Control method and device for decentralized variable frequency air conditioner participating in power grid demand response

Abstract

The present application discloses a control method and device for decentralized variable frequency air conditioners to participate in grid demand response. In this method, user air conditioners are used as grid demand side loads. Since the air conditioner load has the characteristics of high flexibility that can be adjusted at any time, high aggregation of single-type load resources, and real-time control of the adjustment range, it meets the conditions for participating in the flexible interaction of the grid. In addition, each user air conditioner is first clustered to form a user cluster air conditioner, and the power adjustment amount of each user cluster air conditioner is determined based on the temperature difference of each user cluster air conditioner and the deep reinforcement learning algorithm, so as to compensate the power release value of the air conditioner load caused by the power adjustment amount to the grid power. In this way, while ensuring the peak load regulation of the grid demand response, the actual cooling temperature is maintained within the specified range, and the grid demand response capability is finally improved.

Description

Technical Field

[0001] The present application relates to the technical field of power system dispatching, and in particular to a control method and device for a distributed variable frequency air conditioner participating in power grid demand response. Background Art

[0002] The rapid development of renewable energy such as wind power and photovoltaics has reduced society's dependence on traditional fossil energy, but the randomness and volatility of their output have brought new challenges to the safe and stable operation of power systems with a high proportion of new energy access.

[0003] In the related art, conventional grid loads such as district cooling systems usually have difficulty in controlling grid frequency, which limits the grid's demand response capabilities. Summary of the invention

[0004] In view of this, the present application provides a control method and device for a distributed variable frequency air conditioner to participate in grid demand response, which can improve the grid demand response capability.

[0005] In a first aspect, the present application provides a control method for a distributed variable frequency air conditioner to participate in grid demand response, comprising:

[0006] Clustering each user's air conditioner according to the location information of the user's air conditioner to obtain a user cluster air conditioner;

[0007] Determine the temperature difference between the current indoor temperature and the set temperature of each user cluster air conditioner;

[0008] According to the temperature difference, determining the power adjustment amount of the air conditioner of each user cluster based on a deep reinforcement learning algorithm;

[0009] In response to a peak load shaving instruction for the air conditioning load side composed of all user air conditioners, the power release value of the air conditioning load caused by the power adjustment amount is compensated to the power grid power.

[0010] Optionally, determining the temperature difference includes:

[0011] The actual rate of temperature change of a building within a district cooling system is determined by the following formula,

[0012]

[0013] In the above formula, is the heat transfer coefficient of the room envelope, is the area of ​​the room enclosure,

[0014] is the real-time outdoor temperature, It is the real-time room temperature. is the total heat of indoor heat sources, is the cooling power of the air conditioner, is the density of air, is the specific heat capacity of air, is the volume flow rate of ventilation, is the heat capacity of the room air;

[0015] The integral of the actual temperature change rate during the secondary frequency modulation period of the power grid is established as the temperature difference.

[0016] Optionally, clustering each user's air conditioner according to the location information of the user's air conditioner to obtain a user cluster includes:

[0017] Get the geographic coordinates and altitude of the user's air conditioner location information;

[0018] Determine the geographical distance between the user air conditioners according to the geographical coordinates and the altitude;

[0019] Determine a distance weight matrix according to the matrix corresponding to the geographic distance;

[0020] Hierarchical clustering is performed through the distance weight matrix to obtain user clusters.

[0021] Optionally, the distance weight matrix is ​​determined according to the matrix corresponding to the geographic distance, and is performed by the following formula:

[0022] ;

[0023] In the above formula, yes OK The distance matrix of the columns, yes OK The distance weight matrix of the columns, is the attenuation coefficient.

[0024] Optionally, hierarchical clustering is performed through a distance weight matrix to obtain user clusters, including:

[0025] Establish each user's air conditioner as a cluster;

[0026] According to the distance weight matrix, the average distance between all pairs of points in the cluster is established as the distance between each pair of clusters;

[0027] Merge the two clusters with the smallest distance into a new cluster;

[0028] Update the distance weight matrix and determine the distance weight matrix between the new cluster and other clusters;

[0029] Perform the operation "According to the distance weight matrix, the average distance between all point pairs in the cluster is established as the distance between each pair of clusters; the two clusters with the smallest distance are merged into a new cluster; the distance weight matrix is ​​updated, and the distance weight matrix between the new cluster and other clusters is determined" until all points are aggregated into one cluster, until all points are aggregated into one cluster or the preset number of clusters is reached.

