Energy-saving control method of water-cooled air conditioner in manufacturing plant based on load forecasting

Abstract

A method for energy-saving control of water-cooled air conditioners in a manufacturing plant based on load prediction, the method comprising: constructing a heat load mechanism model of a water-cooled air conditioner system, the heat load mechanism model comprising a transient heat conduction model of wall heat transfer, a solar radiation heat model, a heat dissipation model of equipment, and a fresh air latent heat model. Establishing a maximum heat load flow graph model. Establishing a refrigeration mechanism model of a water-cooled air conditioner system, the refrigeration mechanism model comprising a chiller model, a chilled water pipe heat loss model, a cooling tower model, an air handling unit model, and a duct heat loss model. Establishing a maximum refrigeration flow graph model. Training the heat load maximum flow graph model and the refrigeration maximum flow graph model through an attention mechanism to obtain a heat load graph feature matrix and a refrigeration graph feature matrix. Predicting the current operating data of the water-cooled air conditioner system according to the heat load graph feature matrix and the refrigeration graph feature matrix to output an energy-saving control strategy for the water-cooled air conditioner system.

Description

Technical Field

[0001] The present invention belongs to the technical field of water-cooling energy-saving control for manufacturing plants, and more specifically, relates to an energy-saving control method for water-cooling air conditioners in manufacturing plants based on load prediction. Background Art

[0002] At present, for large-scale manufacturing plant water-cooled air conditioning systems, take a large electronic manufacturing plant covering an area of ​​about 50,000 square meters as an example. The factory needs to meet the temperature and humidity requirements of the production process and personnel comfort, and the air conditioning system needs to provide a cooling capacity of about 5,000 refrigeration tons (about 17,600 kilowatts).

[0003] The supporting facilities can be as follows:

[0004] Water-cooled chillers: 4 units, each with a capacity of approximately 1,250 tons (approximately 4,400 kW), used to produce chilled water (generally with an outlet water temperature of 7°C and a return water temperature of 12°C) for supply to air-conditioning terminal equipment to reduce indoor air temperature.

[0005] Cooling towers: 4 units, each of which handles the heat load of a corresponding chiller. It is used to release the heat generated during the operation of the chiller into the atmosphere through heat exchange between water and air, thereby reducing the cooling water temperature (generally 32°C for inlet water and 27°C for outlet water).

[0006] Chilled water pumps: 8 units (4 in use and 4 in reserve, or in parallel operation), each with a flow rate of approximately 1,000 cubic meters per hour and a head of 30 meters, used to circulate chilled water from the chiller to the air-conditioning terminal equipment and back.

[0007] Cooling water pumps: 8 units (4 in use and 4 in reserve, or in parallel operation), each with a flow rate of approximately 1,200 cubic meters per hour and a head of 25 meters, used to circulate cooling water from the chiller to the cooling tower and back.

[0008] Air handling units: 50 units, each with an air handling capacity of approximately 10,000 cubic meters per hour, used to filter, cool, heat, humidify or dehumidify the air to meet indoor air quality and temperature and humidity requirements.

[0009] Fan coil units: 300 units, each with a cooling capacity of approximately 5 kW and an air volume of approximately 1,000 cubic meters per hour, used for temperature control in local areas or office areas and to assist AHU.

[0010] Other items such as pipes, valves, water tanks, and generators are determined by the main components above and will not be elaborated on here.

[0011] However, in large manufacturing plants, there are various sources of heat load, that is, different production lines, equipment and process flows will generate different amounts of heat. For example, the heat loads in electronic assembly areas, mechanical processing areas and storage areas are significantly different. In addition, dynamic changes in heat load further increase the complexity of load calculations. Production plans and personnel flows cause heat loads to change over time, requiring dynamic load analysis. The equipment inside manufacturing plants is also relatively complex, and the factors to be considered in terms of refrigeration energy saving are also relatively complex. Therefore, existing energy-saving optimization methods are difficult to meet the needs of manufacturing plants for complex load calculations, and their applicability needs to be further improved. Summary of the invention

[0012] In order to solve the deficiencies in the prior art, the purpose of the present invention is to solve the above-mentioned defects and further propose an energy-saving control method for water-cooled air conditioners in manufacturing plants based on load forecasting.

[0013] The present invention adopts the following technical solution.

[0014] The first aspect of the present invention discloses a method for energy-saving control of water-cooled air conditioners in a manufacturing plant based on load forecasting, the method comprising:

[0015] Constructing a heat load mechanism model of a water-cooled air conditioning system, the heat load mechanism model includes a transient heat conduction model of wall heat transfer, a solar radiation heat model, a heat dissipation model of equipment, and a fresh air latent heat model;

[0016] Based on the heat load mechanism model, a heat load maximum flow diagram model is established;

[0017] Establishing a refrigeration mechanism model of the water-cooled air-conditioning system, wherein the refrigeration mechanism model includes a chiller model, a chilled water pipeline heat loss model, a cooling tower model, an air handling unit model, and an air duct heat loss model;

[0018] Based on the refrigeration mechanism model, a refrigeration maximum flow diagram model is established;

[0019] The heat load maximum flow graph model and the cooling maximum flow graph model are trained by an attention mechanism to obtain a heat load graph feature matrix and a cooling graph feature matrix;

[0020] According to the heat load map characteristic matrix and the cooling map characteristic matrix, the current operation data of the water-cooled air-conditioning system is predicted to output an energy-saving control strategy for the water-cooled air-conditioning system.

