Control method of air conditioning system and related device

Abstract

The invention discloses a control method of an air conditioning system and a related device, and relates to the technical field of control, the air conditioning system comprises at least two water chilling units and a plurality of air conditioners, and the water chilling units control the temperature in a building through the air conditioners arranged in the building. The control method of the air conditioning system comprises the steps that historical operation data of the air conditioning system operating in a preset historical time period are obtained; a target load of the air conditioning system in a preset target time period is predicted according to the historical operation data through a load prediction model, and the preset target time period is later than the current moment; energy efficiency parameters of all water chilling units in the air conditioning system are determined through the energy efficiency model; and based on the target load of the air conditioning system in the preset target time period and the energy efficiency parameters of all the water chilling units in the air conditioning system, first target working parameters of all the water chilling units in the air conditioning system are determined, and the first target working parameters are used for controlling the corresponding water chilling units to work when the preset target time period is reached.

Description

Technical Field

[0001] The present application relates to the field of control technology, and in particular to a control method and related devices for an air-conditioning system. Background Art

[0002] Building HVAC systems account for 50% of a building’s total energy consumption, so even small improvements in system operating efficiency can result in significant energy savings.

[0003] The intelligent air conditioning control system is a multi-input and multi-output system with a complex control process, large hysteresis, strong coupling of components, and nonlinearity. In addition, there are usually thousands of rooms and hundreds of sub-devices in a large-scale building HVAC system, and since everyone's perception of the comfort level of the ambient temperature is subjective and different, it becomes a difficult problem to reasonably dispatch the load of the air conditioning unit.

[0004] Traditionally, a rough air conditioning operation mode is generally adopted, and the chiller outlet temperature and the start and stop status of the cooling tower, chilled water pump, and cooling pump in the air conditioning system are set according to the experience of operators or experts and factors such as outdoor temperature. This cannot accurately control the load of the air conditioning unit, resulting in failure to achieve the best control effect. Summary of the invention

[0005] In a first aspect, the present application provides a control method for an air conditioning system, wherein the air conditioning system comprises at least two chillers and a plurality of air conditioners, wherein the chillers control the temperature in a building through the air conditioners arranged in the building, and the control method for the air conditioning system comprises:

[0006] Obtaining historical operation data of the air conditioning system during a preset historical time period;

[0007] Using a load prediction model, based on the historical operation data, predicting a target load of the air conditioning system in a preset target time period, wherein the preset target time period is later than the current time;

[0008] Determine the energy efficiency parameters of each chiller in the air-conditioning system through the energy efficiency model;

[0009] Based on the target load of the air-conditioning system in a preset target time period and the energy efficiency parameters of each chiller in the air-conditioning system, the first target operating parameter of each chiller in the air-conditioning system is determined, and the first target operating parameter is used to control the operation of the corresponding water-cooling unit when the preset target time period is reached.

[0010] In a possible implementation, the energy efficiency model is constructed in the following manner:

[0011] Obtaining a sample set, wherein the sample set includes a number of sample pairs consisting of sample input parameters and sample output results;

[0012] According to the number of input nodes of the input layer in the energy efficiency model, the number of nodes of the hidden layer in the energy efficiency model is determined to obtain an initial energy efficiency model, wherein the number of input nodes matches the type of operating data in the air-conditioning system;

[0013] Processing the sample input parameters based on the initial energy efficiency model to obtain a training output result;

[0014] Based on the fact that the gap between the sample output result and the training output result satisfies a set gap condition, the number of nodes in the hidden layer is increased to obtain an adjusted energy efficiency model;

[0015] The adjusted energy efficiency model is continuously trained using the sample set until the gap between the sample output result and the training output result does not satisfy the set gap condition, thereby obtaining a constructed energy efficiency model.

[0016] In a possible implementation, determining the energy efficiency parameters of each chiller in the air-conditioning system by using the energy efficiency model includes:

[0017] Through the energy efficiency model, based on the historical operation data, the energy efficiency parameters corresponding to each chiller in the air-conditioning system under each operating condition are determined. When the load in the air-conditioning system is the same, the loads of each chiller are the same / different, and different operating conditions correspond to different loads.

[0018] In a possible implementation, the energy efficiency model is used to determine the energy efficiency parameters corresponding to each chiller in each operating condition in the air-conditioning system according to the historical operation data, including:

[0019] Determine the energy efficiency parameters of each chiller per unit time based on historical operation data;

[0020] Through the energy efficiency model, the energy efficiency parameters of each chiller in different working conditions are determined for each energy efficiency parameter of each chiller in unit time, and any working condition includes at least one unit time.

[0021] In a possible implementation, predicting the target load of the air-conditioning system in a preset target time period according to the historical operation data using a load prediction model includes:

[0022] Dividing the historical operation data into at least two categories;

[0023] By using a load forecasting model, historical operation data of at least two categories are processed to obtain a first load of each category;

[0024] Through the load prediction model, the first loads of various categories are combined to obtain the target load of the air-conditioning system in a preset target time period.

[0025] In a possible implementation, determining the first target operating parameter of each chiller in the air-conditioning system based on the target load of the air-conditioning system in a preset target time period and the energy efficiency parameter of each chiller in the air-conditioning system includes:

[0026] Determine the allocated load of each chiller according to the target load of the air conditioning system in a preset target time period and the energy efficiency parameters of each chiller through a load scheduling model;

[0027] According to the distributed load of each chiller, a first target operating parameter of each chiller is determined, and the first target operating parameter of each chiller satisfies the operating constraint condition.

[0028] In a possible implementation, the method further includes:

[0029] When the current moment reaches the preset target time period, obtaining real-time operation data of the air conditioning system;

[0030] Obtaining the real-time load of the air conditioning system according to the real-time operation data;

[0031] The second target operating parameter of each chiller in the air-conditioning system is determined according to the real-time load through the load scheduling model, so as to control the operation of the corresponding chiller based on the second target operating parameter.

[0032] A second aspect of the present application provides a control device for an air conditioning system, the air conditioning system comprising at least two chillers and a plurality of air conditioners, the chillers controlling the temperature in a building through the air conditioners arranged in the building, the control device for the air conditioning system comprising:

[0033] An acquisition module, used to obtain historical operation data of the air conditioning system during a preset historical time period;

[0034] A prediction module, configured to predict a target load of the air-conditioning system in a preset target time period according to the historical operation data through a load prediction model, wherein the preset target time period is later than the current time;

[0035] A first determination module is used to determine the energy efficiency parameters of each chiller in the air-conditioning system through an energy efficiency model;

[0036] The second determination module is used to determine the first target operating parameters of each chiller in the air-conditioning system based on the target load of the air-conditioning system in a preset target time period and the energy efficiency parameters of each chiller in the air-conditioning system. The first target operating parameters are used to control the operation of the corresponding water-cooling unit when the preset target time period is reached.

[0037] A third aspect of the present application provides a computer program product, comprising computer-readable instructions, which, when executed on an electronic device, enables the electronic device to implement the control method for an air-conditioning system according to the first aspect or any implementation of the first aspect.

[0038] A fourth aspect of the present application provides an electronic device, comprising at least one processor and a memory connected to the processor, wherein:

[0039] The memory is used to store computer programs;

[0040] The processor is used to execute the computer program so that the electronic device can implement the control method of the air-conditioning system of the above-mentioned first aspect or any implementation manner of the first aspect.