[0030] Optionally, the determination of the number of clusters includes:

[0031] Get the merge distance when merging clusters, and multiple merge distances form a clustering tree;

[0032] Determine the distance change amount according to the merged distance;

[0033] The cluster tree is cut at the step corresponding to the maximum distance change to determine the number of clusters.

[0034] Optionally, according to the temperature difference, determining the power adjustment amount of the air conditioner in each user cluster based on a deep reinforcement learning algorithm includes:

[0035] Determine the temperature difference of each user cluster air conditioner as a state;

[0036] Establishing the power adjustment amount of each user cluster air conditioner as an action;

[0037] Construct a reward as expressed in the following formula ,

[0038]

[0039] In the above formula, is the power adjustment amount of the air conditioner in each user cluster, It is The comfort temperature set by each user cluster, is the number of user cluster air conditioners, , is the weight coefficient, is the time step;

[0040] Maximize the cumulative reward by determining and strategy network to obtain the power adjustment amount of the air conditioner in each user cluster.

[0041] In a second aspect, the present application provides a control device for a distributed variable frequency air conditioner to participate in grid demand response, comprising:

[0042] A clustering module, used to cluster each user's air conditioner according to the location information of the user's air conditioner to obtain a user cluster air conditioner;

[0043] A first determination module is used to determine the temperature difference between the current indoor temperature and the set temperature of the air conditioner of each user cluster;

[0044] A second determination module is used to determine the power adjustment amount of the air conditioner of each user cluster based on the temperature difference and a deep reinforcement learning algorithm;

[0045] The compensation module is used to respond to the peak load instruction for the air conditioning load side composed of all user air conditioners, and compensate the air conditioning load power release value caused by the power adjustment amount to the grid power.

[0046] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the above method when the computer program is executed by a processor.

[0047] In a fourth aspect, the present application provides an execution device, comprising a processor and a memory, wherein the processor is coupled to the memory;

[0048] The memory is used to store programs;

[0049] The processor is used to execute the program in the memory, so that the execution device executes the above method.

[0050] In the method disclosed in the present application, the user air conditioner is used as the load on the demand side of the power grid. Since the air conditioner load has the characteristics of high flexibility that can be adjusted at any time, high aggregation of single-type load resources, and real-time control of the adjustment range, it meets the conditions for participating in the flexible interaction of the power grid. In addition, the air conditioners of each user are first clustered to form user cluster air conditioners, and the power adjustment amount of each user cluster air conditioner is determined based on the temperature difference of each user cluster air conditioner and the deep reinforcement learning algorithm, so as to compensate the power release value of the air conditioner load caused by the power adjustment amount to the power grid power, thereby ensuring that the power grid demand response peaks while maintaining the actual cooling temperature within the specified range, and ultimately improving the power grid demand response capability. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The technical solution and other beneficial effects of the present application will be made apparent by describing in detail the specific implementation methods of the present application in conjunction with the accompanying drawings.

[0052] Figure 1 An operational flow chart of a control method for a decentralized variable frequency air conditioner participating in grid demand response provided by an exemplary embodiment is shown.

[0053] Figure 2 A schematic diagram of a topology connection diagram of a distributed variable frequency air conditioner participating in grid demand response provided by an exemplary embodiment is shown.

[0054] Figure 3 A power curve diagram of a decentralized variable frequency air conditioner provided by an exemplary embodiment when participating in grid demand response is shown.

[0055] Figure 4 A comparison diagram of a regional cooling system participating in secondary frequency regulation of a power grid and a regional cooling system not participating in secondary frequency regulation of a power grid provided by an exemplary embodiment is shown.

[0056] Figure 5 A relationship diagram of indoor temperature changes when a decentralized variable frequency air conditioner participates in grid demand response is shown according to an exemplary embodiment.

[0057] Figure 6 A graph showing the relationship between the cooling capacity, power and energy efficiency ratio of a distributed variable frequency air conditioner as a function of frequency when the air conditioner participates in grid demand response is shown in an exemplary embodiment.