[0021] Furthermore, the construction of the heat load mechanism model of the water-cooled air-conditioning system includes:

[0022] A transient thermal conduction model of the wall heat transfer is constructed, and the expression of the transient thermal conduction model of the wall heat transfer is:

[0023] ,

[0024] ,

[0025] ;

[0026] In the formula, Indicates that the wall is at time Lower position The temperature at is the thermal conductivity of the wall, is the wall density, is the specific heat capacity of the wall, and are the convective heat transfer coefficients of the inner and outer surfaces, and are the indoor and outdoor air temperatures, for The sun's thermal radiation at all times.

[0027] Furthermore, the construction of the heat load mechanism model of the water-cooled air-conditioning system also includes:

[0028] The solar thermal radiation model is constructed, and the expression of the solar thermal radiation model is:

[0029] ;

[0030] In the formula, for The sun's thermal radiation at all times, is the window area, is the heat coefficient of the sun, Indicates unit time The intensity of solar radiation;

[0031] A heat dissipation thermal model of the device is constructed, and the expression of the heat dissipation thermal model of the device is:

[0032] ;

[0033] In the formula, for The thermal radiation of the equipment at all times, and Respectively Devices in Power consumption and efficiency at all times;

[0034] The fresh air latent heat model is constructed, and the expression of the fresh air latent heat model is:

[0035] ;

[0036] In the formula, is the sensible heat load, is the fresh air mass flow rate, is the constant pressure specific heat capacity of air, is the outdoor air temperature, Set the temperature for the room.

[0037] Furthermore, the establishment of the refrigeration mechanism model of the water-cooled air-conditioning system includes:

[0038] The chiller model is established, and the expression of the chiller model is:

[0039] ;

[0040] In the formula, is the cooling capacity of the chiller, is the chilled water mass flow rate, is the specific heat capacity of water, is the chilled water return temperature, The supply temperature of chilled water;

[0041] The heat loss model of the chilled water pipeline is established. The expression of the heat loss model of the chilled water pipeline is:

[0042] ;

[0043] In the formula, is the pipeline heat loss, is the thermal conductivity of the insulation material, is the average temperature of chilled water, is the ambient temperature, is the pipe length, is the inner radius of the pipe, is the outer radius of the insulation layer.

[0044] Furthermore, the establishment of the refrigeration mechanism model of the water-cooled air-conditioning system also includes:

[0045] The cooling tower model is established, and the expression of the cooling tower model is:

[0046] ;

[0047] In the formula, is the heat dissipation of the cooling tower, is the cooling water mass flow rate, is the specific heat capacity of water, is the cooling water return temperature, The supply temperature of cooling water;

[0048] The air handling unit model is established, and the expression of the air handling unit model is:

[0049] ;

[0050] In the formula, is the cooling capacity of the air handling unit, is the air mass flow rate, is the specific heat capacity of air, is the return air temperature, is the supply air temperature;

[0051] The air duct heat loss model is established, and the expression of the air duct heat loss model is:

[0052] ;

[0053] In the formula, is the heat loss in the air duct, is the thermal conductivity of the duct insulation material, is the average cooling air temperature, is the ambient temperature, is the air duct length, is the inner radius of the air duct, It is the outer radius of the air duct insulation layer.

[0054] Furthermore, the heat load maximum flow graph model and the cooling maximum flow graph model are trained by the attention mechanism to obtain a heat load graph feature matrix and a cooling graph feature matrix, including:

[0055] Extracting a first eigenvector of a first vertex in the heat load map feature matrix after being mapped in the heat load maximum flow map model, and extracting a second eigenvector of a second vertex in the cooling map feature matrix after being mapped in the cooling maximum flow map model;

[0056] Calculate the neighborhood feature vectors of the first vertex and the second vertex with respect to any child node respectively, and calculate the attention scores of the first vertex and the second vertex with respect to any energy-saving control strategy according to the neighborhood feature vectors;

[0057] The first eigenvector is obtained based on a first weight learning matrix preset for the first vertex and a first parameter to be learned, and the second eigenvector is obtained based on a second weight learning matrix preset for the second vertex and a second parameter to be learned.

[0058] Furthermore, the current operation data of the water-cooled air-conditioning system is predicted according to the heat load map characteristic matrix and the cooling map characteristic matrix to output an energy-saving control strategy for the water-cooled air-conditioning system, including:

[0059] Calculating the strength of association between any two sub-nodes based on the attention score to obtain weighted feature vectors of the sub-nodes under the first vertex and the second vertex, and aggregating the weighted feature vectors to obtain the heat load map feature matrix and the cooling map feature matrix;

[0060] Acquire the current operation data of the water-cooled air-conditioning system, and predict the current operation data by calling the softmax function according to the heat load map feature matrix and the cooling map feature matrix to obtain the energy-saving control strategy;

[0061] The energy-saving optimization strategy is the predicted operating data after energy-saving optimization based on the current operating data, and at least includes all physical quantity data involved after energy-saving optimization.

