., N, where N represents the number of TCLs in the building, and T... i (t) represents the indoor temperature at time t, T a (t) represents the ambient temperature at time t (e.g., 30 degrees Celsius in this embodiment), C th and R th These represent the building's heat capacity and thermal resistance, respectively. The rated power of the TCL is represented by η, the thermal coefficient of the VFAC (e.g., 12.5 in this embodiment), and α. i (t) represents the percentage of the rated power of the i-th TCL at time t. The equivalent thermal parameter model of a single TCL describes how its internal temperature varies with ambient temperature and operating power.

[0113] S102, set the user's temperature setting to T. s Set the user's acceptable temperature comfort range to [T]. min ,T max The temperature deviation tolerance is ΔT, and T is given. min =T s -ΔT、T max =T s +ΔT. When TCL reaches a stable operating state, dT is... i Substituting (t) / dt=0 into the equivalent thermal parameter model of a single TCL, we can obtain the steady-state operating temperature T. i (t) and rated power percentage α i The correspondence of (t):

[0114]

[0115] S103, when the temperature is the user-set value T sThe percentage of rated power corresponding to that time is set as α. s Considering the different temperature settings of various TCL models, and the varying comfort ranges for users, the TCL power offset index γ is defined. i (t) and user comfort index β i (t) is shown in equations (7) and (8). β can be obtained. i (t)∈[0,1], to ensure user comfort, the comfort state index cannot exceed this range, β i The user's most comfortable state is when (t) = 0.5. And γ i The range of (t) needs to be determined based on parameters such as thermal coefficient and thermal resistance, as well as the user's comfort status at time t.

[0116] S104, define u i (t) represents the percentage change in the rated power of the i-th TCL, i.e., the control input, and we have:

[0117]

[0118] Take x i If (t) is a state variable, then the state-space expression of a single TCL is shown in equation (6). Taking the temperature-controlled load power and user comfort as two state variables, let x... i (t)=[γ i (t); β i [t], combining equations (12) to (14) and equations (7) and (8), the detailed expression of equation (6) can be written as:

[0119]

[0120] Substituting the parameters, we get:

[0121]

[0122] Step S4 requires the use of building and TCL communication topology information, whose adjacency matrix is ​​M = [a ij ]∈R N×N Among them, when there is a communication connection between the i-th and j-th TCLs, there is a ij =1; when there is no communication connection between the two or i=j, a ij =0. The interior degree matrix of the communication graph is D = diag{d1,d2,…,d...} N}∈R N×N The Laplace matrix is ​​L = {l ij}∈R N×N .

[0123] In this embodiment, step S4 includes:

[0124] S4.1, construct the internal degree matrix D = diag{d1,d2,…,d...} based on the communication topology of the N temperature-controlled loads in the upper-level object. N}∈R N×N Where d1, d2, ..., d N R is the internal temperature of the 1st to Nth temperature-controlled loads. N×N Given a dimension, the expression for calculating the intrinsic degree is:

[0125]

[0126] In the above formula, d i Let a be the internal degree of the i-th temperature-controlled load. ij This indicates whether the i-th temperature-controlled load and the j-th temperature-controlled load have a communication connection, with the value being 1 or 0. Based on the communication topology of the N temperature-controlled loads in the upper-level object, an adjacency matrix M = [a...]. ij ]∈R N×N And construct the Laplace matrix L = {l} according to the following formula. ij}∈R N×N :

[0127] L = DM, (17)

[0128] In the above formula, D is the interior degree matrix, M is the adjacency matrix, and l ij Let be the Laplace operator values ​​of the i-th temperature-controlled load and the j-th temperature-controlled load;

[0129] S4.2, based on the scheduling power increment ΔP required by the upper-layer object in the next control cycle. L (t+Δt s The scheduling information received by the controller for each temperature-controlled load is determined according to the following formula;

[0130]

[0131] In the above formula, Δx i (t) represents the scheduling information received by the controller from the i-th temperature-controlled load at time t, and k1 represents the number of temperature-controlled loads directly connected to the controller. The average power of the temperature-controlled load, Δt s q represents the time interval for lower-level control. i Let q be the control coefficient for the i-th temperature-controlled load. If the i-th temperature-controlled load and the controller have a direct communication connection, then q = ... i =1, otherwise q i =0;

[0132] S4.3, it is determined that the states of temperature-controlled loads with communication connections are relatively available, and the function expression controlled by the distributed static consensus protocol is as follows:

[0133]