[0041] A fifth aspect of the present application provides a computer storage medium, which carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement a method for controlling an air-conditioning system according to the first aspect or any implementation of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0043] Figure 1 is a flow chart of a control method of an air conditioning system provided in an embodiment of the present application;

[0044] Figure 2 is a schematic diagram of an air conditioning system provided in an embodiment of the present application;

[0045] Figure 3 It is a schematic diagram of a process for constructing an energy efficiency model provided in an embodiment of the present application;

[0046] Figure 4 It is a schematic diagram of the energy efficiency model structure provided in the embodiment of the present application;

[0047] Figure 5 is a schematic diagram of a curve showing the relationship between the current percentage and time of a chiller provided in an embodiment of the present application;

[0048] Figure 6 It is a schematic diagram of the relationship between the normal operating current and time of the chiller provided in the embodiment of the present application;

[0049] Figure 7It is a flow chart of determining the energy efficiency parameters corresponding to each chiller in each operating condition in the air-conditioning system through the energy efficiency model according to the historical operation data provided by the embodiment of the present application;

[0050] Figure 8 It is a schematic diagram of a process of predicting the target load of the air-conditioning system in a preset target time period according to the historical operation data through a load prediction model provided by an embodiment of the present application;

[0051] Fig. 9 is a schematic diagram of various categories of items in the historical operation data provided by the embodiment of the present application;

[0052] Fig.10 It is a flow chart of determining the first target operating parameter of each chiller in the air-conditioning system based on the target load of the air-conditioning system in a preset target time period and the energy efficiency parameters of each chiller in the air-conditioning system, provided by an embodiment of the present application;

[0053] Fig.11 It is a schematic diagram of a process for fine-tuning a chiller in an air-conditioning system after reaching a preset target period, provided by an embodiment of the present application;

[0054] Fig.12 It is a schematic diagram comparing the target load predicted by the load prediction model provided in the embodiment of the present application with the actual load;

[0055] Fig.13 is a structural schematic diagram of a control device for an air conditioning system provided in an embodiment of the present application;

[0056] Fig.14 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0057] The following describes the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation method section of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0058] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0059] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0060] Figure 1 is a flow chart of a control method of an air conditioning system provided in an embodiment of the present application, such as Figure 1 As shown, a control method for an air-conditioning system provided in an embodiment of the present application may include steps 101 to 104, and these steps are described in detail below.

[0061] 101. Obtain historical operation data of the air conditioning system during a preset historical time period;

[0062] The air conditioning system includes at least two chillers and several air conditioners. The chiller controls the temperature in the building through the air conditioners installed in the building.

[0063] Among them, several air conditioners are installed in the building, and the air conditioners perform heat exchange through chillers to achieve temperature control in the building.

[0064] The chiller may include a cooling tower, a cooling pump, a refrigerator, a cold water pump, and supporting valves, water distributors, water collectors, fan coils and other structures.

[0065] Figure 2 Schematic diagram of an air conditioning system provided in an embodiment of the present application, including a chiller set 201 and an air conditioning set 202. The chiller set includes: 5 cooling towers, 4 cooling pumps, 4 refrigerators, 4 cooling water pumps, a primary water distributor, a primary water collector, 4 circulating pumps, a secondary water distributor and a secondary water collector, and matching valves. The air conditioning set includes 5 air conditioners, which are arranged in the room where the staff are located. The chiller is generally arranged in the basement and other non-staff areas. M in the figure represents a valve.

[0066] Among them, the preset historical period can be set according to actual conditions, such as the last three months, the last week, the last month, etc.

[0067] The historical operation data may include various data involved in the operation of each chiller.

[0068] As an example, the historical operation data may include cooling flow, cooling water inlet temperature, outlet temperature, power consumption, etc.

[0069] 102. Predicting the target load of the air conditioning system in a preset target time period according to the historical operation data through a load prediction model, where the preset target time period is later than the current time;

[0070] Among them, a load prediction model is preset, which can predict the load of the air-conditioning system.

[0071] Wherein, the historical operation data is used as input data, and the load prediction model can predict the target load of the air-conditioning system in a preset target time period.

[0072] Among them, the preset target time period is a specified time period later than the current moment. The length of the specified time period can be set according to actual conditions. It can be timed in hours, days, or longer units. The specific length of the preset target time period is not limited in this application.

[0073] As an example, the preset target time period may be 24 hours in the future from the current moment, a specified future day or a week, etc.

[0074] Among them, the load forecasting model can be established using a time series decomposition algorithm to predict the load demand in the future time period and obtain the target load, so as to combine the energy efficiency parameters of each chiller, fully tap the energy-saving potential, and maximize the balance between supply and demand. Its goal is to establish a quantitative or qualitative relationship between the input and predicted loads, without exploring the complex physical mechanisms within the system. It is suitable for load forecasting in the operation stage. Commonly used supervised algorithms include artificial neural networks and support vector machines.

[0075] In a possible implementation, the historical operation data can be divided into multiple categories, and the load forecasting model is used to process the historical data of each category to finally obtain the target load. Figure 8 The process of predicting the target load of the air conditioning system in a preset target time period is described in detail.

[0076] 103. Determine the energy efficiency parameters of each chiller in the air conditioning system through the energy efficiency model;

[0077] Among them, the energy efficiency model can target the energy efficiency parameters (COP, Coefficient of performance) of each chiller in the air-conditioning system under various working conditions.

[0078] The energy efficiency of the chiller is affected by many factors, such as the heat exchange medium of the chiller, outdoor meteorological conditions, operating conditions of the chiller, load distribution method of the chiller, etc. Among them, the operating parameters and load distribution method of the chiller are the key factors affecting its overall operating energy efficiency.

[0079] Among them, the factors affecting the energy efficiency parameters of the chiller include internal factors and external factors.

[0080] Among them, internal factors include chiller type, manufacturing level, compressor matching, refrigerant type and filling amount, etc. External factors include chiller water temperature and flow, cooling water temperature and flow, etc., which affect the chiller evaporation temperature and condensation temperature, as well as building load characteristics and climatic conditions, etc., which affect the unit load rate. In actual engineering, the external factors that affect the performance of the chiller have been determined and will not change easily, so only the changes in external factors need to be considered. Chiller performance is related to evaporation temperature, condensation temperature and chiller load, while the evaporation temperature of the chiller is mainly affected by the set value of the chilled water outlet temperature, and the condensation temperature is related to the cooling water inlet temperature.

[0081] In order to establish an accurate and reliable energy efficiency that can reflect the operating characteristics of the unit under various operating conditions, the impact of various factors on energy efficiency should be considered as much as possible. The energy efficiency model is the basis of operation optimization control. If the energy efficiency model includes the air conditioning system control parameters, the energy efficiency model has higher flexibility in practical applications. Therefore, the energy efficiency model can adopt a neural network model.

[0082] Among them, the energy efficiency model can determine the energy efficiency parameters of the cold water model under various working conditions based on historical operating data. In subsequent steps, it can also combine the target load of the air-conditioning system to determine energy efficiency parameters that better match the target load, so as to maximize the balance between supply and demand and reduce the energy consumption of the air-conditioning system.

[0083] In a possible implementation, the energy efficiency model uses a neural network model, which can be pre-built and trained. Figure 3 The construction process of the energy efficiency model is described in detail in .

[0084] 104. Based on the target load of the air-conditioning system in a preset target time period and the energy efficiency parameters of each chiller in the air-conditioning system, determine the first target operating parameter of each chiller in the air-conditioning system, and the first target operating parameter is used to control the operation of the corresponding chiller when the preset target time period is reached.