[0062] The second aspect of the present invention discloses a water-cooled air conditioning energy-saving control device for a manufacturing plant based on load forecasting, the device comprising:

[0063] A heat load mechanism model building module is used to build a heat load mechanism model for a water-cooled air conditioning system. The heat load mechanism model includes a transient heat conduction model for wall heat transfer, a solar radiation heat model, a heat dissipation model for equipment, and a fresh air latent heat model.

[0064] A heat load maximum flow diagram model establishment module, used to establish a heat load maximum flow diagram model based on the heat load mechanism model;

[0065] A refrigeration mechanism model building module is used to build a refrigeration mechanism model of the water-cooled air-conditioning system, wherein the refrigeration mechanism model includes a chiller model, a chilled water pipeline heat loss model, a cooling tower model, an air handling unit model, and an air duct heat loss model;

[0066] A refrigeration maximum flow graph model establishment module, used to establish a refrigeration maximum flow graph model based on the refrigeration mechanism model;

[0067] An attention mechanism training module, used for training the heat load maximum flow graph model and the cooling maximum flow graph model through an attention mechanism to obtain a heat load graph feature matrix and a cooling graph feature matrix;

[0068] The energy-saving strategy prediction module is used to predict the current operation data of the water-cooled air-conditioning system according to the heat load map characteristic matrix and the cooling map characteristic matrix to output the energy-saving control strategy of the water-cooled air-conditioning system.

[0069] The third aspect of the present invention discloses a terminal, including a processor and a storage medium; the characteristics are:

[0070] The storage medium is used to store instructions;

[0071] The processor is used to operate according to the instructions to execute the steps of the method described in the first aspect.

[0072] The fourth aspect of the present invention discloses a computer-readable storage medium having a computer program stored thereon, characterized in that when the program is executed by a processor, the steps of the method described in the first aspect are implemented.

[0073] The beneficial effects of the present invention are that, compared with the prior art, the present invention has the following advantages:

[0074] By constructing a heat load mechanism model of a water-cooled air-conditioning system, the heat load mechanism model includes a transient heat conduction model of wall heat transfer, a solar radiation heat model, a heat dissipation model of the equipment, and a fresh air latent heat model, and then based on the heat load mechanism model, a heat load maximum flow graph model is established. At the same time, a refrigeration mechanism model of a water-cooled air-conditioning system is established, which includes a chiller model, a chilled water pipeline heat loss model, a cooling tower model, an air handling unit model, and a duct heat loss model. Then, based on the refrigeration mechanism model, a refrigeration maximum flow graph model is established. Subsequently, the heat load maximum flow graph model and the refrigeration maximum flow graph model are trained through the attention mechanism to obtain the heat load graph feature matrix and the refrigeration graph feature matrix. Finally, according to the heat load graph feature matrix and the refrigeration graph feature matrix, the current operating data of the water-cooled air-conditioning system is predicted to output the energy-saving control strategy of the water-cooled air-conditioning system. This method uses multiple models constructed in the heat load mechanism model and refrigeration mechanism model of the water-cooled air-conditioning system to perform intelligent refrigeration energy-saving optimization control on complex equipment in manufacturing projects. It can meet the needs of manufacturing plants with large areas and complex internal equipment for refrigeration energy-saving optimization and has strong applicability. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 It is a flow chart of a method for energy-saving control of water-cooled air conditioners in a manufacturing plant based on load forecasting provided by the present invention. DETAILED DESCRIPTION

[0076] The present application is further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present application.

[0077] like Figure 1 As shown, in one embodiment, the energy-saving control method of water-cooled air conditioner in a manufacturing plant based on load forecasting provided by the present invention comprises the following steps:

[0078] Step S110, constructing a heat load mechanism model of the water-cooled air conditioning system, the heat load mechanism model including a transient heat conduction model of wall heat transfer, a solar radiation heat model, a heat dissipation model of equipment, and a fresh air latent heat model.

[0079] In some embodiments, the energy-saving control method for water-cooled air conditioners in manufacturing plants based on load forecasting provided by the present invention constructs a heat load mechanism model of the water-cooled air conditioner system, specifically comprising the following steps:

[0080] Step S111, constructing a transient thermal conduction model of wall heat transfer, the expression of the transient thermal conduction model of wall heat transfer is:

[0081] ,

[0082] ,

[0083] .

[0084] In the formula, Indicates that the wall is at time Lower position The temperature at is the thermal conductivity of the wall, is the wall density, is the specific heat capacity of the wall, and are the convective heat transfer coefficients of the inner and outer surfaces, and are the indoor and outdoor air temperatures, for The sun's thermal radiation at all times.

[0085] In some embodiments, the energy-saving control method for water-cooled air conditioners in manufacturing plants based on load forecasting provided by the present invention constructs a heat load mechanism model of the water-cooled air conditioner system, and specifically further includes the following steps:

[0086] Step S112, constructing a solar thermal radiation model, the solar thermal radiation model is expressed as:

[0087] .

[0088] In the formula, for The sun's thermal radiation at all times, is the window area, is the heat coefficient of the sun, Indicates unit time of solar radiation intensity.