[0134] In the above formula, u i (t) represents the control variable at time t, c is the coupling coefficient greater than 0, and L′∈R 1×2 R represents the feedback gain matrix. 1×2 x represents the dimension. i (t) represents the state variable of the i-th temperature-controlled load, a ij x represents the communication connection status between the i-th temperature-controlled load and the j-th temperature-controlled load. j (t) represents the state variable of the j-th temperature-controlled load, Δx i (t) represents the scheduling information received by the controller from the i-th temperature-controlled load; substituting the function expression of the distributed static consensus protocol control into the state-space model of a single temperature-controlled load yields the control model of the i-th temperature-controlled load:

[0135]

[0136] In the above formula, l ij Let Δx be the Laplace operator value of the i-th temperature-controlled load and the j-th temperature-controlled load. j (t) represents the scheduling information received by the controller for the j-th temperature-controlled load;

[0137] S4.4, determine the feedback gain matrix L′ and coupling coefficient c, and substitute the determined feedback gain matrix L′ and coupling coefficient c into the control model of each temperature control load to control the state of each temperature control load.

[0138] According to consensus inferences, for a group of TCLs within a building, the condition is met if and only if matrix A+cλ i When BL′ is a Hurwitz matrix, all TCL state variables can reach a consensus. Where λ i These are the non-zero eigenvalues ​​of the Laplace matrix L. Considering that matrices A and B are both stable, the consensus problem with the protocol (20) can be solved by finding suitable L′ and c.

[0139] In this embodiment, determining the feedback gain matrix L′ in step S4.4 includes:

[0140] S101, find the solution to the Linear Matrix Inequalities (LMI) shown in the following equation when P>0:

[0141] AP+PA T -2BB T <0, (21)

[0142] In the above formula, A and B are the system state matrices, and P is the solution to the linear matrix inequality;

[0143] S102, calculate the feedback gain matrix L′ according to the following formula:

[0144] L′=-B T P -1 ,(twenty two)

[0145] In the above equation, P is a solution to the linear matrix inequality.

[0146] In this embodiment, determining the coupling coefficient c in step S4.4 includes:

[0147] S201, the condition for reaching a consensus on the state variables of each temperature-controlled load in the upper-level object is matrix A+cλ. i BL′ is the Herwitz matrix, and the matrix A+cλ i The characteristic polynomial det(sI-(A+σBL′)) of BL′ is stable, where λ i s is the non-zero eigenvalue of the Laplacian matrix L, and s is the matrix A+cλ i The eigenvalues ​​of BL′ are: I is the second-order identity matrix, A and B are the system state matrices, σ is the coupling coefficient c, and λ is the eigenvalue. i The product of , where L′ is the feedback gain matrix, and σ=cλ i = x + jy, where x and y are the coordinates of the real axis and the imaginary axis, respectively, and j is the imaginary unit;

[0148] S202, based on matrix A+cλ i The eigenvalues ​​s of BL′ establish the complex coefficient polynomial p(s) as shown in the following equation:

[0149] p(s) = s 2 +(a+jb)s+e+jd,(23)

[0150] In the above formula, a, b, d, and e are all polynomial coefficients, j is the imaginary unit, and a, b, e ∈ R, where R is a real number;

[0151] S203, according to the necessary and sufficient condition for the stability of a complex coefficient polynomial p(s), a>0 and abd+a 2 ed 2 If the value is greater than 0, the range of values ​​for the coupling coefficient c is determined, and the value of the coupling coefficient c is determined within this range.

[0152] In this embodiment, after substituting the determined feedback gain matrix L′ and coupling coefficient c into the control model of each temperature-controlled load in step S4.4 to control the state of each temperature-controlled load, the state variable of any i-th temperature-controlled load after each control is:

[0153]

[0154] In the above formula, x i (t+Δt s Let x be the state variable after regulation. i (t) represents the state variable of the i-th temperature-controlled load. Let Δt be the first derivative of the state variable of the i-th temperature-controlled load. s The time interval for lower-level control. After the determined feedback gain matrix L′ and coupling coefficient c are substituted into the control model of each temperature-controlled load to control the state of each temperature-controlled load, each TCL only needs to control according to the consensus protocol shown in equations (18) and (20) to meet the consensus requirements.