[0058] Figure 7 A block diagram of a control device for a distributed variable frequency air conditioner participating in grid demand response provided by an exemplary embodiment is shown.

[0059] Figure 8 A block diagram of the execution device provided by an exemplary embodiment is shown. DETAILED DESCRIPTION

[0060] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present application, so the present application is not limited by the specific embodiments disclosed below.

[0061] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or modules is not necessarily limited to those steps or modules clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices. The naming or numbering of steps in this application does not mean that the steps in the method flow must be executed in the time / logical sequence indicated by the naming or numbering. The process steps that have been named or numbered can change the execution order according to the technical purpose to be achieved, as long as the same or similar technical effects can be achieved. The division of units in this application is a logical division. There may be other division methods when it is implemented in actual applications. For example, multiple units can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection between units can be electrical or other similar forms, which are not limited in this application. In addition, the units or sub-units described as separate components may or may not be physically separated, may or may not be physical units, or may be distributed in multiple circuit units, and some or all of the units may be selected according to actual needs to achieve the purpose of the present application.

[0062] Please refer to Figure 1 ,in Figure 1 The flowchart of a control method for a distributed variable frequency air conditioner to participate in grid demand response is shown in an exemplary embodiment. The method is implemented through steps 102-108.

[0063] In step 102, the air conditioners of each user are clustered according to the location information of the air conditioners of the user to obtain user cluster air conditioners.

[0064] From the power adjustment mentioned above, it can be directly and undoubtedly determined that the air conditioner involved here is a variable frequency air conditioner. The variable frequency air conditioner can adjust the compressor speed by changing the compressor power supply frequency in a control mode that avoids starting and stopping. The purpose of controlling the room temperature is achieved by relying on the speed of the compressor.

[0065] Here, the execution subject of the collection of user air conditioner location information can be a control terminal in the form of a server that realizes communication connection with each user's air conditioner. It can be imagined that since the air conditioner itself is not equipped with hardware with positioning function, the air conditioner location information can be obtained through a mobile terminal such as a mobile phone that realizes a short-range connection with the user's air conditioner, such as the same WiFi connection or infrared connection. Since WiFi connection or infrared connection is limited to short-range connection, the communication distance of short-range connection is generally about ten meters. Therefore, the mobile terminal and the user's air conditioner are in the same position, and the location information of the mobile terminal represents the user's air conditioner location information. Specifically, the mobile terminal can be realized through an air conditioner remote control APP that has been widely used in the field and is installed on the mobile terminal. The relevant characteristic parameters of the air conditioner, such as the air conditioner set temperature, cooling wind speed, etc., can be obtained on the air conditioner remote control APP.

[0066] The data collected by the aforementioned control terminal may include not only the location information of each user's air conditioner, but also data such as real-time temperature, set temperature and power.

[0067] The present application can effectively reduce the processing amount of subsequent steps 104 and 106 through clustering processing. In addition, data sharing in the vicinity of user air conditioners can also be achieved.

[0068] As an exemplary specific example of clustering processing, clustering each user's air conditioner according to the location information of the user's air conditioner to obtain a user cluster includes:

[0069] Step 102A1: Obtain the geographic coordinates and altitude included in the location information of the user's air conditioner.

[0070] Specifically, the geographic coordinates (latitude and longitude) of each user are collected. ,longitude and height ):

[0071] (1)

[0072] In formula (1), is the total number of users, Is a user The latitude (in degrees, ranging from (Antarctica) to (Arctic)), Is a user The longitude of (west) to (east)), Is a user The height of the .

[0073] Step 102A2: Determine the geographical distance between the user's air conditioners according to the geographical coordinates and the altitude.

[0074] Specifically, geographical distance This can be done using the following formula, (2)

[0075] In formula (2), is the radius of the earth (about 6371 km), and other parameters are detailed in formula (1).

[0076] Step 102A3: Determine a distance weight matrix according to the matrix corresponding to the geographic distance.

[0077] Specifically, the distance weight matrix is ​​determined according to the matrix corresponding to the geographic distance, and is performed by the following formula:

[0078] (3);

[0079] In the above formula, yes OK The geographical distance of the column is a matrix, which is used to control the influence of distance on weight. yes OK The distance weight matrix of the columns, is the attenuation coefficient.

