A method and device for joint prediction and control of load and room temperature of an airport cooling system

Abstract

The present invention discloses a load-room temperature joint predictive control method and device for an airport cooling system, which relates to the field of predictive control, including: establishing a cooling energy attenuation estimation model for cooling pipes of an airport cooling system; constructing a load-room temperature joint predictive control model for an airport cooling system by combining a trained cooling load prediction model and a trained indoor cooling temperature prediction model; constructing an objective function of the load-room temperature joint predictive control model based on the indoor cooling temperature of the next time step, the chilled water supply temperature of the current time step, the penalty factor of the current time step, and the indoor comfort reference temperature, taking minimization of the objective function as the optimization goal, and using a particle swarm algorithm to solve the chilled water supply temperature of the next time step. The present invention solves the problems of insufficient prediction accuracy, poor control flexibility, and insufficient energy consumption optimization faced by the predictive control strategy in the existing airport cooling system.

Description

Technical Field

[0001] The invention relates to the field of predictive control, and in particular to a method and device for combined predictive control of load and room temperature of an airport cooling system. Background Art

[0002] With the continuous increase in air transport demand and global climate change, the load demand of large airports is increasing day by day. In the humid and hot southern region, the cooling system accounts for a high proportion of the overall energy consumption of the airport. As an important equipment in the cooling system, the ice storage system (ISS)'s ice storage or ice melting decisions at different times and the corresponding output power are closely related to the economic benefits and cooling quality of the airport cooling system. Therefore, implementing an efficient and energy-saving predictive control strategy for the ice storage system is an effective solution to achieve economical and efficient operation of the airport cooling system (ACS).

[0003] As a high-energy consumption, large-scale energy system, the cooling load and indoor cooling temperature of the airport cooling system have nonlinear and lagging characteristics. Existing prediction technologies mostly focus on modeling and predicting single-target cooling loads or indoor cooling temperatures, and rarely consider the interaction and feedback relationship between the two. On the other hand, traditional ice storage system control strategies mostly make decisions according to fixed time periods or based on market electricity prices. The above control strategies are economical to a certain extent, but lack flexibility, which will lead to problems such as high energy consumption and inaccurate comfort control during the regulation process, and cannot fully tap the energy-saving potential of ice storage systems.

[0004] Since the southern region has a subtropical marine monsoon climate with high temperatures and high humidity, the airport cooling system has stricter requirements for indoor temperature control. If the indoor comfort is damaged during the control process, it will seriously affect the cooling quality of the system. At the same time, the cooling system relies on the cooling medium to store and transport cold energy. Due to the influence of factors such as thermal radiation, the cooling system has the phenomenon of cold energy attenuation. This cold energy attenuation characteristic will reduce the accuracy and economy of system control. The predictive control strategy in the existing airport cooling system faces the limitations of insufficient prediction accuracy, poor control flexibility, and insufficient energy consumption optimization. It is difficult to effectively deal with the nonlinearity, hysteresis and dynamic changes of the system's external environment. Summary of the invention

[0005] The purpose of this application is to propose a method and device for joint prediction and control of load and room temperature of an airport cooling system in response to the above-mentioned technical problems.

[0006] In a first aspect, the present invention provides a method for joint prediction and control of load and room temperature of an airport cooling system, comprising the following steps:

[0007] The airport cooling system is built based on ice storage system and chiller, and the cooling energy decay estimation model of the cooling pipeline of the airport cooling system is established considering the cooling energy decay characteristics of the airport cooling system.

[0008] Build and train a cooling load prediction model for the airport cooling system based on LSTM to obtain a trained cooling load prediction model; build and train an indoor cooling temperature prediction model for the airport cooling system based on Informer to obtain a trained indoor cooling temperature prediction model;

[0009] A load-room temperature joint prediction control model of the airport cooling system is constructed by combining the trained cooling load prediction model and the trained indoor cooling temperature prediction model; the outdoor ambient temperature at the current time step, the outdoor light intensity at the current time step, the outdoor wind speed at the current time step and the historical data of the cooling load data are obtained and normalized respectively, and an input matrix is ​​constructed with the normalized historical data of the outdoor ambient temperature at the current time step, the outdoor light intensity at the current time step, the outdoor wind speed at the current time step and the cooling load data, and the input matrix is ​​input into the trained cooling load prediction model to obtain the predicted value of the cooling load at the current time step, and the predicted value of the cooling load at the current time step is denormalized to obtain the cooling load data at the current time step, according to The indoor cooling temperature of the current time step is calculated by the cooling load data of the current time step and the chilled water supply temperature of the current time step; the outdoor ambient temperature of the next time step, the outdoor light intensity of the next time step, the indoor cooling temperature of the current time step and the chilled water supply temperature of the current time step are obtained to form an input sequence, and the input sequence is input into the trained indoor cooling temperature prediction model to predict the indoor cooling temperature of the next time step; the objective function of the load-room temperature joint prediction control model is constructed based on the indoor cooling temperature of the next time step, the chilled water supply temperature of the current time step, the penalty factor of the current time step and the indoor comfort reference temperature, and the particle swarm algorithm is used to solve the chilled water supply temperature of the next time step with the minimization of the objective function as the optimization goal;

[0010] The chilled water return temperature and the operation decision of the ice storage system at the next time step are obtained, and the chilled water supply temperature and the chilled water return temperature at the next time step are combined with the cooling pipe cold energy decay estimation model to calculate the actual cooling power of the airport cooling system. Based on the actual cooling power of the airport cooling system and the operation decision of the ice storage system, the cooling cost saved by the airport cooling system is calculated.

[0011] Preferably, the historical data of the outdoor ambient temperature at the current time step, the outdoor light intensity at the current time step, the outdoor wind speed at the current time step, and the cooling load data are obtained and normalized respectively, and an input matrix is ​​constructed with the normalized historical data of the outdoor ambient temperature at the current time step, the outdoor light intensity at the current time step, the outdoor wind speed at the current time step, and the cooling load data, and the input matrix is ​​input into the trained cooling load prediction model to obtain the predicted value of the cooling load at the current time step, and the predicted value of the cooling load at the current time step is denormalized to obtain the cooling load data at the current time step, specifically including:

[0012] The outdoor ambient temperature at the kth time step, the outdoor light intensity at the kth time step, the outdoor wind speed at the kth time step, and the cooling load data of the n time steps before the kth time step are obtained, and normalized according to the data of the corresponding categories, respectively, to obtain the outdoor ambient temperature at the kth time step, the outdoor light intensity at the kth time step, the outdoor wind speed at the kth time step, and the cooling load data of the n time steps before the kth time step after normalization;

[0013] The normalized outdoor ambient temperature at the kth time step, the outdoor light intensity at the kth time step, the outdoor wind speed at the kth time step, and the cooling load data of the n time steps before the kth time step are used to construct an input matrix and input it into the trained cooling load prediction model. The trained cooling load prediction model includes an LSTM module and a fully connected layer. The input matrix passes through the LSTM module to obtain a one-dimensional column vector, which is input into the fully connected layer to obtain the predicted value of the cooling load at the kth time step. The predicted value of the cooling load at the kth time step is denormalized to obtain the cooling load data at the kth time step.

[0014] Preferably, the indoor cooling temperature at the current time step is calculated based on the cooling load data at the current time step and the chilled water supply temperature at the current time step, specifically including:

[0015] The indoor cooling temperature at the kth time step is calculated using the following formula:

[0016] ;

[0017] in, is the indoor cooling temperature at the kth time step, is the cooling load data of the kth time step; is the chilled water supply temperature at the kth time step, The building heat capacity of the airport cooling system is used to describe the thermal hysteresis effect of the building. is the cooling efficiency factor.