[0089] Step S113, constructing a heat dissipation model of the device. The heat dissipation model of the device is expressed as:

[0090] .

[0091] In the formula, for The thermal radiation of the equipment at all times, and Respectively Devices in Power consumption and efficiency at all times.

[0092] Step S114, constructing a fresh air latent heat model, the expression of the fresh air latent heat model is:

[0093] .

[0094] In the formula, is the sensible heat load, is the fresh air mass flow rate, is the constant pressure specific heat capacity of air, is the outdoor air temperature, Set the temperature for the room.

[0095] Step S120, establishing a heat load maximum flow diagram model based on the heat load mechanism model.

[0096] Step S130, establishing a refrigeration mechanism model of the water-cooled air-conditioning system, wherein the refrigeration mechanism model includes a chiller model, a chilled water pipeline heat loss model, a cooling tower model, an air handling unit model, and an air duct heat loss model.

[0097] In some embodiments, the energy-saving control method for water-cooled air conditioners in manufacturing plants based on load forecasting provided by the present invention establishes a refrigeration mechanism model of a water-cooled air conditioner system, specifically comprising the following steps:

[0098] Step S131, establish a chiller model, the expression of the chiller model is:

[0099] .

[0100] In the formula, is the cooling capacity of the chiller, is the chilled water mass flow rate, is the specific heat capacity of water, is the chilled water return temperature, The chilled water supply temperature.

[0101] Step S132, establishing a chilled water pipeline heat loss model, the expression of the chilled water pipeline heat loss model is:

[0102] .

[0103] In the formula, is the pipeline heat loss, is the thermal conductivity of the insulation material, is the average temperature of chilled water, is the ambient temperature, is the pipe length, is the inner radius of the pipe, is the outer radius of the insulation layer.

[0104] In some embodiments, the water-cooled air conditioning energy-saving control method for a manufacturing plant based on load forecasting provided by the present invention establishes a refrigeration mechanism model of the water-cooled air conditioning system, and specifically further includes the following steps:

[0105] Step S133, establish a cooling tower model, the expression of the cooling tower model is:

[0106] .

[0107] In the formula, is the heat dissipation of the cooling tower, is the cooling water mass flow rate, is the specific heat capacity of water, is the cooling water return temperature, It is the cooling water supply temperature.

[0108] Step S134, establishing an air handling unit model, the expression of the air handling unit model is:

[0109] .

[0110] In the formula, is the cooling capacity of the air handling unit, is the air mass flow rate, is the specific heat capacity of air, is the return air temperature, is the supply air temperature.

[0111] Step S135, establish a duct heat loss model, the expression of the duct heat loss model is:

[0112] .

[0113] In the formula, is the heat loss in the air duct, is the thermal conductivity of the duct insulation material, is the average cooling air temperature, is the ambient temperature, is the air duct length, is the inner radius of the air duct, It is the outer radius of the air duct insulation layer.

[0114] Step S140, establishing a refrigeration maximum flow diagram model based on the refrigeration mechanism model.

[0115] Step S150, training the heat load maximum flow graph model and the cooling maximum flow graph model through an attention mechanism to obtain a heat load graph feature matrix and a cooling graph feature matrix.

[0116] In some embodiments, the energy-saving control method for water-cooled air conditioners in manufacturing plants based on load forecasting provided by the present invention trains the maximum heat load flow graph model and the maximum cooling flow graph model through an attention mechanism to obtain a heat load graph feature matrix and a cooling graph feature matrix, specifically comprising the following steps:

[0117] Step S151, extracting a first eigenvector of a first vertex in a heat load map feature matrix mapped in a heat load maximum flow map model, and extracting a second eigenvector of a second vertex in a cooling map feature matrix mapped in a cooling maximum flow map model.

[0118] Step S152, respectively calculating the neighborhood feature vectors of the first vertex and the second vertex with respect to any child node, and calculating the attention scores of the first vertex and the second vertex with respect to any energy-saving control strategy according to the neighborhood feature vectors.

[0119] The first eigenvector is obtained based on a first weight learning matrix preset for the first vertex and a first parameter to be learned, and the second eigenvector is obtained based on a second weight learning matrix preset for the second vertex and a second parameter to be learned.

[0120] Step S160, predicting the current operation data of the water-cooled air-conditioning system according to the heat load map characteristic matrix and the cooling map characteristic matrix, so as to output an energy-saving control strategy for the water-cooled air-conditioning system.

[0121] In some embodiments, the energy-saving control method for water-cooled air conditioners in manufacturing plants based on load forecasting provided by the present invention predicts the current operating data of the water-cooled air-conditioning system according to the heat load map feature matrix and the refrigeration map feature matrix to output an energy-saving control strategy for the water-cooled air-conditioning system, specifically comprising the following steps:

[0122] Step S161, calculate the correlation strength between any two sub-nodes based on the attention score to obtain the weighted feature vectors of the sub-nodes under the first vertex and the second vertex, and aggregate the weighted feature vectors to obtain the heat load map feature matrix and the cooling map feature matrix.

