Energy-saving control method and device of central air conditioning system and medium

Abstract

The invention provides an energy-saving control method and device for a central air-conditioning system and a medium, and the method comprises the steps: constructing a cooling load prediction model of the central air-conditioning system, and predicting the cooling load per hour of the next day; under the conditions that the hourly cooling load of the next day is met and the power utilization cost is the lowest, the cooling capacity in a fixed time period is distributed; and an energy efficiency model and a function model of the central air-conditioning system are constructed, and the operation strategy of the central air-conditioning system corresponding to the given cooling capacity is optimized with the purpose of minimizing the power consumption cost, so that the comprehensive energy consumption of the central air-conditioning system is minimized. According to the method, the cooling load prediction result is improved by comprehensively considering various cooling load influence factors and effectively capturing the long-term dependency relationship in historical cooling load data, the comprehensive energy consumption minimization and the power consumption cost minimization of the central air-conditioning system serve as targets, the operation strategy is optimized through simulation combination and energy efficiency comparison, and the energy efficiency of the central air-conditioning system is improved. Therefore, the requirement of a target environment for an efficient and energy-saving central air conditioning system is met.

Description

Technical Field

[0001] This invention belongs to the field of central air conditioning automatic control technology, specifically relating to energy-saving control methods, devices and media for central air conditioning systems. Background Technology

[0002] Current central air conditioning system control methods primarily rely on simple temperature settings and timed on / off cycles. Cooling load forecasting typically employs methods such as linear regression based on historical data. However, these methods have limitations when dealing with complex influencing factors and long-term dependencies. For example, they struggle to comprehensively consider various cooling load influencing factors and cannot effectively capture long-term dependencies in historical cooling load data, leading to significant discrepancies between predicted cooling loads and actual demand. Furthermore, the allocation of cooling capacity and the determination of operating strategies are often based on fixed patterns or single control methods, failing to allow for flexible adjustments according to different operating conditions. This results in low energy efficiency and high electricity costs for central air conditioning systems.

[0003] Therefore, how to solve the aforementioned technical problems in the existing technology and better meet the needs of the target environment for a high-efficiency and energy-saving central air conditioning system has become an urgent technical problem to be solved. Summary of the Invention

[0004] In view of this, this application provides an energy-saving control method, device and medium for a central air conditioning system. By comprehensively considering various cooling load influencing factors and effectively capturing long-term dependencies in historical cooling load data, the cooling load prediction results are improved. With the goal of minimizing the overall energy consumption and electricity cost of the central air conditioning system, the operation strategy is optimized through simulation combination and energy efficiency comparison to meet the requirements of the target environment for a high-efficiency and energy-saving central air conditioning system.

[0005] In a first aspect, this application provides an energy-saving control method for a central air conditioning system, the method comprising at least the following steps: A central air conditioning system cooling load prediction model is constructed by using convolutional neural networks to extract cooling load influencing factors and long short-term memory networks to capture long-term dependencies in historical cooling load data, and the hourly cooling load for the next day is predicted. Under the condition of meeting the hourly cooling load of the next day and minimizing electricity costs, the cooling capacity within a fixed time period is allocated through a deep deterministic strategy gradient algorithm. An energy efficiency model and a functional model of a central air conditioning system are constructed. With the goal of minimizing electricity costs, the operation strategy of the central air conditioning system corresponding to a given cooling capacity is optimized to minimize the overall energy consumption of the central air conditioning system. The operation strategy adopts a combination of fuzzy logic control and model predictive control. For different load conditions under different wet-bulb temperatures, the operating conditions of the chiller, chilled water pump, cooling water pump, and cooling tower are matched. The optimal combination of operating modes is determined through simulation and energy efficiency comparison.