[0018] Preferably, the outdoor ambient temperature at the next time step, the outdoor light intensity at the next time step, the indoor cooling temperature at the current time step, and the chilled water supply temperature at the current time step are obtained and constitute an input sequence, and the input sequence is input into a trained indoor cooling temperature prediction model to predict the indoor cooling temperature at the next time step; the objective function of the load-room temperature joint prediction control model is constructed based on the indoor cooling temperature at the next time step, the chilled water supply temperature at the current time step, the penalty factor at the current time step, and the indoor comfort reference temperature, and the minimization of the objective function is taken as the optimization goal, and the particle swarm algorithm is used to solve the chilled water supply temperature at the next time step, which specifically includes:

[0019] The outdoor ambient temperature at the k+1th time step, the outdoor light intensity at the k+1th time step, the indoor cooling temperature at the kth time step, and the chilled water supply temperature at the kth time step are obtained and constituted into an input sequence, and the input sequence is input into the trained indoor cooling temperature prediction model. The trained indoor cooling temperature prediction model adopts the Informer model to predict the indoor cooling temperature at the k+1th time step, as shown in the following formula:

[0020] ;

[0021] in, represents the indoor cooling temperature at the k+1th time step, Represents the function corresponding to the Informer model, represents the outdoor ambient temperature at the k+1th time step, represents the solar radiation intensity at the k+1th time step, represents the indoor cooling temperature at the kth time step, represents the chilled water supply temperature at the kth time step;

[0022] The objective function J of the load-room temperature joint predictive control model is constructed using the following formula to minimize the deviation of the indoor cooling temperature and the cooling energy consumption of the airport cooling system:

[0023] ;

[0024] Where N is the prediction range, is the penalty coefficient, and is the scale factor, is the indoor cooling temperature at the k+1th time step, is the indoor comfort reference temperature, is the chilled water supply temperature at the kth time step, is the penalty factor for the kth time step;

[0025] The particle swarm algorithm optimizes the objective function by simulating group behavior to ensure that the indoor cooling temperature at the k+1th time step is within the comfortable range and minimize the cooling energy consumption, and finally converges to the global optimal solution and outputs the chilled water supply temperature at the k+1th time step.

[0026] As a preferred embodiment, the expression of the cooling energy attenuation estimation model of the cooling pipe of the airport cooling system is as follows:

[0027] ;

[0028] in, is the cooling capacity lost in the cooling pipe at time t, in GJ / (m•h); is the outer diameter of the rth layer of cold insulation material, is the outer diameter of the r-1th layer of cold insulation material, in meters; is the temperature of the chilled water in the pipeline at time t, is the temperature of the pipe surface at time t, in K; q is the number of layers of cold insulation material; is the thermal conductivity of the rth layer of cold insulation material, in GJ / (m•h).

[0029] As a preferred method, the chilled water return temperature and the operation decision of the ice storage system at the next time step are obtained, and the chilled water supply temperature and the chilled water return temperature at the next time step are combined with the cooling energy attenuation estimation model of the cooling pipeline to calculate the actual cooling power of the airport cooling system. Based on the actual cooling power of the airport cooling system and the operation decision of the ice storage system, the cooling cost saved by the airport cooling system is calculated, specifically including:

[0030] The following formula is used to calculate the cooling power of the airport cooling system at time t:

[0031] ;

[0032] ;

[0033] in, , is the length of each time step, represents the cooling power of the airport cooling system at time t, is the specific heat capacity of water; is the mass flow rate of chilled water, and are the chilled water supply temperature and chilled water return temperature of the ice storage system at time t, respectively; LPM is the chilled water flow rate, in (L / min);

[0034] After considering the cooling capacity attenuation characteristics of the airport cooling system, the cooling power of the airport cooling system at time t is combined with the cooling pipe cooling energy attenuation estimation model to calculate the actual cooling power of the airport cooling system at time t, as shown in the following formula:

[0035] ;

[0036] in, represents the actual cooling power of the airport cooling system at time t;

[0037] The following formula is used to calculate the actual power consumption of the airport cooling system at time t after considering the cooling capacity decay characteristics:

[0038] ;

[0039] in, is the actual power consumption of the airport cooling system at time t after considering the cooling capacity decay characteristics, COP is the cooling energy efficiency coefficient;

[0040] The cooling cost saved by introducing the airport cooling system at time t is shown in the following formula:

[0041] ;

[0042] in, The cooling cost saved by the airport cooling system at time t, is the decision factor of the ice storage system, and its value can only be 1 or -1. The operation decision of the ice storage device in the ice storage system is divided into ice storage charging mode and ice melting energy release mode. When the operation decision of the ice storage device in the ice storage system selects the ice storage charging mode, then , when the operation decision of the ice storage device in the ice storage system selects the ice melting energy release mode, then When the ice storage device in the ice storage system does not work, ; is the actual power consumption at time t after considering the cooling capacity decay; is the peak and valley electricity price used by the airport cooling system at time t, is the penalty function for the airport cooling system violating the constraint at time t, is the weight coefficient corresponding to the penalty function;

[0043] The penalty function for violating the constraint of the airport cooling system at time t is expressed as follows:

[0044] ;

[0045] in, is the ice storage capacity of the ice storage device in the ice storage system, It is the maximum ice storage capacity of the ice storage device in the ice storage system.

[0046] In a second aspect, the present invention provides a load-room temperature joint prediction control device for an airport cooling system, comprising the following steps:

[0047] The cooling energy decay estimation module is configured to construct an airport cooling system based on an ice storage system and a chiller, consider the cooling energy decay characteristics of the airport cooling system, and establish a cooling energy decay estimation model for cooling pipes of the airport cooling system;

[0048] The model building module is configured to build and train a cooling load prediction model of the airport cooling system based on LSTM to obtain a trained cooling load prediction model; build and train an indoor cooling temperature prediction model of the airport cooling system based on Informer to obtain a trained indoor cooling temperature prediction model;

[0049] The prediction module is configured to construct a load-room temperature joint prediction control model of the airport cooling system by combining a trained cooling load prediction model and a trained indoor cooling temperature prediction model; obtain the outdoor ambient temperature of the current time step, the outdoor light intensity of the current time step, the outdoor wind speed of the current time step and the historical data of the cooling load data and perform normalization processing respectively, construct an input matrix with the normalized historical data of the outdoor ambient temperature of the current time step, the outdoor light intensity of the current time step, the outdoor wind speed of the current time step and the cooling load data, input the input matrix into the trained cooling load prediction model, obtain the predicted value of the cooling load of the current time step, perform denormalization processing on the predicted value of the cooling load of the current time step, and obtain the cooling load of the current time step. Data, calculate the indoor cooling temperature of the current time step according to the cooling load data of the current time step and the chilled water supply temperature of the current time step; obtain the outdoor ambient temperature of the next time step, the outdoor light intensity of the next time step, the indoor cooling temperature of the current time step and the chilled water supply temperature of the current time step to form an input sequence, input the input sequence into the trained indoor cooling temperature prediction model, and predict the indoor cooling temperature of the next time step; construct the objective function of the load-room temperature joint prediction control model based on the indoor cooling temperature of the next time step, the chilled water supply temperature of the current time step, the penalty factor of the current time step and the indoor comfort reference temperature, take minimization of the objective function as the optimization goal, and use the particle swarm algorithm to solve the chilled water supply temperature of the next time step;

[0050] The cost calculation module is configured to obtain the chilled water return temperature and the operation decision of the ice storage system at the next time step, combine the chilled water supply temperature and the chilled water return temperature at the next time step with the cooling pipe cold energy decay estimation model to calculate the actual cooling power of the airport cooling system, and calculate the cooling cost saved by the airport cooling system based on the actual cooling power of the airport cooling system and the operation decision of the ice storage system.