[0080] Step 102A4: Perform hierarchical clustering using a distance weight matrix to obtain user clusters.

[0081] Specifically, the process of hierarchical clustering includes the following steps:

[0082] 102A41. Establish each user's air conditioner as a cluster.

[0083] 102A42. According to the distance weight matrix, the average distance between all point pairs in the cluster is established as the distance between clusters.

[0084] Here, the distance between each pair of clusters is expressed in cluster and Cluster For example, the distance between the two Determined by the following formula,

[0085] (4)

[0086] In formula (4), and Clusters and Cluster The number of points in yes OK The distance weight matrix of the columns.

[0087] 102A43. Merge the two clusters with the smallest distance into a new cluster.

[0088] 102A44. Update the distance weight matrix and determine the distance weight matrix between the new cluster and other clusters.

[0089] 102A45. Execute operations 102A42, 102A43, and 102A44 until all points are aggregated into one cluster, or until all points are aggregated into one cluster or a preset number of clusters is reached.

[0090] In this step 102A45, the determination of the preset number of clusters can be performed according to the following steps:

[0091] 102A451. Get the merge distance when merging clusters, and multiple merge distances constitute a clustering tree.

[0092] In the hierarchical clustering process, each time a cluster is merged, the merged distance is recorded. These merged distances form a cluster tree, in which each branch point corresponds to a cluster merge.

[0093] 102A452. Determine the distance change based on the merge distance.

[0094] Located in The distance at the time of the first merge is , then the distance change in the clustering process can be expressed as:

[0095] (5)

[0096] In formula (5): It is The distance at the time of the first merger; It is The distance change after the merge.

[0097] 102A453. Cut the clustering tree at the step corresponding to the maximum distance change to determine the number of clusters.

[0098] By analyzing the distance change during clustering , find the merging step corresponding to the largest distance change, which usually means that cutting here can obtain better clustering results.

[0099] Specifically, suppose that there are Merge operations, the merge distance is , the distance change is , then the step corresponding to the maximum distance change is for:

[0100] (6)

[0101] In formula (6), It is a function that finds the parameters (sets) of a function. It is The distance at the time of the merger, It is The distance at the time of the merger, It is The distance change after the merge.

[0102] Optimal number of clusters for, (7).

[0103] Finally get The result of the division of clusters: .

[0104] In step 104, the temperature difference between the current indoor temperature and the set temperature of the air conditioner of each user cluster is determined.

[0105] Here, as discussed above, the temperature difference It is equal to the real-time indoor temperature value minus the set temperature value, expressed as,

[0106] (8)

[0107] In the above formula (8), For the The temperature setting value corresponding to the building with user cluster air conditioning, No. The real-time indoor temperature value corresponding to the building with air conditioning for each user cluster.

[0108] In a preferred embodiment, determining the temperature difference includes:

[0109] Step 1041, determine the actual temperature change rate of the building in the regional cooling system according to the following formula:

[0110] (9)

[0111] In the above formula, is the heat transfer coefficient of the room envelope, is the area of ​​the room enclosure,

[0112] is the real-time outdoor temperature, It is the real-time room temperature. is the total heat of indoor heat sources, is the cooling power of the air conditioner, is the density of air, is the specific heat capacity of air, is the volume flow rate of ventilation, is the heat capacity of the room air;

[0113] In order to facilitate the understanding of the above formula (9), the derivation process of the above formula (9) is now described:

[0114] In order to express the relationship between the temperature change in the room and factors such as air conditioning power and outdoor temperature under different outdoor temperatures and different building structures (such as building orientation and house structure), we established a house thermodynamic model in the air conditioning demand response regulation system.

[0115] The temperature change in the room can be expressed by the heat balance equation:

[0116] (10)

[0117] In formula (10): is the temperature in the room, is the time (seconds), is the real-time heat entering the room (watts, ), is the heat leaving the room in real time ( ), is the heat capacity of the room ( )

[0118] The heat sources in the room mainly include external temperature conduction, indoor heat sources (such as electrical appliances, human body), etc.:

[0119] (11)

[0120] In formula (11): is the heat transfer coefficient of the room envelope, is the area of ​​the room enclosure, It is the real-time room temperature. is the real-time outdoor temperature, It is the total heat of indoor heat sources.