[0123] Step S162, obtaining the current operation data of the water-cooled air-conditioning system, and calling the softmax function to predict the current operation data according to the heat load map feature matrix and the cooling map feature matrix to obtain the energy-saving control strategy.

[0124] The energy-saving optimization strategy is the predicted operating data after energy-saving optimization based on the current operating data, and at least includes all physical quantity data involved after energy-saving optimization.

[0125] In a specific embodiment, the present invention provides a method for energy-saving control of water-cooled air conditioners in a manufacturing plant based on load forecasting, comprising steps 1 to 6:

[0126] Step 1: Establish a heat load mechanism model for the water-cooled air-conditioning system, including: transient heat conduction model of wall heat transfer, solar radiation heat model, equipment heat dissipation model and fresh air latent heat model.

[0127] Specifically, the transient thermal conduction model of wall heat transfer is expressed as:

[0128] ,

[0129] ,

[0130] .

[0131] In the formula, Indicates that the wall is at time Lower position The temperature at is the thermal conductivity of the wall, is the wall density, is the specific heat capacity of the wall, and are the convective heat transfer coefficients of the inner and outer surfaces, and are the indoor and outdoor air temperatures, for The sun's thermal radiation at all times.

[0132] The expression of the solar thermal radiation model is:

[0133] .

[0134] In the formula, for The sun's thermal radiation at all times, is the window area, is the heat coefficient of the sun, Indicates unit time of solar radiation intensity.

[0135] The expression of the heat dissipation thermal model of the equipment is:

[0136] .

[0137] In the formula, for The thermal radiation of the equipment at all times, and Respectively Devices in Power consumption and efficiency at all times.

[0138] The expression of fresh air latent heat model is:

[0139] .

[0140] In the formula, is the sensible heat load, is the fresh air mass flow rate, is the constant pressure specific heat capacity of air, is the outdoor air temperature, Set the temperature for the room.

[0141] Step 2: Based on the heat load mechanism model, a heat load maximum flow diagram model is established.

[0142] Step 3: Establish a refrigeration mechanism model of the water-cooled air-conditioning system, including: chiller model, chilled water pipeline heat loss model, cooling tower model, air handling unit (AHU) model and duct heat loss model.

[0143] Specifically, the expression of the chiller model is:

[0144] .

[0145] In the formula, is the cooling capacity of the chiller, is the chilled water mass flow rate, is the specific heat capacity of water, is the chilled water return temperature, The chilled water supply temperature.

[0146] The expression of the heat loss model of the chilled water pipeline is:

[0147] .

[0148] In the formula, is the pipeline heat loss, is the thermal conductivity of the insulation material, is the average temperature of chilled water, is the ambient temperature, is the pipe length, is the inner radius of the pipe, is the outer radius of the insulation layer.

[0149] When the temperature of the chilled water increases after passing through the pipes:

[0150] .

[0151] The expression of the cooling tower model is:

[0152] .

[0153] In the formula, is the heat dissipation of the cooling tower, is the cooling water mass flow rate, is the specific heat capacity of water, is the cooling water return temperature, It is the cooling water supply temperature.

[0154] The expression of the air handling unit model is:

[0155] .

[0156] In the formula, is the cooling capacity of the air handling unit, is the air mass flow rate, is the specific heat capacity of air, is the return air temperature, is the supply air temperature.

[0157] When the temperature of the chilled water in the air handling unit increases:

[0158] .

[0159] The expression of the duct heat loss model is:

[0160] .

[0161] In the formula, is the heat loss in the air duct, is the thermal conductivity of the duct insulation material, is the average cooling air temperature, is the ambient temperature, is the air duct length, is the inner radius of the air duct, It is the outer radius of the air duct insulation layer.

[0162] When the temperature of the cooling air increases after passing through the air duct:

[0163] .

[0164] Step 4: Establish a refrigeration maximum flow diagram model based on the refrigeration mechanism model.

[0165] Step 5: Train the heat load maximum flow graph model and the cooling maximum flow graph model through the attention mechanism to obtain the heat load graph feature matrix and the cooling graph feature matrix.

[0166] Specifically, taking the heat load map feature matrix as an example, the feature vector of the vertex mapped in the heat load maximum flow map model is extracted, where the vertex The mapped feature vector The expression is:

[0167] .

[0168] In the formula, is the preset weight learning matrix, is the parameter to be learned, and its initial value is set randomly.

[0169] Then, calculate the vertex The neighborhood feature vector of any child node (i.e., part of the energy-saving control strategy) under this vertex , the expression is:

[0170] .

[0171] In the formula, is the set of vertices in the neighborhood of the vertex in the flow graph model with the maximum heat load, is the adjacency matrix of the vertex in the maximum heat load flow graph model, represents the feature representation after vertex mapping, where , is the identity matrix.

[0172] According to the attention score between the vertex and any child node, the correlation strength between any two nodes is calculated, and based on the correlation strength, the weighted feature vector between nodes is calculated. Then all weighted feature vectors are aggregated to obtain the final heat load map feature matrix.

[0173] Step 6: predict the current operation data of the water-cooled air-conditioning system according to the heat load map feature matrix and the cooling map feature matrix, and output the energy-saving control strategy of the water-cooled air-conditioning system.