[0085] Among them, based on the predicted target load of the air-conditioning system in the preset target time period and combined with the energy parameters of each chiller in the air-conditioning system, the first target operating parameters adopted by each chiller in the air-conditioning system in the preset target time period are determined.

[0086] Among them, since the energy efficiency parameters of each chiller in the air-conditioning system change with the load rate, after determining the overall target load of the air-conditioning system, it is necessary to find the most suitable energy efficiency parameters in combination with the target load and the operating conditions of each chiller, and then determine the first target operating parameters of each chiller.

[0087] Correspondingly, the determined first target operating parameter of each chiller in the air-conditioning system is such that when the chiller operates at the first target operating parameter, it can match the target load and achieve the balance between supply and demand to the maximum extent.

[0088] In a possible implementation, the target load and the energy efficiency parameters of each chiller can be used to determine the total energy consumption of the air conditioning system, and then the first target operating parameters of each chiller can be determined using the load scheduling model. Fig.10 The process of determining the first target operating parameters of each chiller in the air-conditioning system is described in detail.

[0089] In a possible implementation, when the time corresponding to the target preset time period is reached, the real-time operation data of the air conditioning system is also obtained, and the operation of each chiller is adjusted using the real-time operation data to achieve real-time fine-tuning. Fig.11 This process is described in detail in .

[0090] In one possible implementation, the air-conditioning system is provided with a controller, such as a PLC (Programmable Logic Controller), which sends the first target operating parameter to the corresponding chiller when the corresponding moment of the target preset time period is reached, so as to control the chiller to operate according to the first target operating parameter.

[0091] In this embodiment, the air-conditioning system includes at least two chillers and several air conditioners. The chillers control the temperature in the building through the air conditioners installed in the building. The control method of the air-conditioning system includes: obtaining historical operation data of the air-conditioning system in a preset historical time period; predicting the target load of the air-conditioning system in a preset target time period based on the historical operation data through a load prediction model, and the preset target time period is later than the current moment; determining the energy efficiency parameters of each chiller in the air-conditioning system through an energy efficiency model; determining the first target operating parameters of each chiller in the air-conditioning system based on the target load of the air-conditioning system in the preset target time period and the energy efficiency parameters of each chiller in the air-conditioning system, and the first target operating parameters are used to control the operation of the corresponding water-cooled unit when the preset target time period is reached. The load prediction model predicts the historical operating data of the air-conditioning system and predicts the target load of the air-conditioning system in the future period. The energy efficiency model is used to determine the energy efficiency parameters of each chiller in the air-conditioning system. Combined with the target load and the energy efficiency parameters of the chiller, the first target operating parameters of the chiller in the corresponding future period are determined. When the chiller operates at the first target operating parameters, its energy efficiency is matched with the system load to maximize the balance between supply and demand and reduce the overall energy consumption of the air-conditioning system.

[0092] Figure 3 It is a flowchart of building an energy efficiency model provided in an embodiment of the present application, which may include steps 301 to 305. These steps are described in detail below.

[0093] 301. Obtain a sample set, where the sample set includes a number of sample pairs consisting of sample input parameters and sample output results;

[0094] Among them, the energy efficiency model adopts a neural network model. Since the neural network has adaptive and strong fitting capabilities, a neural network containing only one hidden layer can realize nonlinear mapping of multi-dimensional, multi-variable input and output. The hidden layer plays an abstract role, and the neural network model can extract features from the input information.

[0095] Among them, adding a hidden layer can increase the processing capacity of the artificial neural network, but it will complicate the training, increase the number of training samples and the training time, and the local minimum error will also increase, making it easier to fall into the local minimum error. Therefore, the energy efficiency model in this embodiment uses a three-layer BP (BackPropagation) neural network with one hidden layer (a multi-layer feedforward network trained by the error back propagation algorithm) to fit the relationship between the air-conditioning system load ratio, chilled water supply and return water temperature, outdoor temperature, set water outlet temperature, etc. and the unit energy efficiency, and trains the network model through back propagation.

[0096] Among them, the load ratio of the air-conditioning system can be PLR ​​(Part Load Ratio), which is used to describe the performance of air-conditioning equipment under non-design conditions. It can be the ratio of actual load to the rated capacity of the equipment. The rated capacity of the equipment refers to the cooling / heating capacity value specified on the equipment nameplate.

[0097] Among them, the sample input parameters are various parameters involved in the operation of the air-conditioning system.

[0098] As an example, the sample input parameters include the air conditioning system load ratio, chilled water supply and return water temperature, outdoor temperature, set water outlet temperature, etc. The specific content of the sample input parameters is not limited in this application.

[0099] The sample output parameter is specifically an energy efficiency parameter label, which is an energy efficiency parameter corresponding to the sample input parameter.

[0100] 302. Determine the number of nodes in the hidden layer of the energy efficiency model according to the number of input nodes in the input layer of the energy efficiency model to obtain an initial energy efficiency model, wherein the number of input nodes matches the type of operating data in the air-conditioning system;

[0101] Among them, the selection of the number of hidden layer nodes is the most critical step of the BP neural network, which directly affects the network's ability to map complex problems. If the number of nodes is too small, it will lead to poor network tolerance; if the number of nodes is too large, it will increase the network training time and make it difficult or even impossible for the network to converge.

[0102] In this embodiment, a trial and error method is used to determine the optimal number of nodes included in the hidden layer of the energy efficiency model.

[0103] Among them, the number of input nodes in the input layer of the energy efficiency model can be determined according to the type of operating data in the air-conditioning system. One type of operating data is used as an input parameter and is input into the energy efficiency model through a corresponding input node.

[0104] The operating data may include air conditioning system load, chilled water supply and return water temperature, outdoor temperature, set water outlet temperature, etc.

[0105] The number of nodes in the hidden layer can be determined according to the number of nodes in the input layer. The number of nodes in the hidden layer can be determined using the following formula (1):

[0106] (1)

[0107] in, It represents the number of nodes in the hidden layer, α is the trial-and-error coefficient, and the value of α is generally an integer between 0 and 10, and n is the number of input nodes in the input layer.

[0108] In one possible implementation, the formula (1) The value of can be rounded up. Of course, the rounding method can be preset according to actual conditions and is not limited in this application.

[0109] The number of nodes in the hidden layer can be adjusted by adjusting the value of α.

[0110] In the process of using the trial and error method to determine the optimal number of nodes included in the hidden layer, α can be first taken as a smaller value, such as 1, and the number of nodes included in the hidden layer can be determined using the above formula (1) to generate an initial energy efficiency model.

[0111] 303. Process the sample input parameter based on the initial energy efficiency model to obtain a training output result;

[0112] The sample input parameters are input into the initial energy efficiency model through the input nodes of the input layer, and after being processed by the nodes in the hidden layer, the processing results are output through the output nodes of the output layer to obtain the training output results.

[0113] The training output result is that the initial energy efficiency model processes the sample input parameters to obtain the energy efficiency parameters

[0114] 304. Based on the fact that the gap between the sample output result and the training output result satisfies the set gap condition, the number of nodes in the hidden layer is increased to obtain an adjusted energy efficiency model;

[0115] Among them, the training output result obtained by processing the initial energy efficiency model is compared with the sample output result to determine the gap between the two. If the gap meets the set gap condition, it is determined that the error of the initial energy efficiency model is large and the number of nodes in the hidden layer needs to be increased.