[0155] In this embodiment, the installed capacity of the microgrid photovoltaic power generation system in a certain region is 1200kW. Twenty buildings participate in the regulation, and each building has 20 TCLs. The equivalent heat capacity of the buildings, C... th =2kWh / ℃, equivalent thermal resistance R th =2℃ / kW; User's temperature deviation tolerance ΔT = 2℃; Ambient temperature is 30℃; VFAC's rated power The thermal coefficient η = 2.5. The unit control time Δt at the lower level. s =1s, upper layer single update time t d =0.5s, the time interval Δt for load and photovoltaic data acquisition c =1min, coupling coefficient w=0.6. The performance of the TCL system in mitigating photovoltaic and load power fluctuations under this patented scheme is verified. The example is based on the photovoltaic power output of the microgrid from 12:00 to 15:00 on a certain day and the load consumption excluding VFAC. The main grid's dispatch requirement is to transfer the interaction power P between the microgrid and the main grid. d (t) stabilizes at 25kW, with all TCLs participating in dispatch. Ultimately, the method in this embodiment determines c = 12, L′ = [-1.4366 2], and the resulting interaction power between the microgrid and the main grid is as follows: Figure 5 As shown. See also Figure 5 It can be seen that the interaction power between the unoptimized microgrid and the main grid fluctuates greatly. After optimization by the method in this embodiment, the interaction power Pd(t) stabilizes at around 25kW, indicating that the scheduling of the method in this embodiment meets the requirements of the main grid and smooths out the power fluctuations of photovoltaic power generation and load.

[0156] In summary, the method in this embodiment is implemented using aggregators and controllers within a microgrid composed of buildings. The aggregators establish communication connections with the main grid, the photovoltaic power generation system, and some buildings, while the controllers establish communication connections with some TCLs within the buildings. First, a state-space model of a single TCL is established, constructing a communication topology diagram of buildings within the microgrid and a communication topology diagram of TCLs within the buildings, and determining the load and photovoltaic data acquisition time interval Δt. cThe time interval Δt between the lower-level control and the lower-level control s Then, the current load power consumption P is collected using an aggregator. load (t), Photovoltaic output P pv (t), and calculate the TCL dispatch power P based on load power consumption, photovoltaic power output and grid dispatch requirements. req (t); Finally, the upper-level aggregator calculates the real-time dispatch power of each building based on the restraint control and transmits the result to the lower-level controller. The lower-level controller calculates the real-time dispatch power of each TCL according to the consensus protocol. The implementation of the method of this invention can realize the consensus control of the two parameters of TCL power offset and comfort within the building, enabling the microgrid to respond to various dispatch signals from the main grid and accurately and effectively smooth out power fluctuations of photovoltaic power generation and load.

[0157] Furthermore, this embodiment also provides a temperature-controlled load clustering control system for microgrids, including a microprocessor and a memory interconnected thereon. The microprocessor is programmed or configured to execute the aforementioned temperature-controlled load clustering control method for microgrids. Additionally, this embodiment also provides a computer-readable storage medium storing a computer program for use by programming or configuring the microprocessor to execute the aforementioned temperature-controlled load clustering control method for microgrids.

[0158] Example 2:

[0159] This embodiment is basically the same as Embodiment 1, the main difference being the microgrid topology. In this embodiment, compared to Embodiment 1... Figures 2-4 Based on the topology shown, due to communication failures and user demands, the operating status of the TCLs in each building differs. Building 14 did not participate in system scheduling, one TCL in buildings 6-10 was not operating, and two TCLs in buildings 11-13 and 15 were not operating, resulting in different communication diagrams between the TCLs in each building. Additionally, the temperature setpoint Ts of the TCLs within each building also differs. Photovoltaic output and load consumption are the same as in Example 1. Finally, the method in this example determines c = 12, L′ = [-1.4366 2], and the resulting interaction power between the microgrid and the main grid is as follows: Figure 6 As shown. See also Figure 6 It can be seen that the interaction power between the unoptimized microgrid and the main grid fluctuates greatly. After optimization by the method in this embodiment, the interaction power Pd(t) stabilizes at around 25kW, indicating that the control strategy of this embodiment is still applicable under different set temperatures and communication failures.

[0160] Furthermore, this embodiment also provides a temperature-controlled load clustering control system for microgrids, including a microprocessor and a memory interconnected thereon. The microprocessor is programmed or configured to execute the aforementioned temperature-controlled load clustering control method for microgrids. Additionally, this embodiment also provides a computer-readable storage medium storing a computer program for use by programming or configuring the microprocessor to execute the aforementioned temperature-controlled load clustering control method for microgrids.