[0006] In one possible implementation, the method of constructing a cooling load prediction model for a central air conditioning system by employing a convolutional neural network to extract cooling load influencing factors and a long short-term memory network to capture long-term dependencies in historical cooling load data includes: Acquire data related to the cooling load of the central air conditioning system, including historical cooling load data, outdoor meteorological parameters, and indoor environmental parameters; The acquired data is preprocessed, including cleaning and normalization, to make it conform to the input requirements of the convolutional neural network-long short-term memory network model; The parameters of the convolutional neural network-long short-term memory network model are determined based on the characteristics of the acquired data, so that the features extracted by the convolutional neural network are input into the long short-term memory network to achieve feature fusion. Among them, the convolutional kernel of the convolutional neural network can extract local features and spatial information about the cold load influencing factors in the input data, and the long short-term memory network can handle the long-term dependencies in the historical cold load data of the time series. The convolutional neural network-long short-term memory network model is trained using preprocessed data, and the parameters of the convolutional neural network-long short-term memory network model are optimized by backpropagation and optimization algorithms to minimize the error between the predicted and actual values.

[0007] In one possible implementation, the allocation of cooling capacity within a fixed time period using a deep deterministic strategy gradient algorithm, under the condition of satisfying the hourly cooling load of the next day and minimizing electricity costs, includes: The deep deterministic strategy gradient algorithm dynamically adjusts the cooling capacity allocation strategy based on real-time electricity price information and cooling load changes to minimize electricity costs. The allocation strategy includes introducing ice storage or water storage systems to store cooling capacity during periods of low electricity prices and release it during periods of high electricity prices.

[0008] In one possible implementation, the construction of an energy efficiency model and a functional model of the central air conditioning system, and the optimization of the operation strategy of the central air conditioning system corresponding to a given cooling capacity with the goal of minimizing electricity costs, includes: In the energy efficiency model, a data-driven energy efficiency assessment method is introduced, which uses machine learning algorithms to analyze a large amount of operating data and predict the energy efficiency performance of the central air conditioning system under different operating conditions. In the functional model, fuzzy logic control is adopted to perform fuzzification processing and full-condition modeling of the chiller, chilled water pump, cooling water pump, and cooling tower based on actual operation and prediction data under different load conditions at different wet-bulb temperatures, forming a preliminary operation strategy. Through model predictive control, combined with the predicted cooling load and electricity price changes over a period of time, the central air conditioning system is optimized and controlled. The optimal combination of operating modes is determined through simulation combination and energy efficiency comparison.

[0009] In one possible implementation, the central air conditioning system comprises a chiller, chilled water pumps, cooling water pumps, a cooling tower, air handling units, fan coil units, a water collector, and a water distributor, which output water to different air outlets via ductwork; wherein, The input variables for the energy efficiency model of the chiller are cooling load, chilled water outlet temperature, cooling water return temperature, chilled water flow rate, and cooling water flow rate, and the output variables are chiller power and number of units. The input variable for the energy efficiency model of chilled water pumps is chilled water flow rate, and the output variables are chilled water pump frequency / power and number of pumps. The input variable for the energy efficiency model of the cooling water pump is the cooling water flow rate, and the output power is the cooling water pump frequency / power and the number of pumps. The input variables for the energy efficiency model of a cooling tower are the heat dissipation of the cooling tower and the outdoor temperature and humidity, and the output variables are the frequency / power and number of cooling towers.

[0010] In one possible implementation, the overall energy consumption of the central air conditioning system is determined by the following formula:

[0011] in, Let t be the total energy consumption of the central air conditioning system. This represents the real-time power of the chiller unit at time t. This represents the real-time power of the chilled water pump at time t. This represents the real-time power of the cooling water pump at time t. This represents the real-time power of the cooling tower at time t. This represents the penalty factor for not satisfying the constraints; This represents the total heat dissipation of the cooling tower at time t. Let t be the cooling load of the central air conditioning system. Let t be the cooling load demand. , , , , , These are, respectively, the chilled water return temperature, chilled water outlet temperature, cooling water outlet temperature, cooling water return temperature, cooling water return approximation temperature, and wet-bulb temperature at time t; By adding a penalty factor to the objective function given by the above formula, and considering the running time of the equipment when it has been started and the running time when it has been stopped, corresponding constraints are imposed to ensure the stable operation of the central air conditioning system and that the cooling capacity meets the demand.