[0116] In a possible implementation, the number of nodes in the hidden layer can be increased by increasing the value of α in the above formula (1).

[0117] As an example, after a training, it is determined that the number of nodes in the hidden layer needs to be increased. In this calculation, the value of α is 2 and the number of hidden layer nodes is 4. Then, using formula (1), the value of α is set to 3 to determine the new number of hidden layer nodes, which is 5.

[0118] In a possible implementation, the value of α may be increased by 1 at a time, or may be increased by 2 or 3 at a time, etc. The present application does not limit the amount of change in the value of α during the trial and error process.

[0119] 305. Continue to train the adjusted energy efficiency model using the sample set until the gap between the sample output result and the training output result does not meet the set gap condition, thereby obtaining a constructed energy efficiency model.

[0120] Among them, using the sample set to continue training the adjusted energy efficiency model includes: inputting the sample input parameters in the sample set into the adjusted energy efficiency model to obtain the training output result, and then comparing the training output result with the sample output result. If the gap between the two still meets the set gap condition, it indicates that the energy efficiency model has not reached the optimal level, and the number of nodes in the hidden layer is continued to be increased, and the sample set is used again to continue training the adjusted energy efficiency model; if the gap between the two does not meet the set gap condition, the training process is stopped, and the adjusted energy efficiency model is used as the constructed energy efficiency model.

[0121] Among them, in the process of constructing the energy efficiency model, the number of nodes contained in the hidden layer of the energy efficiency model is gradually increased through trial and error method to achieve the optimal energy efficiency model structure. The energy efficiency model can be combined with the input operating data in the air-conditioning system for processing to obtain the energy efficiency parameters of each chiller in the air-conditioning system. The energy efficiency parameters are most matched with the load of the air-conditioning system, so as to achieve the purpose of reducing the power consumption of the air-conditioning system.

[0122] In a possible implementation, the BP neural network may adopt an MLP (Multilayer Perceptron) fitting network structure.

[0123] Figure 4 4 is a schematic diagram of the energy efficiency model structure provided by the embodiment of the present application. The energy efficiency model is a three-layer BP neural network, including an input layer 401, a hidden layer 402 and an output layer 403. Among them, the input layer 401 includes 5 input nodes, which respectively input PLR, chilled water flow Q1, cooling water flow Q2, outdoor temperature T0, and water outlet problem T1. The hidden layer includes 3 nodes, and the output layer includes 1 output node, which is used to output the energy efficiency parameter COP.

[0124] Among them, in the trial and error method in this embodiment, the same sample set is used for training during multiple training processes until an energy efficiency model is obtained in which the gap between the sample output result and the training output result meets the set gap condition, and the construction process is completed to obtain a constructed energy efficiency model.

[0125] In the specific application process, as long as the air conditioning system is fixed, the neural network structure remains unchanged. When the air conditioning system adds or reduces equipment, the neural network structure is determined based on the actual situation. If the equipment is the same as the previous system and no new input parameters are generated, the structure of the energy efficiency model remains unchanged.

[0126] In this embodiment, in the process of constructing the energy efficiency model, the number of input nodes is first determined according to the type of operating data in the air-conditioning system, and then the number of nodes in the hidden layer is determined according to the number of input nodes to construct an energy efficiency model. The constructed energy efficiency model is input into the sample input parameters in the sample set to obtain a training output result. If the difference between the training output result and the sample output parameter meets the set difference condition, it means that the difference between the two is large, and the energy efficiency model needs to be adjusted. Specifically, the number of nodes in the hidden layer of the energy efficiency model is increased, and the sample set is used to continue to train the adjusted energy efficiency model until the difference between the training output result output by the energy efficiency model and the sample output result does not meet the set difference condition, the training process is terminated, and the constructed energy efficiency model is obtained. The trial and error method is adopted in the construction process to reduce the training time and improve the training efficiency under the premise of ensuring the network compatibility of the energy efficiency model.

[0127] In a possible implementation, the energy efficiency parameters of each chiller in the air-conditioning system are determined by the energy efficiency model, including:

[0128] Through the energy efficiency model, based on the historical operation data, the energy efficiency parameters of each chiller in the air-conditioning system under each operating condition are determined. When the load in the air-conditioning system is the same, the loads of each chiller are the same / different, and different operating conditions correspond to different loads.

[0129] Among them, the energy efficiency model targets the historical operation data of the air-conditioning system and divides the historical operation data into several sections according to the different operating conditions of the chiller. Each section corresponds to an operating condition, and the loads under different operating conditions are different.

[0130] Among them, this working condition can be divided by the PLR ​​of the chiller. Since the rated capacity of the chiller is fixed, different PLRs correspond to different loads.

[0131] Among them, the energy efficiency model determines the energy efficiency parameters of the chiller under different operating conditions, and then combines the corresponding operating conditions and corresponding energy efficiency parameters of the chiller under various loads of the air-conditioning system to achieve fitting and obtain the energy efficiency parameters of each chiller for different air-conditioning system loads, and determines the corresponding energy efficiency parameters for the subsequent target load for the preset target time period, so as to match its energy efficiency with the system load, maximize the supply and demand balance, and reduce the overall energy consumption of the air-conditioning system.

[0132] In a possible implementation, when using historical operating data to calculate energy efficiency parameters, the following steps are included: cleaning and preprocessing the historical operating data to eliminate expired data, missing data, and out-of-range data; determining the normal operating time of the chiller based on the operating current percentage of the chiller; and calculating the energy efficiency parameters of the chiller.

[0133] Among them, the chiller can operate normally after the air conditioner is turned on for a period of time and before it is turned off for a period of time. The operating data during this period of time is regarded as normal data, and only the data during this period of time is used as historical operating data to participate in determining the energy efficiency parameters.

[0134] Figure 5 1 is a schematic diagram of the relationship between the current percentage of the chiller and time provided in an embodiment of the present application. In the figure, the X-axis represents time, specifically date, and the Y-axis represents the current percentage of the chiller. According to the figure, the current percentage drops sharply from a higher value to 0 in the first few hours of each day, and then suddenly rises to 1200.

[0135] Figure 6 1 is a schematic diagram of the relationship between the normal operating current and time of the chiller provided in the embodiment of the present application. In the figure, the X-axis represents the time, specifically the date, and the Y-axis represents the normal operating current of the chiller. According to the figure, during the operation of the chiller, the change of the current is gradually decreasing and gradually increasing.

[0136] Based on experience Figure 5 and Figure 6 The place where the operating current is 0 is the air conditioner switching point. After a period of time after the air conditioner is turned on, the current gradually increases and reaches stability after about 30 minutes. Before the air conditioner is turned off, the current is unstable, and it is relatively stable 30 minutes before shutdown.

[0137] Therefore, when determining energy efficiency parameters, historical operating data 30 minutes before shutdown or 30 minutes after startup can be extracted as the normal operating range, and the historical operating data in the normal operating range can be used to calculate COP.

[0138] In a possible implementation, the energy efficiency parameters per unit time can be determined for each chiller, and then the energy efficiency parameters under different working conditions can be determined. Figure 7 The process of determining the energy efficiency parameters of each chiller in the air-conditioning system under various operating conditions is described in detail.