[0051] In a third aspect, the present invention provides an electronic device comprising one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner in the first aspect.

[0052] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in any implementation manner in the first aspect.

[0053] In a fifth aspect, the present invention provides a computer program product, comprising a computer program, which, when executed by a processor, implements the method described in any implementation manner in the first aspect.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] (1) The load-room temperature joint predictive control method for the airport cooling system proposed in this invention is aimed at the cooling system of a large airport. It takes into account the impact of the cooling capacity decay phenomenon in the cooling system on the cooling accuracy and quality, and constructs a cooling energy decay estimation model for the cooling pipes of the airport cooling system.

[0056] (2) The airport cooling system load-room temperature joint prediction and control method proposed in the present invention combines the energy consumption and geographical location characteristics of the airport, and adopts the LSTM-based airport cooling system cooling load prediction model to predict the cooling load data of the airport cooling system. Based on this, the indoor cooling temperature is predicted and controlled, which lays a foundation for realizing accurate cooling of the airport cooling system and creating a comfortable cooling environment for passengers.

[0057] (3) The load-room temperature joint predictive control method for the airport cooling system proposed in the present invention establishes an indoor cooling temperature prediction model for the airport cooling system based on Informer. The Informer model has higher prediction accuracy, and combines the trained cooling load prediction model and the trained indoor cooling temperature prediction model to construct a load-room temperature joint predictive control model for the airport cooling system. The objective function of the load-room temperature joint predictive control model is set, and the particle swarm algorithm is used to solve the chilled water supply temperature of the next time step. By adjusting the control variable of the chilled water supply temperature, this method can more effectively measure the relationship between parameters such as indoor cooling temperature and environmental factors, and provide a model basis and decision support for regulation.

[0058] (4) The method for joint prediction and control of load and room temperature of the airport cooling system proposed in the present invention combines the cooling capacity attenuation characteristics existing in the cooling system to propose an evaluation index that is more in line with the actual operation of the system. The cooling cost saved by the airport cooling system is used to measure the feasibility and effectiveness of the method proposed in the present invention in the prediction and control of room temperature of the airport cooling system, which provides an important reference for the intelligent, fast, economical and efficient development of the airport cooling system. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0060] Figure 1 A schematic flow chart of a method for joint prediction and control of load and room temperature of an airport cooling system according to an embodiment of the present application;

[0061] Figure 2 A schematic diagram of the architecture of an airport cooling system for an airport cooling system load-room temperature joint prediction control method according to an embodiment of the present application;

[0062] Figure 3 A schematic diagram of the structure of a cooling pipe with a multi-layer cold-insulation structure in an airport cooling system of an airport cooling system load-room temperature joint prediction control method according to an embodiment of the present application;

[0063] Figure 4 A schematic diagram of a load-room temperature joint predictive control model of a load-room temperature joint predictive control method for an airport cooling system according to an embodiment of the present application;

[0064] Figure 5A peak-valley electricity price variation trend diagram used in the load-room temperature joint prediction control method for the airport cooling system of an embodiment of the present application;

[0065] Figure 6 A schematic diagram of a load-room temperature joint prediction control device for an airport cooling system according to an embodiment of the present application;

[0066] Figure 7 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0067] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0068] Figure 1 The embodiment of the present application provides a method for joint prediction and control of load and room temperature of an airport cooling system, comprising the following steps:

[0069] S1, construct an airport cooling system based on ice storage system and chillers, consider the cooling capacity decay characteristics of the airport cooling system, and establish a cooling energy decay estimation model for the cooling pipeline of the airport cooling system.

[0070] In a specific embodiment, the expression of the cooling energy attenuation estimation model of the cooling pipe of the airport cooling system is as follows:

[0071] ;

[0072] in, is the cooling capacity lost in the cooling pipe at time t, in GJ / (m•h); is the outer diameter of the rth layer of cold insulation material, is the outer diameter of the r-1th layer of cold insulation material, in meters; is the temperature of the chilled water in the pipeline at time t, is the temperature of the pipe surface at time t, in K; q is the number of layers of cold insulation material; is the thermal conductivity of the rth layer of cold insulation material, in GJ / (m•h).

[0073] Specifically, the Airport Cooling System (ACS) uses an ice storage system (ISS) and a chiller to jointly cool the load side of the airport system. Based on the energy flow in the airport cooling system, a mathematical model of the ice storage system is established, and its operation mechanism is explained. The energy flow and working characteristics of the ice storage system in the airport cooling system are analyzed. The airport cooling system is mainly composed of an ice storage system and a chiller. The chilled water in the airport cooling system is distributed to the load through a water distributor and pipeline transportation. After completing the heat exchange, it becomes high-temperature return water and is sent back to the airport cooling system for re-cooling.

[0074] Further analysis of the characteristics of the airport cooling system shows that there is a phenomenon of cooling energy attenuation (CEA) in the airport cooling system. In the cooling process of the airport cooling system, due to the existence of factors such as thermal radiation, temperature difference between the pipeline and the external environment, and fluid resistance, the cooling energy of the chilled water is attenuated during the transmission process, that is, there is energy loss in the process of transmitting the cooling energy from the cold source to the terminal side. A detailed analysis of the cooling energy attenuation characteristic of the airport cooling system is conducted, and a cooling energy attenuation estimation model for the cooling pipeline with a length of L at time t in the airport cooling system is constructed. The model reflects the changing trend of the cooling energy lost in the cooling pipeline at time t.

[0075] S2, construct and train a cooling load prediction model for the airport cooling system based on LSTM to obtain a trained cooling load prediction model; construct and train an indoor cooling temperature prediction model for the airport cooling system based on Informer to obtain a trained indoor cooling temperature prediction model.

[0076] Specifically, the cooling load in the airport cooling system is affected by many factors and has obvious time series characteristics, such as daily periodicity. At the same time, the cooling load is also affected by factors such as outdoor ambient temperature, passenger flow and behavior, which makes the cooling load show nonlinear changes. LSTM can capture relevant time dependencies through its memory unit structure, has the ability to handle complex nonlinear relationships, can more accurately reflect the actual changes in load, and improve the accuracy of prediction. LSTM is a variant of the recurrent neural network structure, which can solve the gradient vanishing and gradient explosion problems encountered by standard recurrent neural networks when processing long sequence data. Therefore, a cooling load prediction model for the airport cooling system based on LSTM is constructed to predict the cooling load. The cooling load prediction model includes LSTM modules and fully connected layers connected in sequence.