[0012] Secondly, this application provides an energy-saving control device for a central air conditioning system, the device comprising at least: The prediction module is used to construct a cooling load prediction model for the central air conditioning system by using a convolutional neural network to extract cooling load influencing factors and a long short-term memory network to capture long-term dependencies in historical cooling load data, and to predict the hourly cooling load for the next day. The first optimization module is used to allocate the cooling capacity within a fixed time period using a deep deterministic strategy gradient algorithm, under the condition of meeting the hourly cooling load of the next day and minimizing the electricity cost. The second optimization module is used to construct the energy efficiency model and functional model of the central air conditioning system, and optimize the operation strategy of the central air conditioning system corresponding to the given cooling capacity with the goal of minimizing electricity costs, so as to minimize the overall energy consumption of the central air conditioning system. The operation strategy adopts a combination of fuzzy logic control and model predictive control, and matches the operating conditions of the chiller, chilled water pump, cooling water pump and cooling tower for different load conditions under different wet-bulb temperature conditions. The optimal combination of operating modes is determined through simulation combination and energy efficiency comparison.

[0013] In one possible implementation, the prediction module includes: The acquisition module is used to acquire data related to the cooling load of the central air conditioning system, including historical cooling load data, outdoor meteorological parameters, and indoor environmental parameters. The preprocessing module is used to preprocess the acquired data, including cleaning and normalization, to make it conform to the input requirements of the convolutional neural network-long short-term memory network model. The module is used to determine the parameters of the convolutional neural network-long short-term memory network model based on the characteristics of the acquired data, so as to input the features extracted by the convolutional neural network into the long short-term memory network to achieve feature fusion. Among them, the convolutional kernel of the convolutional neural network can extract local features and spatial information about the cold load influencing factors in the input data, and the long short-term memory network can handle the long-term dependencies in the historical cold load data of the time series. The training module is used to train the convolutional neural network-long short-term memory network model using preprocessed data, and optimizes the parameters of the convolutional neural network-long short-term memory network model through backpropagation and optimization algorithms to minimize the error between the predicted and actual values.

[0014] In one possible implementation, the first optimization module is used to dynamically adjust the allocation strategy of cooling capacity based on real-time electricity price information and cooling load changes using a deep deterministic strategy gradient algorithm, so as to achieve the goal of minimizing electricity costs; wherein, the allocation strategy includes introducing an ice storage system or a water storage system to store cooling capacity during periods of low electricity prices and release it during periods of high electricity prices.

[0015] In one possible implementation, the second optimization module is used to introduce a data-driven energy efficiency assessment method into the energy efficiency model, using machine learning algorithms to analyze a large amount of operating data and predict the energy efficiency performance of the central air conditioning system under different operating conditions; in the functional model, fuzzy logic control is used to perform fuzzification processing and full-condition modeling of the chiller, chilled water pump, cooling water pump, and cooling tower based on actual operating and predicted data under different load conditions under different dry / wet bulb temperature conditions, forming a preliminary operating strategy; through model predictive control, combined with the predicted cooling load and electricity price changes over a period of time, the central air conditioning system is optimized and controlled; and through simulation combination and energy efficiency comparison, the optimal combination of operating modes is determined.

[0016] In one possible implementation, the central air conditioning system comprises a chiller, chilled water pumps, cooling water pumps, a cooling tower, air handling units, fan coil units, a water collector, and a water distributor, which output water to different air outlets via ductwork; wherein, The input variables for the energy efficiency model of the chiller are cooling load, chilled water outlet temperature, cooling water return temperature, chilled water flow rate, and cooling water flow rate, and the output variables are chiller power and number of units. The input variable for the energy efficiency model of chilled water pumps is chilled water flow rate, and the output variables are chilled water pump frequency / power and number of pumps. The input variable for the energy efficiency model of the cooling water pump is the cooling water flow rate, and the output power is the cooling water pump frequency / power and the number of pumps. The input variables for the energy efficiency model of a cooling tower are the heat dissipation of the cooling tower and the outdoor temperature and humidity, and the output variables are the frequency / power and number of cooling towers.