[0139] In this embodiment, the energy efficiency parameters corresponding to each chiller in the air-conditioning system under each operating condition are determined through the energy efficiency model based on historical operating data. When the load is the same in the air-conditioning system, the loads of each chiller are the same / different, and different operating conditions correspond to different loads. The energy efficiency model is used to determine the energy efficiency parameters corresponding to the chiller under various operating conditions, providing a basis for subsequently determining the corresponding energy efficiency parameters for the target load in the preset target time period, so that its energy efficiency matches the system load.

[0140] Figure 7The embodiment of the present application provides a flow chart of determining the energy efficiency parameters of each chiller in the air-conditioning system corresponding to each operating condition through an energy efficiency model based on the historical operating data, which may include steps 701 to 702. These steps are described in detail below.

[0141] 701. Determine the energy efficiency parameters of each chiller per unit time based on historical operation data;

[0142] Among them, the historical operation data can be divided into multiple operation data segments according to unit time, and the power consumption and cooling capacity of each chiller per unit time are calculated using each segment of operation data to calculate the energy efficiency parameters of the chiller based on the power consumption and cooling capacity per unit time.

[0143] The normal operating time of the chiller is first determined based on historical operating data, and then the unit time is divided according to the historical operating data of the normal operating time to obtain multiple operating data segments.

[0144] Among them, the energy efficiency parameters of the chiller under a certain working condition can be determined using the following formula (2):

[0145] (2)

[0146] Among them, COP represents the energy efficiency parameter, and C represents the specific heat capacity of water, which is generally taken as 4.2×10 3 (J / kg×℃, joule / kilogram×degree Celsius), F represents the cooling water flow rate, ΔT represents the difference between the cooling water inlet and outlet temperatures, W represents the power consumption per unit time, Represents the cooling capacity of the chiller per unit time.

[0147] Among them, for known historical operating data, the above-mentioned energy efficiency parameter calculation per unit time can be performed only once, and the calculated energy efficiency parameters can be recorded. If there is new historical operating data later, the energy efficiency parameter calculation per unit time can be performed only for the new historical operating data, and the energy efficiency parameters determined by the new historical operating data and the recorded energy efficiency parameters are executed together in step 702.

[0148] 702. Determine the energy efficiency parameters of each chiller in different working conditions through the energy efficiency model for each energy efficiency parameter of each chiller in unit time, wherein any working condition includes at least one unit time.

[0149] Among them, this working condition can be divided by the PLR ​​of the chiller. Since the rated capacity of the chiller is fixed, different PLRs correspond to different loads.

[0150] Among them, different operating conditions of the chiller correspond to different loads. Accordingly, multiple consecutive unit times of the chiller under the same load are divided into the same operating condition, and the energy efficiency parameters under the same operating condition are fitted to obtain the energy efficiency parameters for this operating condition. The energy efficiency parameter obtained by fitting can be an energy efficiency parameter curve or a formula related to a variable. The variable is related to the fitting result, and there is no limitation in this application.

[0151] In a possible implementation, the energy efficiency parameters corresponding to each chiller under each load of the air-conditioning system can be obtained for different air-conditioning system loads, and the energy efficiency parameters of each chiller under the working conditions of the air-conditioning system can be obtained by fitting.

[0152] In this embodiment, the energy efficiency parameters of each chiller per unit time are determined based on the historical operation data; the energy efficiency parameters of each chiller in different working conditions are determined through the energy efficiency model, with respect to the energy efficiency parameters of each chiller per unit time. Any working condition includes at least one unit time, and the historical operation data is divided according to the unit time. After the energy efficiency parameters of each chiller per unit time are determined based on the historical operation data per unit time, the energy efficiency parameters of the chiller in different working conditions are determined based on the energy efficiency parameters of each chiller per unit time through the energy efficiency model, thereby realizing the determination of the energy efficiency parameters of the chiller in each working condition, and subsequently providing a basis for determining the energy efficiency parameters for the target load in the preset target time period.

[0153] Figure 8 It is a flow chart of predicting the target load of the air-conditioning system in a preset target time period through a load prediction model according to the historical operation data provided by an embodiment of the present application, which may include steps 801 to 803, and these steps are described in detail below.

[0154] 801. Divide the historical operation data into at least two categories;

[0155] In this embodiment, the load prediction model uses an execution vector machine to explain the process of predicting the target load.

[0156] Among them, the support vector machine algorithm is based on the VC (Vapnik-Chervonenkis Dimension) dimension theory and structural risk minimization theory, which can better solve problems such as nonlinearity, high dimensionality and local minimum points, and has the advantages of small training samples, strong generalization ability and good practicality. However, the disadvantage of this type of problem is that it is less explainable and does not work well in some scenarios with large periodicity and holiday impacts.

[0157] Therefore, in this embodiment, the historical operation data is divided into multiple categories according to the load, and processing is performed on different categories.

[0158] In a possible implementation, historical load fluctuations and load periodic trends of cooling scenarios are considered, and the load is split into a superposition of factors such as trend terms, periodic terms, and autoregressive terms.

[0159] In a possible implementation, the load of the air-conditioning system may change due to the impact of holidays. Therefore, the holiday factor item may also be used as a supplementary category to improve the prediction accuracy of the load prediction model.

[0160] 802. Processing historical operation data of at least two categories through a load forecasting model to obtain a first load of each category;

[0161] Among them, in the load prediction model, different processing methods are adopted for different categories of historical operation data to predict the corresponding first load.

[0162] In one possible implementation, an autoregressive network (AR net) may be used to construct a time series autocorrelation for an autoregressive item, construct the impact of historical operating data on a forecast period, and predict a first load corresponding to a target preset time period based on the autocorrelation for the historical operating data of the autoregressive item.

[0163] In one possible implementation, Fourier series can be used to model periodic terms, and the load data in the historical operation data can be decomposed by Fourier series. The first K (K is an integer greater than 1) items of data are selected as significant trends, and the first load corresponding to the target preset time period is predicted using the significant trend.

[0164] In one possible implementation, a linear or nonlinear algorithm can be used to model the trend item. The modeling is achieved by analyzing the linear or nonlinear growth trend of the load data in the historical operating data. The modeling results are used to predict the first load corresponding to the target preset time period based on the historical operating data of the trend item.

[0165] In a possible implementation, the first load corresponding to the target preset time period is predicted based on whether the target preset time period is a weekend / holiday or the occurrence time of a specific event.

[0166] Among them, since the impact of holidays is relatively small, they can be used as supplementary items to correct the predicted load.

[0167] 803. By using a load prediction model, the first loads of various categories are combined to obtain a target load of the air-conditioning system in a preset target time period.

[0168] The first loads of various categories may be combined in a pre-set combination manner to obtain the target load of the air-conditioning system in a preset target time period.

[0169] In a possible implementation, the first loads predicted by the trend term, the cycle term, and the autoregressive term are linearly combined to obtain an initial load for a preset target time period. The first load corresponding to the holiday term prediction is further corrected for the initial load to obtain a target load.

[0170] In one possible implementation, when training a load forecasting model, the training samples can be divided into multiple categories according to trend terms, periodic terms, and autoregressive terms, and the load forecasting model can be trained using multiple categories. When the predicted load obtained by the load forecasting model cannot reach consistency with the sample label load, the training samples can be divided into holiday terms to achieve fine-tuning of the predicted load using the holiday terms, so that the prediction result obtained by the load forecasting model is consistent with the sample label load.

[0171] Therefore, when making predictions through the load forecasting model, we can first make predictions based on the trend term, cycle term and autoregressive term in the historical operating data, and then use the holiday term to make corrections.