[0077] The input and output variables are the key to determine the performance of the model prediction. Among them, the input variables of the cooling load prediction model are various attributes that affect the cooling load prediction of the airport cooling system. For the airport cooling system, its cooling load is strongly correlated with environmental factors and is affected by factors such as outdoor temperature, light radiation intensity, and outdoor wind speed. At the same time, the cooling load is a type of load with a large inertia constant coefficient, that is, the cooling load of the predicted period has a large correlation with the cooling load data of the previous day. Therefore, the outdoor ambient temperature at the current time step, the outdoor light intensity at the current time step, the outdoor wind speed at the current time step, and the historical data of the cooling load data are selected as the input variables of the cooling load prediction model, and the predicted value of the cooling load at the current time step is used as the output variable of the cooling load prediction model. In order to solve the problem of different numerical ranges of different input data and speed up the convergence of the model, the outdoor ambient temperature of the current time step, the outdoor light intensity of the current time step, the outdoor wind speed of the current time step and the historical data of the cooling load data are obtained and used as the original data, and the original data is normalized, and the original data is normalized to the interval of [0,1] to obtain the outdoor ambient temperature of the current time step, the outdoor light intensity of the current time step, the outdoor wind speed of the current time step and the historical data of the cooling load data after normalization. In the cooling load prediction model, the LSTM module is composed of an input layer, multiple hidden layers and an output layer. The output of the previous hidden layer is used as the input of the next hidden layer. The input layer and the hidden layer jointly realize the extraction of the input load data features. The output of the last hidden layer is a one-dimensional column vector. After linear regression through the fully connected layer, the predicted value of the cooling load of the current time step of the system cooling load data is obtained. In the embodiment of the present application, the LSTM structure adopted by the LSTM module is an existing neural network model, and its specific structure is not repeated here.

[0078] In the training process of the cooling load prediction model, the mean square error is used as the loss function. After training, the trained cooling load prediction model is obtained. After the predicted value of the cooling load at the current time step is predicted using the trained cooling load prediction model, in order to convert it back to the original load value, it is necessary to perform a denormalization process to obtain the cooling load data at the current time step.

[0079] In the humid and hot southern region, the indoor cooling temperature in the cooling system of a large airport is affected by many factors, such as outdoor ambient temperature, solar radiation intensity, etc. Therefore, the regulation of indoor cooling temperature is a complex nonlinear problem, and it is difficult for traditional linear models to obtain good prediction results. In view of the above problems, the Informer model, with its powerful long sequence processing capabilities, high computing efficiency and accurate prediction performance, is suitable for the prediction of indoor cooling temperature in complex systems such as the airport cooling system, helping to achieve the dual goals of reducing cooling energy consumption, saving cooling costs and creating a comfortable environment for passengers. Therefore, the Informer model is used to predict the indoor cooling temperature in the airport cooling system, that is, the indoor cooling temperature prediction model of the airport cooling system adopts the Informer model.

[0080] The Informer model can effectively process long sequence data through a network structure based on the self-attention mechanism, which makes the Informer model have obvious advantages in predicting data with a long time span; by introducing the ProbSparse self-attention mechanism, the most important feature variables are focused on during the calculation process, thereby improving the calculation efficiency. This advantage is more significant when facing large-scale data processing; by selecting reasonable feature parameters, the Informer model can achieve higher prediction accuracy, smaller errors, and better generalization ability of prediction results.

[0081] Preprocess the input data. Select feature values ​​based on the Transfer Entrop (TE) principle. As a nonlinear method for feature selection, TE determines the importance of features by measuring the impact of one time series on the future state of another time series. Compared with traditional correlation analysis, TE can capture complex homomorphic dependencies and is more suitable for causal feature screening of time series data. After performing correlation analysis on the input data through TE, the outdoor ambient temperature of one time step, the outdoor light intensity of the next time step, the indoor cooling temperature of the current time step, and the chilled water supply temperature of the current time step are selected as the input data of the indoor cooling temperature prediction model, and the indoor cooling temperature of the next time step is used as the output data of the indoor cooling temperature prediction model.

[0082] The architecture of the Informer model consists of four main parts: input layer, output layer, encoder, and decoder.

[0083] Input layer: The input of the model is multidimensional time series data, including historical data and meteorological parameters. The historical data is the indoor cooling temperature at the current time step, which is used to capture the dynamic changes of the system; the meteorological parameters include the outdoor ambient temperature at the next time step, the outdoor light intensity at the next time step, and the chilled water supply temperature at the current time step, which are used to predict the indoor cooling temperature at the next time step. For each time step, the sliding window technology is used to select C time steps of data to form an input sequence with a dimension of C×D, where C represents the number of historical time steps considered by the model and D represents the number of input features at each time step.

[0084] Encoder: Through the ProbSparse self-attention mechanism, the input sequence is encoded to capture long-term dependencies and reduce computational complexity to obtain encoded information.

[0085] Decoder: The decoder processes the encoded information, combines historical data and prediction step length, and generates the indoor cooling temperature for the next time step.

[0086] Output layer: Output the indoor cooling temperature at the next time step.

[0087] In the embodiments of the present application, the following three indicators are used to evaluate the prediction results of the indoor cooling temperature prediction model: mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE). The indoor cooling temperature predicted by the cooling temperature prediction model and the actual value of the indoor cooling temperature are input into the formulas corresponding to the mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE), and the indicator results can be obtained. According to the indicator results, the best cooling temperature prediction model is determined as the trained cooling temperature prediction model, so that it can better match the cooling characteristics of the airport cooling system.

[0088] S3, combining the trained cooling load prediction model and the trained indoor cooling temperature prediction model to build a load-room temperature joint prediction control model for the airport cooling system; obtaining the outdoor ambient temperature at the current time step, the outdoor light intensity at the current time step, the outdoor wind speed at the current time step and the historical data of the cooling load data and normalizing them respectively, constructing an input matrix with the normalized historical data of the outdoor ambient temperature at the current time step, the outdoor light intensity at the current time step, the outdoor wind speed at the current time step and the cooling load data, inputting the input matrix into the trained cooling load prediction model to obtain the predicted value of the cooling load at the current time step, performing a denormalization process on the predicted value of the cooling load at the current time step to obtain the cooling load data at the current time step, The indoor cooling temperature of the current time step is calculated according to the cooling load data of the current time step and the chilled water supply temperature of the current time step; the outdoor ambient temperature of the next time step, the outdoor light intensity of the next time step, the indoor cooling temperature of the current time step and the chilled water supply temperature of the current time step are obtained to form an input sequence, and the input sequence is input into the trained indoor cooling temperature prediction model to predict the indoor cooling temperature of the next time step; the objective function of the load-room temperature joint prediction control model is constructed based on the indoor cooling temperature of the next time step, the chilled water supply temperature of the current time step, the penalty factor of the current time step and the indoor comfort reference temperature, and the particle swarm algorithm is used to solve the chilled water supply temperature of the next time step with the minimization of the objective function as the optimization goal.

[0089] In a specific embodiment, the outdoor ambient temperature at the current time step, the outdoor light intensity at the current time step, the outdoor wind speed at the current time step, and the historical data of the cooling load data are obtained and normalized respectively, and an input matrix is ​​constructed with the normalized historical data of the outdoor ambient temperature at the current time step, the outdoor light intensity at the current time step, the outdoor wind speed at the current time step, and the cooling load data, and the input matrix is ​​input into a trained cooling load prediction model to obtain a predicted value of the cooling load at the current time step, and the predicted value of the cooling load at the current time step is denormalized to obtain the cooling load data at the current time step, specifically including:

[0090] The outdoor ambient temperature at the kth time step, the outdoor light intensity at the kth time step, the outdoor wind speed at the kth time step, and the cooling load data of the n time steps before the kth time step are obtained, and normalized according to the data of the corresponding categories, respectively, to obtain the outdoor ambient temperature at the kth time step, the outdoor light intensity at the kth time step, the outdoor wind speed at the kth time step, and the cooling load data of the n time steps before the kth time step after normalization;

[0091] The normalized outdoor ambient temperature at the kth time step, the outdoor light intensity at the kth time step, the outdoor wind speed at the kth time step, and the cooling load data of the n time steps before the kth time step are used to construct an input matrix and input it into the trained cooling load prediction model. The trained cooling load prediction model includes an LSTM module and a fully connected layer. The input matrix passes through the LSTM module to obtain a one-dimensional column vector, which is input into the fully connected layer to obtain the predicted value of the cooling load at the kth time step. The predicted value of the cooling load at the kth time step is denormalized to obtain the cooling load data at the kth time step.