[0121] The heat consumption in the room mainly includes the cooling effect of the air conditioner and ventilation:

[0122] (12)

[0123] In formula (12): is the cooling power of the air conditioner, It is the heat taken away by ventilation.

[0124] The heat removed by ventilation can be expressed as:

[0125] (13)

[0126] In formula (13): is the density of air, is the specific heat capacity of air, usually taken as 1005 , is the volume flow rate of ventilation.

[0127] Combining the above formulas (11)-(13), the above formula (9) can be derived.

[0128] Step 1042: Establish the integral of the actual temperature change rate within the grid response time period as the temperature difference.

[0129] The process of step 1042 is represented as follows: (14);

[0130] In formula (14): is the temperature difference, is the grid demand response time, is the temperature in the room.

[0131] In step 106, according to the temperature difference, a power adjustment amount of the air conditioner of each user cluster is determined based on a deep reinforcement learning algorithm.

[0132] Specifically, according to the temperature difference, the power adjustment amount of the air conditioner of each user cluster is determined based on the deep reinforcement learning algorithm, including:

[0133] Step 1061: Determine the temperature difference of the air conditioners of each user cluster as a state.

[0134] state It is expressed as:

[0135] (15)

[0136] In formula (15): is the temperature difference.

[0137] Step 1062: Establish the power adjustment amount of the air conditioner in each user cluster as an action.

[0138] Specifically, the time step From 1- All air conditioning power adjustments are determined as actions ,

[0139] (16).

[0140] Step 1063: Construct a reward represented by the following formula: ,

[0141] (17)

[0142] In formula (17), is the power adjustment amount of the air conditioner in each user cluster, It is The comfort temperature set by each user cluster, It is the real-time room temperature. is the number of user cluster air conditioners, , is the weight coefficient, is the time step.

[0143] Step 1064: Maximize the cumulative reward by determining and strategy network to obtain the power adjustment amount of the air conditioner in each user cluster.

[0144] Specifically, the policy network Using deep neural network to implement Generate Action The probability distribution of . Its structure is as follows: Input layer: state ; Hidden layer: Several fully connected layers, each containing Activation function output layer: action The probability distribution of Activation function.

[0145] The policy gradient algorithm is used to update the parameters of the policy network , by maximizing the cumulative reward :

[0146] (18)

[0147] In formula (18), is the discount factor, usually a constant between 0 and 1, Indicates expected value.

[0148] And the maximum reward Gradient update formula:

[0149] (19)

[0150] In formula (19): The parameter is strategy, From the time step The cumulative reward at the beginning, T is the total time step.

[0151] In the specific example of step 1064, the rule is followed in the process of maximizing the cumulative reward: the adjusted temperature exceeds the maximum temperature value allowed by the user. That is, (20) For the Initial indoor temperature of the building for user cluster air conditioning, For the The maximum temperature allowed in the building where the user cluster air conditioning is used. For the first Temperature difference of buildings conditioned by user clusters.

[0152] In step 108, in response to the peak load shaving instruction for the air conditioning load side composed of all user air conditioners, the power release value of the air conditioning load caused by the power adjustment amount is compensated to the grid power.

[0153] In this step 108, the implementation of compensation is known to those skilled in the art. Figure 2 , 3 Indicated.

[0154] Please refer to Figure 4-6 ,It can be seen from the figure that the participation of ,distributed variable frequency air conditioner in grid demand response ,control can significantly improve the power on the load side of the grid while still maintaining ,stability.

[0155] Please refer to Figure 7 , which shows a block diagram of a control device for a distributed variable frequency air conditioner participating in grid demand response provided by the present application. The device 200 includes:

[0156] A clustering module 202, for clustering each user's air conditioner according to the location information of the user's air conditioner to obtain a user cluster air conditioner;

[0157] A first determination module 204 is used to determine the temperature difference between the current indoor temperature and the set temperature of the air conditioner of each user cluster;

[0158] A second determination module 206 is used to determine the power adjustment amount of the air conditioner of each user cluster based on the temperature difference and based on a deep reinforcement learning algorithm;

[0159] The compensation module 208 is used to respond to the peak load instruction for the air conditioning load side composed of all user air conditioners, and compensate the power release value of the air conditioning load caused by the power adjustment amount to the grid power.