[0172] Fig. 9 It is a schematic diagram of each category item in the historical operation data provided by the embodiment of the present application. Among them, the horizontal axis in (a) represents the date, and the vertical axis represents the load value, and (a) shows the change trend of the load over time; the horizontal axis in (b) represents the date, and the vertical axis represents the change amount of the load, and (b) shows the trend change amount of the load over time, which reflects the situation of the autoregressive term in the load; the horizontal axis in (c) represents time, specifically the time within a week, and the vertical axis represents the periodic change law based on a week; the horizontal axis in (d) represents time, specifically the time within a month, and the vertical axis represents the periodic change law based on days.

[0173] In order to achieve better generalization and portability between different projects, the load forecasting model combines transfer learning and expert experience, adjusts the model based on a small amount of current load data (target domain), and realizes cross-domain learning of historical data (source domain), thereby providing cross-project adaptive load forecasting capabilities. The load forecasting model can be directly applied to other air-conditioning systems with different structures to reduce the time required for initial model training.

[0174] In this embodiment, the historical operating data is divided into at least two categories. The load prediction model is used to process the historical operating data of at least two categories to obtain the first load of each category. The first load of each category is then combined to obtain the target load of the air-conditioning system in a preset target time period. Different categories are used to divide the historical operating data and predictions are made for different categories. The prediction results are highly interpretable, so that the prediction accuracy of the load prediction model is higher.

[0175] Fig.10 The present application provides a flow chart for determining the first target operating parameters of each chiller in the air-conditioning system based on the target load of the air-conditioning system in a preset target time period and the energy efficiency parameters of each chiller in the air-conditioning system. The flow chart may include steps 1001 to 1002, and these steps are described in detail below.

[0176] 1001. Determine the allocated load of each chiller according to the target load of the air-conditioning system in a preset target time period and the energy efficiency parameters of each chiller through a load scheduling model;

[0177] The target time period includes multiple time points, and a load value is predicted for each time point. Accordingly, the target load includes multiple load values.

[0178] As an example, the preset target time period is 24 hours tomorrow, and the target load may include multiple load values ​​within 24 hours tomorrow, and the multiple load values ​​may form a load curve.

[0179] Among them, for different loads, each chiller needs to adjust its load to match the load of the air-conditioning system. In order to reduce the total energy consumption of the system, each chiller also needs to select energy efficiency parameters that match the air-conditioning system.

[0180] Among them, when modeling the load scheduling model, it is necessary to consider the total output of the air-conditioning unit to achieve supply and demand balance and reasonably allocate the load of each chiller to achieve the goal of minimizing the energy consumption of the entire system.

[0181] In order to balance the supply and demand of cooling capacity, it is necessary to ensure that the demand load at the next moment does not exceed the total load at the current moment, and the goal is to balance the supply and demand and minimize the energy consumption of the entire system.

[0182] The following formula (3) can be used to express the balance between cooling supply and demand:

[0183] (3)

[0184] Where, T represents the load required by the air-conditioning system at the next moment; n represents the number of chillers in the air-conditioning system. Represents the start-up load of chiller i at the current moment.

[0185] The following formula (4) can be used to express the distribution load of each chiller under the goal of meeting the supply and demand balance and minimizing the system energy consumption:

[0186] (4)

[0187] Among them, the represents the total energy consumption of the air-conditioning system, α is the balance factor, which can be used to assign different weights to supply-demand balance and energy saving by changing its size, and β is the dimensional coefficient used to balance the dimensions of the two sub-goals. n means that the air-conditioning system contains n chillers, represents the start-up load of chiller i at the current moment; represents the distributed load of chiller i at the next moment; represents the energy efficiency parameter of chiller i under the load at the current moment, and T represents the load required by the air-conditioning system at the next moment.

[0188] Among them, each chiller needs to comply with safe operation constraints to ensure the safe operation of the air-conditioning system.

[0189] Among them, under the condition that each chiller complies with the safe operation constraints, the balance between cooling supply and demand is met and the system energy consumption is minimal, the distribution load of each chiller in the air-conditioning system is determined.

[0190] Among them, the safe operation constraints include the following five:

[0191] Constraint 1: Each chiller needs to control its load between the upper and lower limits of its rated output;

[0192] The constraint condition 1 is expressed by the following formula (5):

[0193] (5)

[0194] in, represents the start-up load of chiller i at the next moment, represents the upper limit of the rated load of chiller i, represents the optimal operating load factor, .

[0195] Constraint 2: Only one device is started or stopped at a time to ensure the stability of the air conditioning system;

[0196] The constraint condition 2 is expressed by the following formula (6):

[0197] (6)

[0198] in, represents the start-up load of chiller i at the next moment, Represents the real-time operating load of chiller i.

[0199] Constraint 3: For the chillers to be started or shut down, the thresholds for maintaining the start / stop status must be met;

[0200] The constraint condition 3 is expressed by the following formula (7):

[0201] (7)

[0202] in, represents the start-up load of chiller i at the next moment, represents the real-time operating load of chiller i, represents the time threshold for chiller i to remain off, represents the time threshold for chiller i to remain on, Indicates the time threshold.

[0203] Constraint 4: The ramp rate limit does not need to be met in the first t minutes before the chiller is started (the load per unit time is less than the limit value). Except for this time, the load must be within the allowable range.

[0204] The constraint condition 4 is expressed by the following formula (8):

[0205] (8)

[0206] in, represents the start-up load of chiller i at the next moment, represents the upper limit of the rated load of chiller i, represents the operating time of chiller i, Indicates the load threshold.

[0207] Constraint 5: If the chiller is turned on, the distributed load needs to be above 60% of the rated load.

[0208] The constraint condition 5 is expressed by the following formula (9):

[0209] (9)

[0210] in, represents the start-up load of chiller i at the next moment, Indicates the upper limit of the rated load of chiller i.

[0211] The objectives represented by the above formulas (3) to (9) and all parameters in the constraints are transmitted to the load forecasting model in a configured manner, so that in similar scenarios, the load forecasting model can make forecast adjustments according to the on-site working conditions to obtain the distributed load of each chiller at the next moment that meets the above requirements.

[0212] In a possible implementation, a solver is set in the load scheduling model. The solver adopts a multi-population collaborative genetic algorithm with enhanced elite retention, combines objectives and constraints to perform optimization, and finds a suitable distributed load.

[0213] Randomly assign loads to each chiller, and use the assigned loads to determine whether the above objectives and constraints are met. If not, randomly change the load of a chiller, calculate the difference between the load assigned to each chiller and the objectives and constraints, and determine the effect of this change. Based on whether the effect is close to the objective or constraint, choose the way to change the chiller, and continue to change the load of a chiller in the air-conditioning system until the objective and constraint are balanced, and take the determined load of each chiller as the optimal solution.

[0214] 1002. Determine a first target operating parameter of each chiller according to the allocated load of each chiller, wherein the first target operating parameter of each chiller satisfies an operating constraint condition.

[0215] After the allocated load for each chiller is determined, the operating parameters of each chiller can be calculated using the equipment parameters of each chiller.

[0216] Among them, the equipment parameters may include rated voltage and rated current, which can be set according to actual conditions and are not limited in this application.

[0217] Among them, since the load allocated to each chiller meets the aforementioned allocation target and constraint conditions, the first target operating parameter determined by each chiller also meets the operating constraint conditions.