[0161] Example 3:

[0162] This embodiment is basically the same as Embodiment 1, with the main difference being the different building topologies. Based on the microgrid in Embodiment 1, this embodiment assigns different values ​​to the heat capacity, thermal resistance, and thermal coefficient of each building, distributed within the ranges of [1.5, 2.5], [1, 3], and [1.5, 3.5] respectively. The operation of the buildings and TCLs is the same as in Embodiment 2. Finally, the method in this embodiment determines c = 12, L′ = [-1.4366 2], and finally obtains the TCL user comfort index β within each building. i (t) changes as follows Figures 7-10 As shown. Among them, Figure 7 The comfort index β for TCL users inside buildings 1-5 i (t) change graph, Figure 8 The comfort index β for TCL users inside buildings 6-10 i (t) change graph, Figure 9 The comfort index β for TCL users inside buildings 11-15 i (t) change graph, Figure 10 The comfort index β for TCL users inside buildings 16-20 i (t) Change graph. See also... Figures 7-10 It can be seen that the user comfort index is within the range of [0,1]. That is, the dispatcher controls the interactive power at around 75kW while meeting comfort requirements. This is because, during stable operation, the temperature T... i (t) and percentage of rated power α i The correspondence of (t) is subject to the parameter ηR th The effect of the same power offset γ. i (t), corresponding temperature offset ΔT i Different. Parameter R th This affects the building's scheduling capabilities, leading to β. L and α ML The changes in [the parameter ηR] are considered in the power scheduling of each building in this embodiment. th This approach not only ensures user comfort but also improves the system's scheduling capabilities.

[0163] Furthermore, this embodiment also provides a temperature-controlled load clustering control system for microgrids, including a microprocessor and a memory interconnected thereon. The microprocessor is programmed or configured to execute the aforementioned temperature-controlled load clustering control method for microgrids. Additionally, this embodiment also provides a computer-readable storage medium storing a computer program for use by programming or configuring the microprocessor to execute the aforementioned temperature-controlled load clustering control method for microgrids.

[0164] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations 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, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0165] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for temperature-controlled load cluster control in microgrids, characterized in that, include: S1, collect data on the microgrid at the current moment. t load electricity P load ( t ) and photovoltaic output P pv ( t ), initialization time Δ t =0; S2, Power consumption based on load P load ( t Photovoltaic power output P pv ( t ) and scheduling requirements to calculate the total scheduling power of all temperature-controlled loads. P req ( t ); S3, for each upper-layer object containing multiple temperature-controlled loads, is based on the total power dispatched by the temperature-controlled loads. P req ( t ) Calculate the upper-level object L The scheduling power increment Δ to be provided in the next control cycle P L ( t +Δ t s ); S4, for the temperature control load in each upper-level object, calculate the scheduling power increment Δ that the corresponding upper-level object needs to provide in the next control cycle. P L ( t +Δ t s ) Calculate the scheduling power of each temperature-controlled load and achieve consensus control of the temperature-controlled load cluster within the same upper-level object; S5, time Δ t Add the time interval Δ of the lower-level control t s If the new time Δ t Equal to the preset load and photovoltaic data acquisition time interval Δ t c If the condition is met, proceed to step S6; otherwise, proceed to step S2. S6, determine whether the scheduling has ended. If the scheduling has ended, then end and exit; otherwise, jump to step S1.

2. The temperature-controlled load cluster control method for microgrids according to claim 1, characterized in that, In step S2, the power of temperature-controlled load dispatching is calculated. P req ( t The function expression for ) is: P req ( t )= P pv ( t )- P dreq ( t )- P load ( t )- P TCL ( t ) ， In the above formula, P dreq ( t The output power required for scheduling is... P TCL ( t ) represents the total power consumed by the temperature-controlled load cluster.