[0174] Specifically, firstly, the correlation strength between any two sub-nodes is calculated based on the attention score to obtain the weighted feature vector of the sub-node under the corresponding vertex, and the weighted feature vector is aggregated to obtain the heat load map feature matrix and the cooling map feature matrix. The current operation data of the water-cooled air-conditioning system is obtained, and the softmax function is called to predict the current operation data according to the heat load map feature matrix and the cooling map feature matrix to obtain the energy-saving control strategy.

[0175] Among them, the energy-saving optimization strategy is the operating data corresponding to the water-cooled air-conditioning system after energy-saving optimization, including the temperature, flow and other physical quantity data involved in each model, and also includes the specific optimization strategy for each load in the sub-node, such as adjusting the temperature of chilled water, optimizing the cooling water flow, adjusting the air temperature delivered by the air handling unit, etc.

[0176] The following describes the energy-saving control device for water-cooled air conditioners in manufacturing plants based on load forecasting provided by the present invention. The energy-saving control device for water-cooled air conditioners in manufacturing plants based on load forecasting described below and the energy-saving control method for water-cooled air conditioners in manufacturing plants based on load forecasting described above can be referenced to each other.

[0177] In one embodiment, a water-cooled air conditioning energy-saving control device for a manufacturing plant based on load prediction includes a heat load mechanism model building module, a heat load maximum flow graph model building module, a refrigeration mechanism model building module, a refrigeration maximum flow graph model building module, an attention mechanism training module, and an energy-saving strategy prediction module.

[0178] The heat load mechanism model building module is used to construct the heat load mechanism model of the water-cooled air-conditioning system. The heat load mechanism model includes the transient thermal conduction model of wall heat transfer, the solar radiation heat model, the heat dissipation model of the equipment, and the fresh air latent heat model.

[0179] The heat load maximum flow diagram model establishment module is used to establish a heat load maximum flow diagram model based on the heat load mechanism model.

[0180] The refrigeration mechanism model building module is used to build the refrigeration mechanism model of the water-cooled air-conditioning system. The refrigeration mechanism model includes a chiller model, a chilled water pipeline heat loss model, a cooling tower model, an air handling unit model, and an air duct heat loss model.

[0181] The refrigeration maximum flow graph model establishment module is used to establish a refrigeration maximum flow graph model based on a refrigeration mechanism model.

[0182] The attention mechanism training module is used to train the heat load maximum flow graph model and the cooling maximum flow graph model through the attention mechanism to obtain the heat load graph feature matrix and the cooling graph feature matrix.

[0183] The energy-saving strategy prediction module is used to predict the current operation data of the water-cooled air-conditioning system according to the heat load map characteristic matrix and the cooling map characteristic matrix, so as to output the energy-saving control strategy of the water-cooled air-conditioning system.

[0184] In this embodiment, the heat load mechanism model building module of the water-cooled air-conditioning energy-saving control device for a manufacturing plant based on load forecasting provided by the present invention is specifically used for:

[0185] A transient thermal conduction model of wall heat transfer is constructed. The expression of the transient thermal conduction model of wall heat transfer is:

[0186] ,

[0187] ,

[0188] .

[0189] In the formula, Indicates that the wall is at time Lower position The temperature at is the thermal conductivity of the wall, is the wall density, is the specific heat capacity of the wall, and are the convective heat transfer coefficients of the inner and outer surfaces, and are the indoor and outdoor air temperatures, for The sun's thermal radiation at all times.

[0190] In this embodiment, the water-cooled air-conditioning energy-saving control device for a manufacturing plant based on load forecasting provided by the present invention, the heat load mechanism model establishment module is specifically used for:

[0191] Construct the solar thermal radiation model. The expression of the solar thermal radiation model is:

[0192] .

[0193] In the formula, for The sun's thermal radiation at all times, is the window area, is the heat coefficient of the sun, Indicates unit time of solar radiation intensity.

[0194] Construct a heat dissipation model for the equipment. The heat dissipation model for the equipment is expressed as:

[0195] .

[0196] In the formula, for The thermal radiation of the equipment at all times, and Respectively Devices in Power consumption and efficiency at all times.

[0197] Construct a fresh air latent heat model, the expression of the fresh air latent heat model is:

[0198] .

[0199] In the formula, is the sensible heat load, is the fresh air mass flow rate, is the constant pressure specific heat capacity of air, is the outdoor air temperature, Set the temperature for the room.

[0200] In this embodiment, the water-cooled air-conditioning energy-saving control device for a manufacturing plant based on load forecasting provided by the present invention, the refrigeration mechanism model establishment module is specifically used for:

[0201] Establish a chiller model. The expression of the chiller model is:

[0202] .

[0203] In the formula, is the cooling capacity of the chiller, is the chilled water mass flow rate, is the specific heat capacity of water, is the chilled water return temperature, The chilled water supply temperature.

[0204] A chilled water pipeline heat loss model is established. The expression of the chilled water pipeline heat loss model is:

[0205] .

[0206] In the formula, is the pipeline heat loss, is the thermal conductivity of the insulation material, is the average temperature of chilled water, is the ambient temperature, is the pipe length, is the inner radius of the pipe, is the outer radius of the insulation layer.