[0160] Since the above methods have been discussed in detail, the specific implementation of the above modules will not be repeated here.

[0161] Next, an execution device provided by an embodiment of the present application is introduced. Figure 8 , Figure 8This is a schematic diagram of a structure of an execution device provided in an embodiment of the present application. The execution device 300 can be specifically manifested as an autonomous driving vehicle, a mobile phone, a tablet, a laptop computer, a desktop computer, a monitoring data processing device, etc., which is not limited here. Figure 1 The functions of the execution device in the corresponding embodiment. Specifically, the execution device 300 includes: a receiver 301, a transmitter 302, a processor 303 and a memory 304 (wherein the number of the processor 303 in the execution device 300 can be one or more, Figure 8 In the example of FIG. 3 , the processor 303 may include an application processor 3031 and a communication processor 3032. In some embodiments of the present application, the receiver 301, the transmitter 302, the processor 303 and the memory 304 may be connected via a bus or other means.

[0162] The memory 304 may include a read-only memory and a random access memory, and provides instructions and data to the processor 303. A portion of the memory 304 may also include a non-volatile random access memory (NVRAM). The memory 304 stores processor and operation instructions, executable modules or data structures, or subsets thereof, or extended sets thereof, wherein the operation instructions may include various operation instructions for implementing various operations.

[0163] The processor 303 controls the operation of the execution device. In a specific application, the various components of the execution device are coupled together through a bus system, wherein the bus system includes not only a data bus but also a power bus, a control bus, and a status signal bus, etc. However, for the sake of clarity, various buses are referred to as bus systems in the figure.

[0164] The method disclosed in the above embodiment of the present application can be applied to the processor 303, or implemented by the processor 303. The processor 303 can be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 303. The above processor 303 can be a general processor, a digital signal processor (digital signal processing, DSP), a microprocessor or a microcontroller, and can further include an application specific integrated circuit (application specific integrated circuit, ASIC), a field programmable gate array (field-programmable gate array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The processor 303 can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiments of the present application. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor to execute, or it can be executed by a combination of hardware and software modules in the decoding processor. The software module may be located in a storage medium mature in the art, such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 304, and the processor 303 reads the information in the memory 304 and completes the steps of the above method in combination with its hardware.

[0165] The receiver 301 can be used to receive input digital or character information and generate signal input related to the relevant settings and function control of the execution device. The transmitter 302 can be used to output digital or character information through the first interface; the transmitter 302 can also be used to send instructions to the disk group through the first interface to modify the data in the disk group; the transmitter 302 can also include a display device such as a display screen.

[0166] In the embodiment of the present application, the processor 303 is used to execute Figure 1 The specific manner in which the application processor 3031 in the processor 303 performs the above steps is the same as that in the present application. Figure 1 The corresponding method embodiments are based on the same concept, and the technical effects they bring are the same as those in this application. Figure 1 The corresponding method embodiments are the same. For specific contents, please refer to the description in the method embodiments shown above in this application, which will not be repeated here.

[0167] The above is only a preferred specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed in the present application should be covered within the protection scope of the present application.

Claims

1. A control method for a distributed variable frequency air conditioner to participate in grid demand response, characterized in that: include: Clustering each user's air conditioner according to the location information of the user's air conditioner to obtain a user cluster air conditioner; Determine the temperature difference between the current indoor temperature and the set temperature of each user cluster air conditioner; According to the temperature difference, determining the power adjustment amount of the air conditioner of each user cluster based on a deep reinforcement learning algorithm; In response to a peak load regulation instruction for the air conditioning load side composed of all user air conditioners, the power release value of the air conditioning load caused by the power regulation amount is compensated to the power of the power grid; Determine the temperature difference, including: The actual rate of temperature change of a building within a district cooling system is determined by the following formula, In the above formula, is the heat transfer coefficient of the room envelope, is the area of ​​the room enclosure, is the real-time outdoor temperature, It is the real-time room temperature. is the total heat of indoor heat sources, is the cooling power of the air conditioner, is the density of air, is the specific heat capacity of air, is the volume flow rate of ventilation, is the heat capacity of the room air; Establishing the integral of the actual temperature change rate within the power grid response time period as a temperature difference; According to the temperature difference, the power adjustment amount of the air conditioner of each user cluster is determined based on a deep reinforcement learning algorithm, including: Determine the temperature difference of each user cluster air conditioner as a state; Establishing the power adjustment amount of each user cluster air conditioner as an action; Construct a reward as expressed in the following formula , In the above formula, is the power adjustment of the air conditioner in each user cluster, It is The comfort temperature set by each user cluster, It is the real-time room temperature. is the number of user cluster air conditioners, , is the weight coefficient, is the time step; Maximize the cumulative reward by determining and strategy network to obtain the power adjustment amount of the air conditioner in each user cluster.