3. The temperature-controlled load cluster control method for microgrids according to claim 1, characterized in that, In step S3, the upper-level object is calculated. L The scheduling power increment Δ to be provided in the next control cycle P L ( t +Δ t s )include: First, calculate the upper-level object. L exist t + t d Scheduling power at any time P L ( t + t d ): ， In the above formula, the subscript L and M All of these are the identifiers of upper-level objects, which are directly connected to the microgrid via the aggregator or indirectly connected to the microgrid via other upper-level objects. Δ P L ( t + t d ) is the upper-level object L exist t + t d The scheduling power increment at any given moment. t d The update time of the upper-level object. w Represents the coupling coefficient. N bc The number of upper-level objects, c LM For the upper-level object L and upper-level objects M The correlation coefficient between them indicates whether there is communication between them. c LM The value can be 1 or 0; P M ( t ) is the upper-level object M exist t The scheduling power required at any given time. P L ( t ) is the upper-level object L exist t The scheduling power required at any given time. q L The constraint coefficient determines whether the upper-level object is directly connected to the aggregator. q L The value can be 1 or 0; P req ( t )for t Total power dispatched for all temperature-controlled loads at any given time. P L ( t + t d ) is the upper-level object L exist t + t d The required scheduling power at any given time. P L-max and P L-min Representing the upper-level object respectively L The maximum and minimum scheduling power that can be provided β L This represents the consensus-checking coefficient. α ML This is the scheduling capacity ratio coefficient, and it has Δ t s =2 t d ; Then, t + t d As the new present moment t Recalculate the parent object again. L exist t + t d Scheduling power at any time P L ( t + t d And as the upper-level object L The scheduling power required in the next control cycle P L ( t +Δ t s ); Finally, the upper-level object is calculated according to the following formula. L The scheduling power increment Δ to be provided in the next control cycle P L ( t +Δ t s ): ， In the above formula, P L ( t ) is the upper-level object L exist t The scheduling power required at any given time.

4. The temperature-controlled load cluster control method for microgrids according to claim 3, characterized in that, The upper-level object is a building, and one building corresponds to one upper-level object. The calculation function expressions for the constraint consensus coefficient and the scheduling capability ratio coefficient are as follows: ， ， In the above formula, β L This represents the consensus-checking coefficient. α ML This is the scheduling capacity ratio coefficient. N L and N M Represents the upper-level object L and upper-level objects M The number of temperature-controlled loads participating in the scheduling. This represents the average power of the temperature-controlled load. R L and η L Representing the upper-level object respectively L The equivalent thermal resistance and thermal coefficient, R M and η M Representing the upper-level object respectively M The equivalent thermal resistance and thermal coefficient, N 1~ N Nbc These represent the upper-level object 1 respectively. ～N bc The number of temperature-controlled loads participating in the scheduling. R 1~ R Nbc These represent the upper-level object 1 respectively. ～N bc The equivalent thermal resistance, η 1~ η Nbc These represent the upper-level object 1 respectively. ～N bc The coefficient of thermal conductivity.

5. The temperature-controlled load cluster control method for microgrids according to claim 1, characterized in that, Step S1 is preceded by establishing a state-space model for a single temperature-controlled load, as shown in the following equation: ， In the above formula, For the first i State variables of a temperature-controlled load x i ( t )=[ γ i ( t ); β i ( t The first derivative of )] γ i ( t ) is the first i The power of the temperature-controlled load. β i ( t ) is the first i User comfort under temperature control load A and B The system state matrix, W It is a constant matrix. u i ( t ) is the first i The control variable for a temperature-controlled load, subscript i To indicate the first i A temperature-controlled load, and a control variable u i ( t The percentage change in rated power is used, and the following applies: ， ， ， ， ， In the above formula, α i ( t )express t Time of the first i The percentage of the rated power of each temperature-controlled load. α s This indicates that the temperature is the user-defined value. T s The corresponding percentage of rated power at that time T i ( t )for t Time of the first i The temperature at which a temperature-controlled load operates stably, Δ T For temperature deviation from tolerance, η The coefficient of performance (COP) of a variable frequency air conditioner. Rated power, C th and R th Represent the building's heat capacity and thermal resistance respectively; Steps S4 includes: S4.1, based on the upper-level object N The communication topology of each temperature-controlled load constitutes an interior degree matrix. D=diag{d 1 , d 2 , …, d N } R N N ,in, d 1 , d 2 , …, d N For the 1st to N The internal temperature of a temperature-controlled load, R N N Given a dimension, the function expression for calculating the inner degree is: ， In the above formula, d i For the first i The internal temperature of a temperature-controlled load, a ij For the first i The temperature control load and the first j The communication connection status of each temperature-controlled load is checked, with a value of 1 or 0; based on the upper-level object... N The communication topology of each temperature-controlled load constructs an adjacency matrix as follows: M = [ a ij ] R N N And construct the Laplace matrix according to the following formula. L= { l ij } R N N : ， In the above formula, D Let be the interior degree matrix. M It is an adjacency matrix. l ij For the first i The temperature control load and the first j The Laplace operator value for a temperature-controlled load; S4.2, based on the scheduling power increment Δ that the upper-layer object needs to provide in the next control cycle. P L ( t+ Δ t s The scheduling information received by the controller for each temperature-controlled load is determined according to the following formula; ， In the above formula, Δ x i ( t ) is the first i A temperature-controlled load in t It constantly receives scheduling information from the controller. k 1 represents the number of temperature-controlled loads directly connected to the controller. Δ is the average power of the temperature-controlled load. t s This is the time interval for lower-level control. q i For the first i The control coefficient of the temperature-controlled load, if the first temperature-controlled load... i If the temperature-controlled load and the controller have a direct communication connection, then... q i = 1, otherwise q i = 0; S4.3, it is determined that the states of temperature-controlled loads with communication connections are relatively available, and the function expression controlled by the distributed static consensus protocol is as follows: ， In the above formula, u i ( t ) for in t Control variables at time, c The coupling coefficient is greater than 0. L R 1 2 Represents the feedback gain matrix. R 1 2 Representing dimension, x i ( t ) is the first i A state variable of a temperature-controlled load. a ij For the first i The temperature control load and the first j The communication connection status of each temperature-controlled load. x j ( t ) is the first j The state variable of a temperature-controlled load, Δ x i ( t ) is the first i Each temperature-controlled load receives scheduling information from the controller; the function expression controlled by the distributed static consensus protocol is substituted into the state-space model of a single temperature-controlled load to obtain the first... i A temperature-controlled load control model: ， In the above formula, l ij For the first i The temperature control load and the first j The Laplace operator value of a temperature-controlled load, Δ x j ( t ) is the first j Each temperature-controlled load receives scheduling information from the controller; S4.4, Determine the feedback gain matrix L and coupling coefficient c and the determined feedback gain matrix L and coupling coefficient c Substitute the control model of each temperature-controlled load into the control model to control the state of each temperature-controlled load.