[0092] In a specific embodiment, the indoor cooling temperature at the current time step is calculated based on the cooling load data at the current time step and the chilled water supply temperature at the current time step, specifically including:

[0093] The indoor cooling temperature at the kth time step is calculated using the following formula:

[0094] ;

[0095] in, is the indoor cooling temperature at the kth time step, is the cooling load data of the kth time step; is the chilled water supply temperature at the kth time step, The building heat capacity of the airport cooling system is used to describe the thermal hysteresis effect of the building. is the cooling efficiency factor.

[0096] In a specific embodiment, the outdoor ambient temperature at the next time step, the outdoor light intensity at the next time step, the indoor cooling temperature at the current time step, and the chilled water supply temperature at the current time step are obtained to form an input sequence, and the input sequence is input into a trained indoor cooling temperature prediction model to predict the indoor cooling temperature at the next time step; the objective function of the load-room temperature joint prediction control model is constructed based on the indoor cooling temperature at the next time step, the chilled water supply temperature at the current time step, the penalty factor at the current time step, and the indoor comfort reference temperature, and the objective function is minimized as the optimization goal, and the chilled water supply temperature at the next time step is solved by using a particle swarm algorithm, which specifically includes:

[0097] The outdoor ambient temperature at the k+1th time step, the outdoor light intensity at the k+1th time step, the indoor cooling temperature at the kth time step, and the chilled water supply temperature at the kth time step are obtained and constituted into an input sequence, and the input sequence is input into the trained indoor cooling temperature prediction model. The trained indoor cooling temperature prediction model adopts the Informer model to predict the indoor cooling temperature at the k+1th time step, as shown in the following formula:

[0098] ;

[0099] in, represents the indoor cooling temperature at the k+1th time step, Represents the function corresponding to the Informer model, represents the outdoor ambient temperature at the k+1th time step, represents the solar radiation intensity at the k+1th time step, represents the indoor cooling temperature at the kth time step, represents the chilled water supply temperature at the kth time step;

[0100] The objective function J of the load-room temperature joint predictive control model is constructed using the following formula to minimize the deviation of the indoor cooling temperature and the cooling energy consumption of the airport cooling system:

[0101] ;

[0102] Where N is the prediction range, is the penalty coefficient, and is the scale factor, is the indoor cooling temperature at the k+1th time step, is the indoor comfort reference temperature, is the chilled water supply temperature at the kth time step, is the penalty factor for the kth time step;

[0103] The particle swarm algorithm optimizes the objective function by simulating group behavior to ensure that the indoor cooling temperature at the k+1th time step is within the comfortable range and minimize the cooling energy consumption. It finally converges to the global optimal solution and outputs the chilled water supply temperature at the k+1th time step.

[0104] Specifically, the trained cooling load prediction model and the trained indoor cooling temperature prediction model are combined to construct a load-room temperature joint prediction control model for the airport cooling system. The control objective of the load-room temperature joint prediction control model is to minimize the cooling cost of the airport cooling system while ensuring passenger comfort. The outdoor ambient temperature at the kth time step, the outdoor light intensity at the kth time step, the outdoor wind speed at the kth time step, and the cooling load data of the n time steps before the kth time step are normalized according to the data of the corresponding categories to obtain the normalized outdoor ambient temperature at the kth time step, the outdoor light intensity at the kth time step, the outdoor wind speed at the kth time step, and the cooling load data of the n time steps before the kth time step. The input matrix is ​​constructed with the normalized outdoor ambient temperature at the kth time step, the outdoor light intensity at the kth time step, the outdoor wind speed at the kth time step, and the cooling load data of the n time steps before the kth time step and is input into the trained cooling load prediction model to obtain the cooling load data of the kth time step. The chilled water supply temperature at the k-th time step is known, and the indoor cooling temperature at the k-th time step is calculated in combination with the cooling load data of the k-th time step predicted by the trained cooling load prediction model. The temperature is used as one of the input data of the trained indoor cooling temperature prediction model, and then the input sequence is obtained by combining the outdoor ambient temperature at the k+1-th time step, the outdoor light intensity at the k+1-th time step, and the chilled water supply temperature at the k-th time step and inputting it into the trained indoor cooling temperature prediction model to predict the indoor cooling temperature at the k+1-th time step.

[0105] Combined with the objective function of the defined load-room temperature joint predictive control model and the set value of the indoor comfort reference temperature, the particle swarm optimization (PSO) algorithm is used to optimize and solve the chilled water supply temperature at the k+1th time step. The PSO algorithm optimizes the chilled water supply temperature by simulating group behavior to ensure that the indoor cooling temperature is within the comfortable range and minimize the cooling energy consumption. In the process of the PSO algorithm, the position (i.e., the state of the system at this time) and speed of the particles are randomly initialized in the search space. These particles represent different chilled water supply temperature strategies. For each particle, its fitness value is calculated, and a lower fitness value indicates a better solution. According to the calculation process, based on the individual historical optimal position and the global optimal position, the speed and position of the particle are updated to explore a better chilled water supply temperature. Through iterative updates, the PSO algorithm gradually converges to the global optimal solution and finally outputs the best chilled water supply temperature strategy, that is, the chilled water supply temperature at the k+1th time step.

[0106] S4, obtain the chilled water return temperature and the operation decision of the ice storage system at the next time step, combine the chilled water supply temperature and the chilled water return temperature at the next time step with the cooling pipe cold energy attenuation estimation model to calculate the actual cooling power of the airport cooling system, and calculate the cooling cost saved by the airport cooling system based on the actual cooling power of the airport cooling system and the operation decision of the ice storage system.

[0107] In a specific embodiment, step S4 specifically includes:

[0108] The following formula is used to calculate the cooling power of the airport cooling system at time t:

[0109] ;

[0110] ;

[0111] in, , is the length of each time step, represents the cooling power of the airport cooling system at time t, is the specific heat capacity of water; is the mass flow rate of chilled water, and are the chilled water supply temperature and chilled water return temperature of the ice storage system at time t, respectively; LPM is the chilled water flow rate, in (L / min);

[0112] After considering the cooling capacity attenuation characteristics of the airport cooling system, the cooling power of the airport cooling system at time t is combined with the cooling pipe cooling energy attenuation estimation model to calculate the actual cooling power of the airport cooling system at time t, as shown in the following formula:

[0113] ;

[0114] in, represents the actual cooling power of the airport cooling system at time t;

[0115] The following formula is used to calculate the actual power consumption of the airport cooling system at time t after considering the cooling capacity decay characteristics:

[0116] ;

[0117] in, is the actual power consumption of the airport cooling system at time t after considering the cooling capacity decay characteristics, COP is the cooling energy efficiency coefficient;

[0118] The cooling cost saved by introducing the airport cooling system at time t is shown in the following formula:

[0119] ;

[0120] in, The cooling cost saved by the airport cooling system at time t, is the decision factor of the ice storage system, and its value can only be 1 or -1. The operation decision of the ice storage device in the ice storage system is divided into ice storage charging mode and ice melting energy release mode. When the operation decision of the ice storage device in the ice storage system selects the ice storage charging mode, then , when the operation decision of the ice storage device in the ice storage system selects the ice melting energy release mode, then When the ice storage device in the ice storage system does not work, ; is the actual power consumption at time t after considering the cooling capacity decay; is the peak and valley electricity price used by the airport cooling system at time t, is the penalty function for the airport cooling system violating the constraint at time t, is the weight coefficient corresponding to the penalty function;

[0121] The penalty function for violating the constraint of the airport cooling system at time t is expressed as follows:

[0122] ;

[0123] in, is the ice storage capacity of the ice storage device in the ice storage system, It is the maximum ice storage capacity of the ice storage device in the ice storage system.