[0017] In one possible implementation, the overall energy consumption of the central air conditioning system is determined by the following formula:

[0018] in, Let t be the total energy consumption of the central air conditioning system. This represents the real-time power of the chiller unit at time t. This represents the real-time power of the chilled water pump at time t. This represents the real-time power of the cooling water pump at time t. This represents the real-time power of the cooling tower at time t. This represents the penalty factor for not satisfying the constraints; This represents the total heat dissipation of the cooling tower at time t. Let t be the cooling load of the central air conditioning system. Let t be the cooling load demand. , , , , , These are, respectively, the chilled water return temperature, chilled water outlet temperature, cooling water outlet temperature, cooling water return temperature, cooling water return approximation temperature, and wet-bulb temperature at time t; By adding a penalty factor to the objective function given by the above formula, and considering the running time of the equipment when it has been started and the running time when it has been stopped, corresponding constraints are imposed to ensure the stable operation of the central air conditioning system and that the cooling capacity meets the demand.

[0019] Thirdly, this application provides a terminal, which includes a touch screen, a memory, a transceiver, one or more processors, multiple applications, and one or more programs; wherein the one or more programs are stored in the memory, and the transceiver is used to transmit or receive wireless signals; when the one or more processors execute the one or more programs, the terminal enables the terminal to implement the energy-saving control method of the central air conditioning system as described above.

[0020] Fourthly, this application provides a computer-readable storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor of an electronic device, cause the processor to perform the energy-saving control method for a central air conditioning system as described above.

[0021] In the energy-saving control scheme for the central air conditioning system provided in this application embodiment, a cooling load prediction model is constructed by using a convolutional neural network to extract cooling load influencing factors and a long short-term memory network to capture long-term dependencies in historical cooling load data. This model can more accurately predict the hourly cooling load of the next day, providing more precise guidance for the operation of the central air conditioning system. Through a deep deterministic strategy gradient algorithm, cooling capacity is allocated within a fixed time period under the condition of satisfying the hourly cooling load of the next day and minimizing electricity costs, thereby minimizing electricity costs. An energy efficiency model and a functional model of the central air conditioning system are constructed. With the goal of minimizing electricity costs, a combination of fuzzy logic control and model predictive control is adopted. For different load conditions under different wet-bulb temperatures, the operating conditions of the chiller, chilled water pump, cooling water pump, and cooling tower are matched. Through simulation combinations and energy efficiency comparisons, the optimal combination of operating modes is determined, which can improve the energy efficiency of the central air conditioning system and reduce overall energy consumption.

[0022] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0024] Figure 1 This is a schematic flowchart of an energy-saving control method for a central air conditioning system provided in an embodiment of this application; Figure 2 yes Figure 1 A flowchart illustrating step S100; Figure 3 This is a schematic diagram of the structure of an energy-saving control device for a central air conditioning system provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of an energy-saving control device for a central air conditioning system provided in an embodiment of this application; Figure 5 A schematic diagram of the hardware structure of an energy-saving control device for a central air conditioning system provided in an embodiment of this application. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0026] In the specification, claims, and accompanying drawings of this application, the use of terms such as "first" and "second" is for descriptive purposes only, to distinguish different objects, and not to describe a specific order, nor should it be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated.

[0027] The terms “comprising” and “having”, and any variations thereof, used in this document are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps, units, or modules is not limited to the steps, units, or modules listed, but may optionally include steps, units, or modules not listed, or may optionally include other steps, units, or modules inherent to such process, method, product, or device.

[0028] The use of "and / or" or "and / or" in the text implies three parallel options. For example, "A and / or B" could mean option A, option B, or a combination of both A and B.

[0029] The term "embodiment" in this document means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0030] The central air conditioning system consists of a cooling tower, cooling water pump, chilled water pump, central air conditioning chiller, air handling unit, fan coil unit, water collector, and water distributor, which output water to different air outlets through ductwork. Its specific functions are as follows: 1) The main function of a cooling tower is to dissipate heat. The cooling tower removes heat from the condenser through water circulation, and the heat is dissipated into the atmosphere through heat exchange between the air and water.

[0031] 2) The task of the cooling water pump is to drive the cooling water to circulate in the cooling system. It connects the condenser and the cooling tower, ensuring that the cooling water can flow and remove heat from the condenser.

[0032] 3) The chilled water pump is responsible for circulating chilled water, which is the medium used to transport the cooling capacity of the refrigerant in the central air conditioning system. In the refrigeration unit, the refrigerant absorbs heat in the evaporator and becomes low-temperature, low-pressure vapor, then flows to the condenser to release heat and condense into a liquid.