[0218] In this embodiment, through the load scheduling model, the distributed load of each chiller is determined according to the target load of the air-conditioning system in a preset target time period and the energy efficiency parameters of each chiller; based on the distributed load of each chiller, the first target operating parameter of each chiller is determined, and the first target operating parameter of each chiller meets the working constraints, thereby realizing the use of a load scheduling model to determine a distributed load that matches the target load and has lower power consumption for each chiller, thereby reducing the overall energy consumption of the air-conditioning system.

[0219] Fig.11 It is a flowchart of fine-tuning a chiller in an air-conditioning system after reaching a preset target time period provided by an embodiment of the present application, which may include steps 1101 to 1103, and these steps are described in detail below.

[0220] 1101. When the current time reaches the preset target time period, real-time operation data of the air conditioning system is obtained;

[0221] Among them, after the target load of the air-conditioning system in the preset target time period is predicted and the first target operating parameters of each chiller are determined, when the time reaches the preset target time period, each chiller in the air-conditioning system operates using the first target operating parameters.

[0222] The real-time operation data may be data of the same type as the aforementioned historical operation data, and may include various data involved in the operation of each chiller.

[0223] 1102. Obtain the real-time load of the air conditioning system according to the real-time operation data;

[0224] Among them, based on the real-time operation data of each chiller, the load of each chiller can be determined.

[0225] Among them, by accumulating the loads of each chiller, the overall real-time load of the air-conditioning system can be obtained.

[0226] When the current moment reaches the preset target period, the real-time operation data of the air-conditioning system continues to be acquired, so as to fine-tune the real-time operation data and improve the accuracy of the load prediction model.

[0227] Among them, since the refrigeration load is greatly affected by the production cycle, the load forecasting model focuses on the analysis and processing of complex nonlinear and non-stationary load data, determines the overall trend of the time series, and fine-tunes the load forecast value according to real-time load data feedback, so as to achieve more accurate subsequent forecasts and dynamic adjustments based on feedback.

[0228] 1103. Determine the second target operating parameter of each chiller in the air-conditioning system according to the real-time load through a load scheduling model, so as to control the operation of the corresponding chiller based on the second target operating parameter.

[0229] Among them, the difference between the current real-time load and the predicted target load can be determined by point-to-point comparison. If the difference is small, no processing is performed.

[0230] Among them, if the gap between the real-time load at the current moment and the predicted target load is large, the gap is used to adjust the subsequent predicted target load, and the adjusted load is used to allocate load to each chiller in the air-conditioning system, and the second target working parameters are determined based on the load allocated to each chiller.

[0231] In one possible implementation, the ratio of the real-time load at the current moment to the predicted target load is calculated, and the target load corresponding to the time after the current moment is adjusted using the ratio to obtain the adjusted target load. The load scheduling model is used to reallocate the load to each negative chiller based on the adjusted target load.

[0232] Fig.12This is a schematic diagram comparing the target load predicted by the load prediction model provided in the embodiment of the present application with the actual load, wherein the horizontal axis represents time, the vertical axis represents load, the dotted line represents the predicted target load curve, and the solid line represents the actual load. When there is a large gap between the two, the real-time load is used to adjust the subsequent target load so that the two tend to be consistent.

[0233] In this embodiment, when the current moment reaches the preset target time period, the real-time operation data of the air-conditioning system is obtained; based on the real-time operation data, the real-time load of the air-conditioning system is obtained; through the load scheduling model, the second target operating parameters of each chiller in the air-conditioning system are determined according to the real-time load, so as to control the operation of the corresponding water-cooling unit based on the second target operating parameters, thereby utilizing the real-time load in the actual operation of the air-conditioning system to adjust the predicted target load, and utilizing the load scheduling model to reallocate the load for each chiller, so as to reduce the system energy consumption while ensuring the needs of the air-conditioning system.

[0234] A control method for an air-conditioning system provided in an embodiment of the present application is introduced above, and a device for executing the control method for an air-conditioning system will be introduced below.

[0235] Fig.13 1 is a schematic diagram of a control device for an air conditioning system provided in an embodiment of the present application. The air conditioning system includes at least two chillers and several air conditioners. The chillers control the temperature in the building through the air conditioners installed in the building. Fig.13 As shown, the control device 1300 of the air conditioning system includes:

[0236] The acquisition module 1301 is used to obtain the historical operation data of the air conditioning system in a preset historical time period;

[0237] A prediction module 1302 is used to predict the target load of the air conditioning system in a preset target time period according to the historical operation data through a load prediction model, and the preset target time period is later than the current time;

[0238] The first determination module 1303 is used to determine the energy efficiency parameters of each chiller in the air-conditioning system through an energy efficiency model;

[0239] The second determination module 1304 is used to determine the first target operating parameter of each chiller in the air-conditioning system based on the target load of the air-conditioning system in a preset target time period and the energy efficiency parameters of each chiller in the air-conditioning system. The first target operating parameter is used to control the operation of the corresponding chiller when the preset target time period is reached.

[0240] In one possible implementation, the energy efficiency model is constructed as follows:

[0241] Obtain a sample set, the sample set comprising a number of sample pairs consisting of sample input parameters and sample output results;

[0242] According to the number of input nodes of the input layer in the energy efficiency model, the number of nodes of the hidden layer in the energy efficiency model is determined to obtain an initial energy efficiency model, and the number of input nodes matches the type of operating data in the air-conditioning system;

[0243] Processing the sample input parameters based on the initial energy efficiency model to obtain a training output result;

[0244] Based on the fact that the gap between the sample output result and the training output result satisfies the set gap condition, the number of nodes in the hidden layer is increased to obtain an adjusted energy efficiency model;

[0245] The adjusted energy efficiency model is continuously trained using the sample set until the gap between the sample output result and the training output result does not satisfy the set gap condition, thereby obtaining a constructed energy efficiency model.

[0246] In a possible implementation, the first determining module is specifically configured to:

[0247] Through the energy efficiency model, based on the historical operation data, the energy efficiency parameters of each chiller in the air-conditioning system under each operating condition are determined. When the load in the air-conditioning system is the same, the loads of each chiller are the same / different, and different operating conditions correspond to different loads.

[0248] In a possible implementation, the first determining module includes:

[0249] Determine the energy efficiency parameters of each chiller per unit time based on historical operation data;

[0250] Through the energy efficiency model, the energy efficiency parameters of each chiller in different working conditions are determined for each energy efficiency parameter of each chiller in unit time, and any working condition includes at least one unit time.

[0251] In a possible implementation, the prediction module includes:

[0252] A division unit, used for dividing the historical operation data into at least two categories;

[0253] A processing unit, configured to process historical operation data of at least two categories through a load forecasting model to obtain a first load of each category;

[0254] The combining unit is used to combine the first loads of various categories to obtain the target load of the air-conditioning system in a preset target time period through a load prediction model.

[0255] In a possible implementation, the second determining module includes:

[0256] A first determination unit is used to determine the distribution load of each chiller according to the target load of the air-conditioning system in a preset target time period and the energy efficiency parameters of each chiller through a load scheduling model;

[0257] The second determination unit is used to determine the first target operating parameter of each chiller according to the allocated load of each chiller, and the first target operating parameter of each chiller satisfies the operating constraint condition.