[0207] In this embodiment, the water-cooled air-conditioning energy-saving control device for a manufacturing plant based on load forecasting provided by the present invention, the refrigeration mechanism model establishment module is specifically used for:

[0208] Establish a cooling tower model. The expression of the cooling tower model is:

[0209] .

[0210] In the formula, is the heat dissipation of the cooling tower, is the cooling water mass flow rate, is the specific heat capacity of water, is the cooling water return temperature, It is the cooling water supply temperature.

[0211] Establish an air handling unit model. The expression of the air handling unit model is:

[0212] .

[0213] In the formula, is the cooling capacity of the air handling unit, is the air mass flow rate, is the specific heat capacity of air, is the return air temperature, is the supply air temperature.

[0214] The duct heat loss model is established, and the expression of the duct heat loss model is:

[0215] .

[0216] In the formula, is the heat loss in the air duct, is the thermal conductivity of the duct insulation material, is the average cooling air temperature, is the ambient temperature, is the air duct length, is the inner radius of the air duct, It is the outer radius of the air duct insulation layer.

[0217] In this embodiment, the water-cooled air-conditioning energy-saving control device for a manufacturing plant based on load forecasting provided by the present invention, the attention mechanism training module is specifically used for:

[0218] A first eigenvector is extracted after a first vertex in a heat load map feature matrix is ​​mapped in a heat load maximum flow map model, and a second eigenvector is extracted after a second vertex in a cooling map feature matrix is ​​mapped in a cooling maximum flow map model.

[0219] The neighborhood feature vectors of the first vertex and the second vertex with respect to any child node are calculated respectively, and the attention scores of the first vertex and the second vertex with respect to any energy-saving control strategy are calculated according to the neighborhood feature vectors.

[0220] The first eigenvector is obtained based on a first weight learning matrix preset for the first vertex and a first parameter to be learned, and the second eigenvector is obtained based on a second weight learning matrix preset for the second vertex and a second parameter to be learned.

[0221] In this embodiment, the energy-saving strategy prediction module of the water-cooling air-conditioning energy-saving control device for a manufacturing plant based on load prediction provided by the present invention is specifically used for:

[0222] The correlation strength between any two sub-nodes is calculated based on the attention score to obtain the weighted feature vectors of the sub-nodes under the first vertex and the second vertex, and the weighted feature vectors are aggregated to obtain the heat load map feature matrix and the cooling map feature matrix.

[0223] The current operating data of the water-cooled air-conditioning system is obtained, and the softmax function is called to predict the current operating data according to the heat load map feature matrix and the cooling map feature matrix to obtain the energy-saving control strategy.

[0224] The energy-saving optimization strategy is the predicted operating data after energy-saving optimization based on the current operating data, and at least includes all physical quantity data involved after energy-saving optimization.

[0225] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0226] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the above. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the above. The computer-readable storage medium used herein is not to be interpreted as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0227] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0228] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect through the Internet). In some embodiments, by using the state information of the computer-readable program instructions to personalize an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), the electronic circuit may execute the computer-readable program instructions, thereby implementing various aspects of the present disclosure.

[0229] Various aspects of the present disclosure are described herein with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of boxes in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0230] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0231] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0232] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of special hardware and computer instructions.