2. The method according to claim 1, characterized in that: Cluster each user's air conditioner according to its location information to obtain user clusters, including: Get the geographic coordinates and altitude of the user's air conditioner location information; Determine the geographical distance between the user air conditioners according to the geographical coordinates and the altitude; Determine a distance weight matrix according to the matrix corresponding to the geographic distance; Hierarchical clustering is performed through the distance weight matrix to obtain user clusters.

3. The method according to claim 2, characterized in that The distance weight matrix is ​​determined according to the matrix corresponding to the geographic distance, and is executed by the following formula: ； In the above formula, yes OK The distance matrix of the columns, yes OK The distance weight matrix of the columns, is the attenuation coefficient.

4. The method according to claim 2, characterized in that: Hierarchical clustering is performed through the distance weight matrix to obtain user clusters, including: Establish each user's air conditioner as a cluster; According to the distance weight matrix, the average distance between all point pairs in the cluster is established as the distance between clusters; Merge the two clusters with the smallest distance into a new cluster; Update the distance weight matrix and determine the distance weight matrix between the new cluster and other clusters; Perform the operation "According to the distance weight matrix, the average distance between all point pairs in the cluster is established as the distance between clusters; the two clusters with the smallest distance are merged into a new cluster; the distance weight matrix is ​​updated, and the distance weight matrix between the new cluster and other clusters is determined" until all points are aggregated into one cluster, until all points are aggregated into one cluster or the preset number of clusters is reached.

5. The method according to claim 4, characterized in that The determination of the number of clusters includes: Get the merge distance when merging clusters, and multiple merge distances form a clustering tree; Determine the distance change amount according to the merged distance; The cluster tree is cut at the step corresponding to the maximum distance change to determine the number of clusters.

6. A control device for a distributed variable frequency air conditioner participating in power grid demand response, characterized in that: include: A clustering module, used to cluster each user's air conditioner according to the location information of the user's air conditioner to obtain a user cluster air conditioner; A first determination module is used to determine the temperature difference between the current indoor temperature and the set temperature of the air conditioner of each user cluster; A second determination module is used to determine the power adjustment amount of the air conditioner of each user cluster based on the temperature difference and a deep reinforcement learning algorithm; A compensation module, for responding to a peak load instruction for the air conditioning load side composed of all user air conditioners, and compensating the air conditioning load power release value caused by the power regulation amount to the grid power; Determine the temperature difference, including: The actual rate of temperature change of a building within a district cooling system is determined by the following formula, In the above formula, is the heat transfer coefficient of the room envelope, is the area of ​​the room enclosure, is the real-time outdoor temperature, It is the real-time room temperature. is the total heat of indoor heat sources, is the cooling power of the air conditioner, is the density of air, is the specific heat capacity of air, is the volume flow rate of ventilation, is the heat capacity of the room air; Establishing the integral of the actual temperature change rate within the power grid response time period as a temperature difference; According to the temperature difference, the power adjustment amount of the air conditioner of each user cluster is determined based on a deep reinforcement learning algorithm, including: Determine the temperature difference of each user cluster air conditioner as a state; Establishing the power adjustment amount of each user cluster air conditioner as an action; Construct a reward as expressed in the following formula , In the above formula, is the power adjustment of the air conditioner in each user cluster, It is The comfort temperature set by each user cluster, It is the real-time room temperature. is the number of user cluster air conditioners, , is the weight coefficient, is the time step; Maximize the cumulative reward by determining and strategy network to obtain the power adjustment amount of the air conditioner in each user cluster.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

8. An execution device, characterized in that: comprising a processor and a memory, wherein the processor is coupled to the memory; The memory is used to store programs; The processor is configured to execute the program in the memory so that the execution device executes the method according to any one of claims 1 to 5.

Images