[0124] Specifically, while ensuring the comfort of passengers in the airport cooling system, an evaluation index of the airport cooling system is established to ensure passenger comfort, that is, to ensure that the indoor cooling temperature is controlled within the passenger comfort range. Based on the analysis of the cooling capacity attenuation characteristics of the airport cooling system, combined with the obtained chilled water supply temperature at the k+1th time step, the cooling cost saved by the airport cooling system under the load-room temperature joint predictive control model proposed in the embodiment of the present application is calculated. When calculating the cooling power of the airport cooling system, the chilled water return temperature at the k+1th time step can be obtained from the records in the control process.

[0125] The following is described by specific examples:

[0126] Taking a large international airport in the southern region as an example, the schematic diagram of the airport cooling system (ACS) is as follows: Figure 2As shown. It can be seen that in this cooling system, the chilled water cooled to the set temperature is transported to the side that needs cooling through the pipeline system. The chilled water absorbs ambient heat through the heat exchanger, thereby reducing the indoor temperature. After absorbing heat, the temperature of the chilled water rises and becomes return water, which is then re-cooled by the pump to form a cycle. In the transmission pipeline, due to the existence of thermal radiation and other phenomena, there is a phenomenon of cold attenuation during the cold transport process. The structure of the cooling pipeline with a multi-layer cold insulation structure in the cold network system is shown in the figure. Figure 3 shown.

[0127] For large international airports in the southern region, the demand for cooling load is relatively large. The trained cooling load forecasting model is used to predict cooling load data. By learning the time series pattern of the cooling load data, its complex linear relationship is captured. The trained cooling load forecasting model can effectively handle long-term dependencies and predict the cooling load data at future moments by dynamically memorizing and updating the input sequence.

[0128] In the indoor cooling temperature prediction of the airport cooling system, the trained indoor cooling temperature prediction model processes multi-dimensional time series data, captures long-term dependencies and complex spatiotemporal features, and achieves efficient and accurate temperature prediction. The Informer model uses sparse matrix calculation based on the attention mechanism, which can maintain computational efficiency under large-scale data, adapt to the complex environmental conditions of the airport, provide refined room temperature control solutions, and optimize energy management.

[0129] Reference Figure 4 , the load-room temperature joint predictive control model adjusts the ice storage water supply temperature, and realizes precise control of the chilled water supply temperature by predicting the future trend of the chilled water supply temperature. The Informer model uses time series data to capture complex temperature dynamics, and the load-room temperature joint predictive control model optimizes the chilled water supply temperature according to the prediction results to maintain the stability of the indoor cooling temperature. After adopting the particle swarm optimization algorithm, the minimum objective function is used as the optimization goal. The solved chilled water supply temperature of the ice storage system can maintain the indoor cooling temperature of the airport cooling system at ±1.7°C of the set value, and can maintain the room temperature in a comfortable temperature range (23.9°C-27.3°C), creating a comfortable indoor cooling environment for passengers and meeting the indoor cooling environment requirements of the airport cooling system. In the embodiment of the present application, when calculating the cooling cost saved by the airport cooling system, the peak and valley electricity price change trend used is as follows Figure 5 See Figure 5 , we can know that the peak hours are: 8:30-11:30; 14:30-17:30; 19:00-21:00; normal hours: 7:00-8:30; 11:30-14:30; 17:30-19:00; 21:00-23:00; off-peak hours: 23:00-7:00.

[0130] It can be seen that the load-room temperature joint prediction control method of the airport cooling system proposed in the embodiment of the present application is feasible and effective. By combining the cooling characteristics of the ice storage device in the airport cooling system, a reasonable prediction control strategy is adopted to ensure that the indoor cooling temperature of the airport is in a comfortable range by regulating the chilled water supply temperature. At the same time, under the premise of meeting the energy consumption of the airport, the cooling cost of the airport is reduced as much as possible.

[0131] The embodiment of the present application, taking into account the cooling capacity decay characteristics, can improve the cooling quality of the airport cooling system and ensure the comfort experience of passengers by regulating the ice storage and ice melting decisions of the ice storage system. In the context of the continuous increase in the throughput of the aviation industry, a more accurate indoor cooling temperature prediction and control method is proposed for the cooling system of large airports, which controls the ice storage and ice melting decisions of the ice storage system, so as to improve the economic benefits of the cooling system while ensuring the cooling comfort of passengers in the airport.

[0132] Further references Figure 6 As an implementation of the methods shown in the above figures, the present application provides an embodiment of a load-room temperature joint prediction control device for an airport cooling system. Figure 1 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0133] The embodiment of the present application provides a load-room temperature joint prediction control device for an airport cooling system, comprising the following steps:

[0134] The cooling energy decay estimation module 1 is configured to construct an airport cooling system based on an ice storage system and a chiller, consider the cooling energy decay characteristics of the airport cooling system, and establish a cooling energy decay estimation model for cooling pipes of the airport cooling system;

[0135] The model building module 2 is configured to build and train a cooling load prediction model of the airport cooling system based on LSTM to obtain a trained cooling load prediction model; build and train an indoor cooling temperature prediction model of the airport cooling system based on Informer to obtain a trained indoor cooling temperature prediction model;

[0136] Prediction module 3 is configured to construct a load-room temperature joint prediction control model of the airport cooling system by combining a trained cooling load prediction model and a trained indoor cooling temperature prediction model; obtain the outdoor ambient temperature of the current time step, the outdoor light intensity of the current time step, the outdoor wind speed of the current time step and the historical data of the cooling load data and perform normalization processing respectively, construct an input matrix with the normalized historical data of the outdoor ambient temperature of the current time step, the outdoor light intensity of the current time step, the outdoor wind speed of the current time step and the cooling load data, input the input matrix into the trained cooling load prediction model, obtain the predicted value of the cooling load of the current time step, perform denormalization processing on the predicted value of the cooling load of the current time step, and obtain the cooling load of the current time step. Data, calculate the indoor cooling temperature of the current time step according to the cooling load data of the current time step and the chilled water supply temperature of the current time step; obtain the outdoor ambient temperature of the next time step, the outdoor light intensity of the next time step, the indoor cooling temperature of the current time step and the chilled water supply temperature of the current time step to form an input sequence, input the input sequence into the trained indoor cooling temperature prediction model, and predict the indoor cooling temperature of the next time step; construct the objective function of the load-room temperature joint prediction control model based on the indoor cooling temperature of the next time step, the chilled water supply temperature of the current time step, the penalty factor of the current time step and the indoor comfort reference temperature, take minimization of the objective function as the optimization goal, and use the particle swarm algorithm to solve the chilled water supply temperature of the next time step;

[0137] The cost calculation module 4 is configured to obtain the chilled water return temperature and the operation decision of the ice storage system at the next time step, and combine the chilled water supply temperature and the chilled water return temperature at the next time step with the cooling pipe cold energy attenuation estimation model to calculate the actual cooling power of the airport cooling system, and calculate the cooling cost saved by the airport cooling system based on the actual cooling power of the airport cooling system and the operation decision of the ice storage system.