[0233] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for energy-saving control of water-cooled air conditioners in a manufacturing plant based on load forecasting, characterized in that: The method comprises: Constructing a heat load mechanism model of a water-cooled air conditioning system, the heat load mechanism model includes a transient heat conduction model of wall heat transfer, a solar radiation heat model, a heat dissipation model of equipment, and a fresh air latent heat model; Based on the heat load mechanism model, a heat load maximum flow diagram model is established; Establishing a refrigeration mechanism model of the water-cooled air-conditioning system, wherein the refrigeration mechanism model includes a chiller model, a chilled water pipeline heat loss model, a cooling tower model, an air handling unit model, and an air duct heat loss model; Based on the refrigeration mechanism model, a refrigeration maximum flow diagram model is established; The heat load maximum flow graph model and the cooling maximum flow graph model are trained by an attention mechanism to obtain a heat load graph feature matrix and a cooling graph feature matrix; According to the heat load map characteristic matrix and the cooling map characteristic matrix, current operation data of the water-cooled air-conditioning system is predicted to output an energy-saving control strategy for the water-cooled air-conditioning system; Wherein, the construction of the heat load mechanism model of the water-cooled air-conditioning system includes: A transient thermal conduction model of the wall heat transfer is constructed, and the expression of the transient thermal conduction model of the wall heat transfer is: ， In the formula, Indicates that the wall is at time Lower position The temperature at is the thermal conductivity of the wall, is the wall density, is the specific heat capacity of the wall, and are the convective heat transfer coefficients of the inner and outer surfaces, and are the indoor and outdoor air temperatures, for The sun's thermal radiation at all times; The construction of the heat load mechanism model of the water-cooled air conditioning system also includes: The solar thermal radiation model is constructed, and the expression of the solar thermal radiation model is: In the formula, for The sun's thermal radiation at all times, is the window area, is the heat coefficient of the sun, Indicates unit time The intensity of solar radiation; A heat dissipation thermal model of the device is constructed, and the expression of the heat dissipation thermal model of the device is: In the formula, for The thermal radiation of the equipment at all times, and Respectively Devices in Power consumption and efficiency at all times; The fresh air latent heat model is constructed, and the expression of the fresh air latent heat model is: In the formula, is the sensible heat load, is the fresh air mass flow rate, is the constant pressure specific heat capacity of air, is the outdoor air temperature, Set the temperature for the room; The step of establishing the refrigeration mechanism model of the water-cooled air-conditioning system comprises: The chiller model is established, and the expression of the chiller model is: In the formula, is the cooling capacity of the chiller, is the chilled water mass flow rate, is the specific heat capacity of water, is the chilled water return temperature, The supply temperature of chilled water; The heat loss model of the chilled water pipeline is established. The expression of the heat loss model of the chilled water pipeline is: In the formula, is the pipeline heat loss, is the thermal conductivity of the insulation material, is the average temperature of chilled water, is the ambient temperature, is the pipe length, is the inner radius of the pipe, is the outer radius of the insulation layer; Establishing the refrigeration mechanism model of the water-cooled air-conditioning system also includes: The cooling tower model is established, and the expression of the cooling tower model is: In the formula, is the heat dissipation of the cooling tower, is the cooling water mass flow rate, is the specific heat capacity of water, is the cooling water return temperature, The supply temperature of cooling water; The air handling unit model is established, and the expression of the air handling unit model is: In the formula, is the cooling capacity of the air handling unit, is the air mass flow rate, is the specific heat capacity of air, is the return air temperature, is the supply air temperature; The air duct heat loss model is established, and the expression of the air duct heat loss model is: In the formula, is the heat loss in the air duct, is the thermal conductivity of the duct insulation material, is the average cooling air temperature, is the ambient temperature, is the air duct length, is the inner radius of the air duct, is the outer radius of the duct insulation layer; The heat load maximum flow graph model and the cooling maximum flow graph model are trained by the attention mechanism to obtain a heat load graph feature matrix and a cooling graph feature matrix, including: Extracting a first eigenvector of a first vertex in the heat load map feature matrix after being mapped in the heat load maximum flow map model, and extracting a second eigenvector of a second vertex in the cooling map feature matrix after being mapped in the cooling maximum flow map model; Calculate the neighborhood feature vectors of the first vertex and the second vertex with respect to any child node respectively, and calculate the attention scores of the first vertex and the second vertex with respect to any energy-saving control strategy according to the neighborhood feature vectors; The first eigenvector is obtained based on a first weight learning matrix preset for the first vertex and a first parameter to be learned, and the second eigenvector is obtained based on a second weight learning matrix preset for the second vertex and a second parameter to be learned.

2. The method for energy-saving control of water-cooled air conditioners in manufacturing plants based on load forecasting according to claim 1 is characterized in that: The method of predicting the current operation data of the water-cooled air-conditioning system according to the heat load map characteristic matrix and the cooling map characteristic matrix to output an energy-saving control strategy for the water-cooled air-conditioning system includes: Calculating the strength of association between any two sub-nodes based on the attention score to obtain weighted feature vectors of the sub-nodes under the first vertex and the second vertex, and aggregating the weighted feature vectors to obtain the heat load map feature matrix and the cooling map feature matrix; Acquire the current operation data of the water-cooled air-conditioning system, and predict the current operation data by calling the softmax function according to the heat load map feature matrix and the cooling map feature matrix to obtain the energy-saving control strategy; The energy-saving optimization strategy is the predicted operating data after energy-saving optimization based on the current operating data, and at least includes all physical quantity data involved after energy-saving optimization.

3. A water-cooled air conditioning energy-saving control device for a manufacturing plant based on load forecasting, characterized in that: The device is used to implement the energy-saving control method of water-cooled air conditioner in a manufacturing plant based on load forecasting as described in any one of claims 1 to 2, and comprises: A heat load mechanism model building module is used to build a heat load mechanism model for a water-cooled air conditioning system. The heat load mechanism model includes a transient heat conduction model for wall heat transfer, a solar radiation heat model, a heat dissipation model for equipment, and a fresh air latent heat model. A heat load maximum flow diagram model establishment module, used to establish a heat load maximum flow diagram model based on the heat load mechanism model; A refrigeration mechanism model building module is used to build a refrigeration mechanism model of the water-cooled air-conditioning system, wherein the refrigeration mechanism model includes a chiller model, a chilled water pipeline heat loss model, a cooling tower model, an air handling unit model, and an air duct heat loss model; A refrigeration maximum flow graph model establishment module, used to establish a refrigeration maximum flow graph model based on the refrigeration mechanism model; An attention mechanism training module, used for training the heat load maximum flow graph model and the cooling maximum flow graph model through an attention mechanism to obtain a heat load graph feature matrix and a cooling graph feature matrix; The energy-saving strategy prediction module is used to predict the current operation data of the water-cooled air-conditioning system according to the heat load map characteristic matrix and the cooling map characteristic matrix to output the energy-saving control strategy of the water-cooled air-conditioning system.

4. A terminal comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1-2.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 2 are implemented.

Images