[0033] 4) Central air conditioning chillers usually refer to refrigeration units, which are the core of the entire central air conditioning system and are responsible for generating cooling or heating. Refrigeration units contain key components such as compressors, condensers, evaporators, and expansion valves, and achieve cooling or heating through the circulation of refrigerant.

[0034] 5) An air handling unit is an important component of a central air conditioning system. It is responsible for processing and regulating the amount of air supplied to the room.

[0035] 6) A fan coil unit is a terminal air handling unit that is usually installed on the ceiling or wall indoors.

[0036] 7) In central air conditioning systems, water collectors are mainly used to collect and distribute condensate.

[0037] 8) A water distributor is a pipe distribution device in a central air conditioning system. Its function is to distribute chilled water or hot water from the central air conditioning chiller to each air handling unit or fan coil unit.

[0038] Firstly, please refer to Figure 1 , Figure 1 This is a schematic flowchart of an energy-saving control method for a central air conditioning system provided in an embodiment of this application.

[0039] S100. By using convolutional neural networks to extract factors affecting cooling load and long short-term memory networks to capture long-term dependencies in historical cooling load data, a cooling load prediction model for the central air conditioning system is constructed, and the hourly cooling load for the next day is predicted.

[0040] In this embodiment, the convolutional neural network can effectively extract the features of cooling load influencing factors and capture the degree of influence of different factors on the cooling load. The long short-term memory network can fully utilize the long-term dependencies in historical cooling load data to improve the accuracy of future cooling load predictions. The cooling load prediction model constructed by combining these two technologies can comprehensively consider multiple factors and historical data, thereby significantly improving the accuracy of cooling load prediction. For example, in practical applications, by accurately extracting influencing factors such as outdoor temperature, humidity, and solar radiation, and capturing long-term dependencies such as seasonal variations and differences between weekdays and non-weekdays in historical cooling load data, the network model can more accurately predict the cooling load at different times of the following day, reducing prediction errors.

[0041] Accurate cooling load forecasting provides crucial information for optimizing the operation strategy of central air conditioning systems. Based on the predicted hourly cooling load for the following day, parameters such as start-up and shutdown times and operating power of the central air conditioning system can be rationally arranged to avoid over-cooling or under-cooling. For example, during periods of low cooling load, the operating power of the central air conditioning system can be appropriately reduced or some equipment can be shut down to save energy. During periods of high cooling load, the equipment can be started earlier to ensure that the indoor temperature remains within a comfortable range. By optimizing the operation strategy, the energy efficiency ratio of the central air conditioning system can be improved, and energy consumption can be reduced.

[0042] Please see Figure 2 In one possible implementation, the method of constructing a cooling load prediction model for a central air conditioning system by using a convolutional neural network to extract cooling load influencing factors and a long short-term memory network to capture long-term dependencies in historical cooling load data includes: S110. Obtain data related to the cooling load of the central air conditioning system, including historical cooling load data, outdoor meteorological parameters, and indoor environmental parameters.

[0043] S120. Preprocess the acquired data, including cleaning and normalization, to make it conform to the input requirements of the convolutional neural network-long short-term memory network model.

[0044] S130. Determine the parameters of the convolutional neural network-long short-term memory network model based on the characteristics of the acquired data, so as to input the features extracted by the convolutional neural network into the long short-term memory network to achieve feature fusion.

[0045] Among them, the convolutional kernel of the convolutional neural network can extract local features and spatial information about the cold load influencing factors in the input data, while the long short-term memory network can process long-term dependencies in the historical cold load data of the time series.

[0046] Specifically, convolutional neural networks excel at processing data with spatial structure. In central air conditioning systems, factors influencing cooling load may include indoor and outdoor temperatures, humidity, occupancy density, and equipment operating status. These factors can be viewed as data with certain spatial characteristics; for example, temperature and humidity may vary in different areas. By performing local perception and weight sharing on the input data, convolutional kernels can automatically extract key features from these influencing factors, thereby reducing the model's reliance on manual feature engineering.

[0047] The cooling load of a central air conditioning system exhibits time-series characteristics, and historical cooling load data is of significant reference value for predicting future cooling loads. Long Short-Term Memory (LSTM) networks are neural network structures specifically designed for processing time-series data. Through its unique gating mechanism, LSTM can effectively capture long-term dependencies in historical data. It can remember long-term information while simultaneously updating and forgetting based on current input, thus better adapting to the dynamic changes in cooling load data.