[0138] Figure 7 Schematic diagram of the hardware structure of the electronic device provided by the embodiment of the present invention. Figure 7 As shown, the electronic device of this embodiment includes: a processor 701 and a memory 702; wherein the memory 702 is used to store computer-executable instructions; the processor 701 is used to execute the computer-executable instructions stored in the memory to implement the various steps performed by the electronic device in the above embodiment. For details, please refer to the relevant description in the above method embodiment.

[0139] Optionally, the memory 702 may be independent or integrated with the processor 701 .

[0140] When the memory 702 is independently provided, the electronic device further includes a bus 703 for connecting the memory 702 and the processor 701 .

[0141] The embodiment of the present invention further provides a computer storage medium, in which computer execution instructions are stored. When the processor 701 executes the computer execution instructions, the above method is implemented.

[0142] The embodiment of the present invention further provides a computer program product, including a computer program. When the computer program is executed by the processor 701, the above method is implemented.

[0143] In the embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.

[0144] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to implement the solution of this embodiment.

[0145] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each module may exist physically separately, or two or more modules may be integrated into one unit. The unit formed by the above modules may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0146] The above-mentioned integrated module implemented in the form of a software function module can be stored in a computer-readable storage medium. The above-mentioned software function module is stored in a storage medium, including a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor 701 to perform some steps of the methods of various embodiments of the present application.

[0147] It should be understood that the processor 701 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or the processor 701 may be any conventional processor 701, etc. The steps of the method disclosed in the invention may be directly embodied in the execution of the hardware processor 701, or may be executed by a combination of hardware and software modules in the processor 701.

[0148] The memory 702 may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least one disk memory, and may also be a USB flash drive, a mobile hard disk, a read-only memory, a magnetic disk, or an optical disk.

[0149] The bus 703 may be an Industry Standard Architecture (ISA), a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus 703 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus 703 in the drawings of the present application is not limited to only one bus 703 or one type of bus 703.

[0150] The above storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The storage medium can be any available medium that can be accessed by a general or special purpose computer.

[0151] An exemplary storage medium is coupled to the processor 701, so that the processor 701 can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor 701. The processor 701 and the storage medium can be located in an application specific integrated circuit (Application Specific Integrated Circuits, referred to as ASIC). Of course, the processor 701 and the storage medium can also exist as discrete components in an electronic device or a main control device.

[0152] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0153] 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 aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for joint prediction and control of load and room temperature of an airport cooling system, characterized in that: The following steps are involved: An airport cooling system is constructed based on an ice storage system and a chiller, and a cooling energy attenuation estimation model for cooling pipelines of the airport cooling system is established by taking into account the cooling energy attenuation characteristics of the airport cooling system; Build and train a cooling load prediction model for the airport cooling system based on LSTM to obtain a trained cooling load prediction model; build and train an indoor cooling temperature prediction model for the airport cooling system based on Informer to obtain a trained indoor cooling temperature prediction model; A load-room temperature joint prediction control model of the airport cooling system is constructed by combining the trained cooling load prediction model and the trained indoor cooling temperature prediction model; the outdoor ambient temperature of the current time step, the outdoor light intensity of the current time step, the outdoor wind speed of the current time step and the historical data of the cooling load data are obtained and normalized respectively, an input matrix is ​​constructed with the normalized historical data of the outdoor ambient temperature of the current time step, the outdoor light intensity of the current time step, the outdoor wind speed of the current time step and the cooling load data, the input matrix is ​​input into the trained cooling load prediction model to obtain the predicted value of the cooling load of the current time step, the predicted value of the cooling load of the current time step is denormalized to obtain the cooling load data of the current time step, and the cooling load data of the current time step is calculated according to the training data. The indoor cooling temperature of the current time step is calculated based on the cooling load data of the current time step and the chilled water supply temperature of the current time step; the outdoor ambient temperature of the next time step, the outdoor light intensity of the next time step, the indoor cooling temperature of the current time step and the chilled water supply temperature of the current time step are obtained to form an input sequence, and the input sequence is input into the trained indoor cooling temperature prediction model to predict the indoor cooling temperature of the next time step; the objective function of the load-room temperature joint prediction control model is constructed based on the indoor cooling temperature of the next time step, the chilled water supply temperature of the current time step, the penalty factor of the current time step and the indoor comfort reference temperature, and the minimization of the objective function is taken as the optimization goal, and the chilled water supply temperature of the next time step is solved by using a particle swarm algorithm; The chilled water return temperature and the operation decision of the ice storage system at the next time step are obtained, and the chilled water supply temperature and the chilled water return temperature at the next time step are combined with the cooling pipe cold energy decay estimation model to calculate the actual cooling power of the airport cooling system, and the cooling cost saved by the airport cooling system is calculated based on the actual cooling power of the airport cooling system and the operation decision of the ice storage system.

2. The method for joint prediction and control of load and room temperature of an airport cooling system according to claim 1, characterized in that: The historical data of the outdoor ambient temperature at the current time step, the outdoor light intensity at the current time step, the outdoor wind speed at the current time step, and the cooling load data are obtained and normalized respectively, and an input matrix is ​​constructed with the normalized historical data of the outdoor ambient temperature at the current time step, the outdoor light intensity at the current time step, the outdoor wind speed at the current time step, and the cooling load data, and the input matrix is ​​input into the trained cooling load prediction model to obtain the predicted value of the cooling load at the current time step, and the predicted value of the cooling load at the current time step is denormalized to obtain the cooling load data at the current time step, specifically including: The outdoor ambient temperature at the kth time step, the outdoor light intensity at the kth time step, the outdoor wind speed at the kth time step, and the cooling load data of the n time steps before the kth time step are obtained, and normalized according to the data of the corresponding categories, respectively, to obtain the outdoor ambient temperature at the kth time step, the outdoor light intensity at the kth time step, the outdoor wind speed at the kth time step, and the cooling load data of the n time steps before the kth time step after normalization; The normalized outdoor ambient temperature at the k-th time step, the outdoor light intensity at the k-th time step, the outdoor wind speed at the k-th time step, and the cooling load data of the n time steps before the k-th time step are used to construct an input matrix and input it into the trained cooling load prediction model. The trained cooling load prediction model includes an LSTM module and a fully connected layer. The input matrix passes through the LSTM module to obtain a one-dimensional column vector, which is input into the fully connected layer to obtain a predicted value of the cooling load at the k-th time step. The predicted value of the cooling load at the k-th time step is denormalized to obtain the cooling load data at the k-th time step.

3. The method for joint prediction and control of load and room temperature of an airport cooling system according to claim 1, characterized in that: The indoor cooling temperature at the current time step is calculated according to the cooling load data at the current time step and the chilled water supply temperature at the current time step, specifically including: The indoor cooling temperature at the kth time step is calculated using the following formula: in, is the indoor cooling temperature at the kth time step, is the cooling load data of the kth time step; is the chilled water supply temperature at the kth time step, C zone is the building heat capacity of the airport cooling system, used to describe the thermal hysteresis effect of the building, η c is the cooling efficiency factor.