6. The temperature-controlled load cluster control method for microgrids according to claim 5, characterized in that, In step S4.4, the feedback gain matrix is ​​determined. L include: S101, find the linear matrix inequality shown in the following equation. P Solution when >0: ， In the above formula, A and B The system state matrix, P The solution to the linear matrix inequality; S102, calculate the feedback gain matrix according to the following formula. L : L = - B T P -1 ， In the above formula, P This is a solution to a linear matrix inequality.

7. The temperature-controlled load cluster control method for microgrids according to claim 6, characterized in that, In step S4.4, the coupling coefficient is determined. c include: S201, the condition for consensus among the state variables of each temperature-controlled load in the upper-level object is determined by a matrix. A + cλ i BL It is a Herwitz matrix, and the matrix A + cλ i BL The characteristic polynomial det( sI -( A + σBL Stable, among which λ i It is the Laplace matrix L non-zero eigenvalues, s For matrix A + cλ i BL eigenvalues, I It is a second-order identity matrix. A and B The system state matrix, σ Coupling coefficient c and eigenvalues λ i The product of L For the feedback gain matrix, and have σ = cλ i = x + jy ,in x,y These are the coordinates of the real axis and the imaginary axis, respectively. j The imaginary unit; S202, according to the matrix A + cλ i BL The eigenvalues ​​s establish the complex coefficient polynomial shown in the following equation. p ( s ): ， In the above formula, a, b, d and e All are polynomial coefficients. j It is the imaginary unit, and has a , b , e R ,in R It is a real number; S203, based on complex coefficient polynomials p ( s Stability and necessary and sufficient conditions a >0 and abd + a 2 e - d 2 >0, determine the coupling coefficient c The range of values ​​for is used to determine the coupling coefficient within that range. c The value of .

8. The temperature-controlled load cluster control method for microgrids according to claim 6, characterized in that, The feedback gain matrix determined in step S4.4 L and coupling coefficient c After substituting the control models of each temperature-controlled load to control the state of each temperature-controlled load, any... i The state variables of a temperature-controlled load after each adjustment are: ， In the above formula, For the state variables after regulation, x i ( t ) is the first i A state variable of a temperature-controlled load. For the first i The first derivative of the state variable of the temperature-controlled load, Δ t s This is the time interval for lower-level control.

9. A temperature-controlled load cluster control system for microgrids, comprising a microprocessor and a memory interconnected, characterized in that, The microprocessor is programmed or configured to execute the temperature-controlled load cluster control method for microgrids as described in any one of claims 1 to 8.

10. A computer-readable storage medium storing a computer program, characterized in that, The computer program is used to be programmed or configured by a microprocessor to execute the temperature-controlled load cluster control method for microgrids as described in any one of claims 1 to 8.

Images