[0048] S140. The convolutional neural network-long short-term memory network model is trained using the preprocessed data, and the parameters of the convolutional neural network-long short-term memory network model are optimized by backpropagation algorithm and optimization algorithm to minimize the error between the predicted value and the actual value.

[0049] A well-developed cooling load forecasting model can be applied to the operation optimization and energy-saving control of central air conditioning systems. By accurately predicting future cooling loads, the operation of refrigeration equipment can be rationally scheduled, improving energy efficiency and reducing operating costs.

[0050] S200: Under the condition of meeting the hourly cooling load of the next day and minimizing electricity costs, the cooling capacity within a fixed time period is allocated through a deep deterministic strategy gradient algorithm.

[0051] Specifically, a deep deterministic strategy gradient algorithm dynamically adjusts the allocation strategy of cooling capacity based on real-time electricity price information and cooling load changes to minimize electricity costs. This allocation strategy includes introducing ice storage or water storage systems to store cooling capacity or electrical energy during periods of low electricity prices and release it during periods of high electricity prices.

[0052] Understandably, the operation of central air conditioning systems is affected by various factors, such as outdoor temperature, number of people indoors, and equipment performance, all of which are dynamic and uncertain. The cooling load for the following day is difficult to predict accurately, and electricity costs are also affected by fluctuations in electricity market prices. Cooling capacity allocation is a continuous action space problem, meaning the cooling capacity can vary continuously within a certain range. Deep deterministic policy gradient algorithms are suitable for decision-making problems with continuous action spaces, directly outputting continuous cooling capacity values ​​without discretizing the action space, thus improving the accuracy and flexibility of decision-making.

[0053] In this embodiment, based on cooling load forecasting and combined with time-of-use electricity pricing information, a sequential least squares programming algorithm is used to allocate cooling capacity within a fixed time period, reducing electricity costs while meeting load demand. For the central air conditioning system, a data-driven approach is used to establish equipment energy efficiency and functional models. With the goal of minimizing system power consumption and electricity costs, optimization algorithms, such as genetic algorithms, particle swarm optimization algorithms, annealing algorithms, or ant colony optimization algorithms, are used to optimize system control parameters for a given cooling capacity.

[0054] S300. Construct an energy efficiency model and a functional model for the central air conditioning system, and optimize the operation strategy of the central air conditioning system corresponding to the given cooling capacity with the goal of minimizing electricity costs, so as to minimize the overall energy consumption of the central air conditioning system.

[0055] The operating strategy employs a combination of fuzzy logic control and model predictive control. Under different load conditions and different wet-bulb temperatures, the optimal combination of operating modes is determined by matching the operating conditions of the chiller, chilled water pump, cooling water pump, and cooling tower through simulation and energy efficiency comparison.

[0056] In one possible implementation, the central air conditioning system comprises a chiller, chilled water pumps, cooling water pumps, a cooling tower, air handling units, fan coil units, a water collector, and a water distributor, which output water to different air outlets via ductwork; wherein, The input variables for the energy efficiency model of the chiller are cooling load, chilled water outlet temperature, cooling water return temperature, chilled water flow rate, and cooling water flow rate, and the output variables are chiller power and number of units. The input variable for the energy efficiency model of chilled water pumps is chilled water flow rate, and the output variables are chilled water pump frequency / power and number of pumps. The input variable for the energy efficiency model of the cooling water pump is the cooling water flow rate, and the output power is the cooling water pump frequency / power and the number of pumps. The input variables for the energy efficiency model of a cooling tower are the heat dissipation of the cooling tower and the outdoor temperature and humidity, and the output variables are the frequency / power and number of cooling towers.