4. The method for joint prediction and control of load and room temperature of an airport cooling system according to claim 1, characterized in that: The outdoor ambient temperature at the next time step, the outdoor light intensity at the next time step, the indoor cooling temperature at the current time step, and the chilled water supply temperature at the current time step are obtained to form an input sequence, and the input sequence is input into the trained indoor cooling temperature prediction model to predict the indoor cooling temperature at the next time step; the objective function of the load-room temperature joint prediction control model is constructed based on the indoor cooling temperature at the next time step, the chilled water supply temperature at the current time step, the penalty factor at the current time step, and the indoor comfort reference temperature, and the minimization of the objective function is taken as the optimization goal, and the particle swarm algorithm is used to solve the chilled water supply temperature at the next time step, which specifically includes: The outdoor ambient temperature at the k+1th time step, the outdoor light intensity at the k+1th time step, the indoor cooling temperature at the kth time step, and the chilled water supply temperature at the kth time step are obtained and constituted into an input sequence, and the input sequence is input into the trained indoor cooling temperature prediction model. The trained indoor cooling temperature prediction model adopts the Informer model to predict the indoor cooling temperature at the k+1th time step, as shown in the following formula: in, represents the indoor cooling temperature at the k+1th time step, f(▪) represents the function corresponding to the Informer model, represents the outdoor ambient temperature at the k+1th time step, SRI k+1 represents the solar radiation intensity at the k+1th time step, represents the indoor cooling temperature at the kth time step, represents the chilled water supply temperature at the kth time step; The objective function J of the load-room temperature joint predictive control model is constructed by the following formula, so that the deviation of the indoor cooling temperature and the cooling energy consumption of the airport cooling system are minimized: Among them, N is the prediction range, α is the penalty coefficient, β and λ are the scaling factors, is the indoor cooling temperature at the k+1th time step, T base is the indoor comfort reference temperature, is the chilled water supply temperature at the kth time step, ∈ k is the penalty factor for the kth time step; The particle swarm algorithm optimizes the objective function by simulating group behavior to ensure that the indoor cooling temperature at the k+1th time step is within a comfortable range and minimize the cooling energy consumption, and finally converges to the global optimal solution and outputs the chilled water supply temperature at the k+1th time step.

5. The method for joint prediction and control of load and room temperature of an airport cooling system according to claim 1, characterized in that: The expression of the cooling energy attenuation estimation model of the cooling pipe of the airport cooling system is as follows: Among them, ΔQ L is the cooling capacity lost in the cooling pipe at time t, in GJ / (m·h); d r is the outer diameter of the rth layer of cold insulation material, d r-1 is the outer diameter of the r-1th layer of cold insulation material, in meters; T pcw is the temperature of the chilled water in the pipeline at time t, T ps is the temperature of the pipe surface at time t, in K; q is the number of layers of cold insulation material; μ r is the thermal conductivity of the rth layer of cold insulation material, in GJ / (m·h), and L is the length of the cooling pipe.

6. The method for joint prediction and control of load and room temperature of an airport cooling system according to claim 5, characterized in that: Obtain the chilled water return temperature and the operation decision of the ice storage system at the next time step, combine the chilled water supply temperature and the chilled water return temperature at the next time step with the cooling pipeline cold energy decay estimation model to calculate the actual cooling power of the airport cooling system, and calculate the cooling cost saved by the airport cooling system based on the actual cooling power of the airport cooling system and the operation decision of the ice storage system, specifically including: The cooling power of the airport cooling system at time t is calculated using the following formula: m w =LPM·60min; Where t = (k+1)·Δt, k represents the kth time step, and Δt is the length of each time step. represents the cooling power of the airport cooling system at time t, c w is the specific heat capacity of water; m w is the mass flow rate of chilled water, T IWS and T IWR are the chilled water supply temperature and chilled water return temperature of the ice storage system at time t, respectively; LPM is the chilled water flow rate, in (L / min); After considering the cooling capacity decay characteristics of the airport cooling system, the cooling power of the airport cooling system at time t is combined with the cooling pipe cooling energy decay estimation model to calculate the actual cooling power of the airport cooling system at time t, as shown in the following formula: in, represents the actual cooling power of the airport cooling system at time t; The following formula is used to calculate the actual power consumption of the airport cooling system at time t after considering the cooling capacity decay characteristics: Among them, P t act is the actual power consumption of the airport cooling system at time t after considering the cooling capacity decay characteristics, COP is the cooling energy efficiency coefficient; The cooling cost saved by introducing the airport cooling system at time t is as follows: Among them, S t is the cooling cost saved by the airport cooling system at time t, ξ t is the decision factor of the ice storage system, and its value can only be 1, 0 or -1. The operation decision of the ice storage device in the ice storage system is divided into ice storage charging mode and ice melting energy release mode. When the operation decision of the ice storage device in the ice storage system selects the ice storage charging mode, then ξ t =1, when the operation decision of the ice storage device in the ice storage system selects the ice melting energy release mode, then ξ t = -1, when the ice storage device in the ice storage system does not work, then ξ t =0;P t act is the actual power consumption at time t after considering the cooling capacity decay; is the peak-valley electricity price used by the airport cooling system at time t, p t is the penalty function for the airport cooling system violating the constraint at time t, and γ is the weight coefficient corresponding to the penalty function; The penalty function for violating the constraint of the airport cooling system at time t is expressed as follows: in, is the ice storage capacity of the ice storage device in the ice storage system, It is the maximum ice storage capacity of the ice storage device in the ice storage system.

7. An airport cooling system load-room temperature joint prediction control device, characterized in that: include: A cooling energy decay estimation module is configured to construct an airport cooling system based on an ice storage system and a chiller, and to establish a cooling energy decay estimation model for cooling pipelines of the airport cooling system by taking into account cooling energy decay characteristics of the airport cooling system; The model building module is configured to build and train a cooling load prediction model of the airport cooling system based on LSTM to obtain a trained cooling load prediction model; build and train an indoor cooling temperature prediction model of the airport cooling system based on Informer to obtain a trained indoor cooling temperature prediction model; The prediction module is configured to construct a load-room temperature joint prediction control model of the airport cooling system by combining the trained cooling load prediction model and the trained indoor cooling temperature prediction model; obtain the outdoor ambient temperature of the current time step, the outdoor light intensity of the current time step, the outdoor wind speed of the current time step and the historical data of the cooling load data and perform normalization processing respectively, construct an input matrix with the normalized historical data of the outdoor ambient temperature of the current time step, the outdoor light intensity of the current time step, the outdoor wind speed of the current time step and the cooling load data, input the input matrix into the trained cooling load prediction model, obtain the predicted value of the cooling load of the current time step, perform denormalization processing on the predicted value of the cooling load of the current time step, and obtain the cooling load value of the current time step. According to the cooling load data of the current time step and the chilled water supply temperature of the current time step, the indoor cooling temperature of the current time step is calculated; the outdoor ambient temperature of the next time step, the outdoor light intensity of the next time step, the indoor cooling temperature of the current time step and the chilled water supply temperature of the current time step are obtained to form an input sequence, and the input sequence is input into the trained indoor cooling temperature prediction model to predict the indoor cooling temperature of the next time step; the objective function of the load-room temperature joint prediction control model is constructed based on the indoor cooling temperature of the next time step, the chilled water supply temperature of the current time step, the penalty factor of the current time step and the indoor comfort reference temperature, and the minimization of the objective function is taken as the optimization goal, and the chilled water supply temperature of the next time step is solved by using a particle swarm algorithm; The cost calculation module is configured to obtain the chilled water return temperature and the operation decision of the ice storage system at the next time step, combine the chilled water supply temperature and the chilled water return temperature at the next time step with the cooling pipe cold energy decay estimation model to calculate the actual cooling power of the airport cooling system, and calculate the cooling cost saved by the airport cooling system based on the actual cooling power of the airport cooling system and the operation decision of the ice storage system.

8. An electronic device comprising: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images