[0258] In a possible implementation, the method further includes:

[0259] A real-time acquisition module obtains the real-time operation data of the air conditioning system when the current moment reaches the preset target period;

[0260] An obtaining module, used for obtaining the real-time load of the air conditioning system according to the real-time operation data;

[0261] The third determination module is used to determine the second target operating parameters of each chiller in the air-conditioning system according to the real-time load through a load scheduling model, so as to control the operation of the corresponding chiller based on the second target operating parameters.

[0262] It should be noted that for the functional explanation of each component structure in the control device of an air-conditioning system provided in this embodiment, please refer to the explanation in the aforementioned method embodiment, and it will not be repeated in this embodiment.

[0263] In this embodiment, the air-conditioning system includes at least two chillers and several air conditioners. The chillers control the temperature in the building through the air conditioners installed in the building. The historical operation data of the air-conditioning system is predicted by the load prediction model to predict the target load of the air-conditioning system in the future time period, and the energy efficiency parameters of each chiller in the air-conditioning system are determined by the energy efficiency model. In combination with the target load and the energy efficiency parameters of the chiller, the first target operating parameters of the chiller in the corresponding future time period are determined. When the chiller operates at the first target operating parameters, its energy efficiency is matched with the system load to maximize the supply and demand balance and reduce the overall energy consumption of the air-conditioning system.

[0264] The present application also provides an electronic device in an embodiment. Fig.14 As shown, it shows a schematic diagram of the structure of an electronic device suitable for implementing the embodiment of the present application. The electronic device in the embodiment of the present application may include but is not limited to fixed terminals such as mobile phones, laptops, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Fig.14 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0265] like Fig.14As shown, the electronic device may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 1401, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1402 or a program loaded from a storage device 1408 to a random access memory (RAM) 1403. When the electronic device is powered on, various programs and data required for the operation of the electronic device are also stored in the RAM 1403. The processing device 1401, the ROM 1402, and the RAM 1403 are connected to each other via a bus 1404. An input / output (I / O) interface 1405 is also connected to the bus 1404.

[0266] Typically, the following devices may be connected to the I / O interface 1405: an input device 1406 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 1407 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 1408 including, for example, a memory card, a hard disk, etc.; and a communication device 1409. The communication device 1409 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although Fig.14 An electronic device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0267] An embodiment of the present application also provides a computer program product including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements any of the air-conditioning system control methods provided in the embodiments of the present application.

[0268] A computer-readable storage medium is also provided in an embodiment of the present application. The storage medium carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement any air-conditioning system control method provided in the embodiment of the present application.

[0269] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0270] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0271] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0272] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

Claims

1. A control method for an air conditioning system, the air conditioning system comprising at least two chillers and a plurality of air conditioners, the chillers controlling the temperature in a building through the air conditioners arranged in the building, the control method for the air conditioning system comprising: Obtaining historical operation data of the air conditioning system during a preset historical time period; Using a load prediction model, based on the historical operation data, predicting a target load of the air conditioning system in a preset target time period, wherein the preset target time period is later than the current time; Determine the energy efficiency parameters of each chiller in the air-conditioning system through the energy efficiency model; Based on the target load of the air-conditioning system in a preset target time period and the energy efficiency parameters of each chiller in the air-conditioning system, the first target operating parameter of each chiller in the air-conditioning system is determined, and the first target operating parameter is used to control the operation of the corresponding chiller when the preset target time period is reached.

2. According to the control method of the air conditioning system of claim 1, the energy efficiency model is constructed in the following manner: Obtaining a sample set, wherein the sample set includes a number of sample pairs consisting of sample input parameters and sample output results; According to the number of input nodes of the input layer in the energy efficiency model, the number of nodes of the hidden layer in the energy efficiency model is determined to obtain an initial energy efficiency model, wherein the number of input nodes matches the type of operating data in the air-conditioning system; Processing the sample input parameters based on the initial energy efficiency model to obtain a training output result; Based on the fact that the gap between the sample output result and the training output result satisfies a set gap condition, increasing the number of nodes in the hidden layer to obtain an adjusted energy efficiency model; The adjusted energy efficiency model is continuously trained using the sample set until the gap between the sample output result and the training output result does not satisfy the set gap condition, thereby obtaining a constructed energy efficiency model.

3. The control method of the air conditioning system according to claim 1, wherein determining the energy efficiency parameters of each chiller in the air conditioning system through the energy efficiency model comprises: Through the energy efficiency model, based on the historical operation data, the energy efficiency parameters corresponding to each chiller in the air-conditioning system under each operating condition are determined. When the load in the air-conditioning system is the same, the loads of each chiller are the same / different, and different operating conditions correspond to different loads.

4. The control method of the air conditioning system according to claim 3, wherein the energy efficiency parameters corresponding to each chiller in the air conditioning system under each operating condition are determined according to the historical operation data through the energy efficiency model, including: Determine the energy efficiency parameters of each chiller per unit time based on historical operation data; Through the energy efficiency model, the energy efficiency parameters of each chiller in different working conditions are determined for each energy efficiency parameter of each chiller in unit time, and any working condition includes at least one unit time.

5. The control method of the air conditioning system according to claim 1, wherein the target load of the air conditioning system in a preset target time period is predicted by the load prediction model according to the historical operation data, comprising: Dividing the historical operation data into at least two categories; By using a load forecasting model, historical operation data of at least two categories are processed to obtain a first load of each category; Through the load prediction model, the first loads of various categories are combined to obtain the target load of the air-conditioning system in a preset target time period.

6. The control method of the air conditioning system according to claim 1, wherein the first target operating parameter of each chiller in the air conditioning system is determined based on the target load of the air conditioning system in a preset target time period and the energy efficiency parameter of each chiller in the air conditioning system, comprising: Determine the allocated load of each chiller according to the target load of the air conditioning system in a preset target time period and the energy efficiency parameters of each chiller through a load scheduling model; According to the distributed load of each chiller, a first target operating parameter of each chiller is determined, and the first target operating parameter of each chiller satisfies the operating constraint condition.

7. The control method of the air conditioning system according to claim 1, further comprising: When the current moment reaches the preset target time period, obtaining real-time operation data of the air conditioning system; Obtaining the real-time load of the air conditioning system according to the real-time operation data; The second target operating parameter of each chiller in the air-conditioning system is determined according to the real-time load through the load scheduling model, so as to control the operation of the corresponding chiller based on the second target operating parameter.

8. A control device for an air conditioning system, the air conditioning system comprising at least two chillers and a plurality of air conditioners, the chillers controlling the temperature in a building through the air conditioners arranged in the building, the control device for the air conditioning system comprising: An acquisition module, used to obtain historical operation data of the air conditioning system during a preset historical time period; A prediction module, configured to predict a target load of the air-conditioning system in a preset target time period according to the historical operation data through a load prediction model, wherein the preset target time period is later than the current time; A first determination module is used to determine the energy efficiency parameters of each chiller in the air-conditioning system through an energy efficiency model; The second determination module is used to determine the first target operating parameters of each chiller in the air-conditioning system based on the target load of the air-conditioning system in a preset target time period and the energy efficiency parameters of each chiller in the air-conditioning system. The first target operating parameters are used to control the operation of the corresponding water-cooling unit when the preset target time period is reached.

9. A computer program product, comprising computer-readable instructions, which, when executed on an electronic device, enable the electronic device to implement the control method for an air-conditioning system according to any one of claims 1 to 7.

10. An electronic device, characterized in that: The method comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the electronic device can implement the control method of the air-conditioning system according to any one of claims 1 to 7.

Images