[0057] In one possible implementation, the overall energy consumption of the central air conditioning system is determined by the following formula:

[0058] in, Let t be the total energy consumption of the central air conditioning system. This represents the real-time power of the chiller unit at time t. This represents the real-time power of the chilled water pump at time t. This represents the real-time power of the cooling water pump at time t. This represents the real-time power of the cooling tower at time t. This represents the penalty factor for not satisfying the constraints; This represents the total heat dissipation of the cooling tower at time t. Let t be the cooling load of the central air conditioning system. Let t be the cooling load demand. , , , , , These are, respectively, the chilled water return temperature, chilled water outlet temperature, cooling water outlet temperature, cooling water return temperature, cooling water return approximation temperature, and wet-bulb temperature at time t; By adding a penalty factor to the objective function given by the above formula, and considering the running time of the equipment when it has been started and the running time when it has been stopped, corresponding constraints are imposed to ensure the stable operation of the central air conditioning system and that the cooling capacity meets the demand.

[0059] In this embodiment, by optimizing several controllable parameters of the central air conditioning system, such as the setpoints for chilled water supply temperature, cooling water return temperature, operating frequencies of each chilled water pump, operating frequencies of each cooling water pump, and operating frequencies of each cooling tower, the total energy consumption of the system is minimized. This is transformed into an optimization problem of the objective function under the aforementioned multiple constraints.

[0060] In this embodiment, a data-driven energy efficiency assessment method is introduced into the energy efficiency model. Machine learning algorithms are used to analyze a large amount of operating data to predict the energy efficiency performance of the central air conditioning system under different operating conditions. In the functional model, fuzzy logic control is used to perform fuzzification processing and full-condition modeling of the chiller, chilled water pump, cooling water pump, and cooling tower based on actual operating and predicted data under different load conditions at different wet-bulb temperatures. This forms a preliminary operating strategy. Through model predictive control, combined with the predicted cooling load and electricity price changes over a future period, the central air conditioning system is optimized and controlled. The optimal combination of operating modes is determined through simulation combinations and energy efficiency comparisons.

[0061] In the above embodiments, this application improves the cooling load prediction results by comprehensively considering various cooling load influencing factors and effectively capturing the long-term dependencies in historical cooling load data. With the goal of minimizing the overall energy consumption and electricity cost of the central air conditioning system, the application optimizes the operation strategy through simulation combination and energy efficiency comparison to meet the target environment's demand for a high-efficiency and energy-saving central air conditioning system.

[0062] The methods of the embodiments of this application have been described in detail above, and the apparatus of the embodiments of this application is provided below.

[0063] Secondly, please refer to Figure 3 , Figure 3This is a schematic diagram of the structure of an energy-saving control device for a central air conditioning system provided in an embodiment of this application.

[0064] The prediction module 100 is used to construct a cooling load prediction model for the central air conditioning system by using a convolutional neural network to extract cooling load influencing factors and a long short-term memory network to capture long-term dependencies in historical cooling load data, and to predict the hourly cooling load for the next day.

[0065] The first optimization module 200 is used to allocate cooling capacity within a fixed time period through a deep deterministic strategy gradient algorithm, under the condition of meeting the hourly cooling load of the next day and minimizing electricity costs.

[0066] The second optimization module 300 is used to construct the energy efficiency model and functional model of the central air conditioning system, and optimize the operation strategy of the central air conditioning system corresponding to a given cooling capacity with the goal of minimizing electricity costs, so as to minimize the overall energy consumption of the central air conditioning system. The operation strategy employs a combination of fuzzy logic control and model predictive control, matching the operating conditions of the chiller, chilled water pump, cooling water pump, and cooling tower under different load conditions at different wet-bulb temperatures, and determining the optimal combination of operating modes through simulation and energy efficiency comparison.

[0067] Please see Figure 4 In one possible implementation, the prediction module 100 includes: The acquisition module 110 is used to acquire data related to the cooling load of the central air conditioning system, including historical cooling load data, outdoor meteorological parameters, and indoor environmental parameters. The preprocessing module 120 is used to preprocess the acquired data, including cleaning and normalization, so that it meets the input requirements of the convolutional neural network-long short-term memory network model. The construction module 130 is used to determine the parameters of the convolutional neural network-long short-term memory network model based on the characteristics of the acquired data, so as to input the features extracted by the convolutional neural network into the long short-term memory network to achieve feature fusion; wherein, the convolutional kernel of the convolutional neural network can extract local features and spatial information about the cold load influencing factors in the input data, and the long short-term memory network can process the long-term dependencies in the historical cold load data of the time series. The training module 140 is used to train the convolutional neural network-long short-term memory network model using preprocessed data, and to optimize the parameters of the convolutional neural network-long short-term memory network model through backpropagation and optimization algorithms to minimize the error between the predicted value and the actual value.

[0068] In one possible implementation, the first optimization module 200 is used to dynamically adjust the allocation strategy of cooling capacity based on real-time electricity price information and cooling load changes using a deep deterministic strategy gradient algorithm, so as to achieve the goal of minimizing electricity costs; wherein, the allocation strategy includes introducing an ice storage system or a battery energy storage system to store cooling capacity or electrical energy during periods of low electricity prices and release it during periods of high electricity prices.

[0069] In one possible implementation, the second optimization module 300 is used to introduce a data-driven energy efficiency assessment method into the energy efficiency model, using machine learning algorithms to analyze a large amount of operating data and predict the energy efficiency performance of the central air conditioning system under different operating conditions; in the functional model, fuzzy logic control is used to perform fuzzification processing and full-condition modeling of the chiller, chilled water pump, cooling water pump, and cooling tower based on actual operating and predicted data under different load conditions under different wet-bulb temperatures, forming a preliminary operating strategy; through model predictive control, combined with the predicted cooling load and electricity price changes over a period of time, the central air conditioning system is optimized and controlled; and through simulation combination and energy efficiency comparison, the optimal operating mode combination is determined.

[0070] In one possible implementation, the central air conditioning system comprises a chiller, chilled water pumps, cooling water pumps, a cooling tower, air handling units, fan coil units, a water collector, and a water distributor, which output water to different air outlets via ductwork; wherein, The input variables for the energy efficiency model of the chiller are cooling load, chilled water outlet temperature, cooling water return temperature, chilled water flow rate, and cooling water flow rate, and the output variables are chiller power and number of units. The input variable for the energy efficiency model of chilled water pumps is chilled water flow rate, and the output variables are chilled water pump frequency / power and number of pumps. The input variable for the energy efficiency model of the cooling water pump is the cooling water flow rate, and the output power is the cooling water pump frequency / power and the number of pumps. The input variables for the energy efficiency model of a cooling tower are the heat dissipation of the cooling tower and the outdoor temperature and humidity, and the output variables are the frequency / power and number of cooling towers.

[0071] In one possible implementation, the overall energy consumption of the central air conditioning system is determined by the following formula:

[0072] in, Let t be the total energy consumption of the central air conditioning system. This represents the real-time power of the chiller unit at time t. This represents the real-time power of the chilled water pump at time t. This represents the real-time power of the cooling water pump at time t. This represents the real-time power of the cooling tower at time t. This represents the penalty factor for not satisfying the constraints; This represents the total heat dissipation of the cooling tower at time t. Let t be the cooling load of the central air conditioning system. Let t be the cooling load demand. , , , , , These are, respectively, the chilled water return temperature, chilled water outlet temperature, cooling water outlet temperature, cooling water return temperature, cooling water return approximation temperature, and wet-bulb temperature at time t; By adding a penalty factor to the objective function given by the above formula, and considering the running time of the equipment when it has been started and the running time when it has been stopped, corresponding constraints are imposed to ensure the stable operation of the central air conditioning system and that the cooling capacity meets the demand.

[0073] It should be noted that in some embodiments, the functions or modules of the device provided in this application embodiment can be used to execute the methods described in the above method embodiments. The specific implementation can be referred to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0074] Please see Figure 5 , Figure 5 A schematic diagram of the hardware structure of an energy-saving control device for a central air conditioning system provided in an embodiment of this application.

[0075] The device 2 includes a processor 21, a memory 24, an input device 22, and an output device 23. The processor 21, memory 24, input device 22, and output device 23 are coupled together via connectors, which may include various interfaces, transmission lines, or buses, etc., and are not limited in this application embodiment. It should be understood that in the various embodiments of this application, coupling refers to mutual connection through a specific method, including direct connection or indirect connection through other devices, such as through various interfaces, transmission lines, buses, etc.

[0076] Processor 21 can be one or more graphics processing units (GPUs). If processor 21 is a GPU, the GPU can be a single-core GPU or a multi-core GPU. Optionally, processor 21 can be a processor group composed of multiple GPUs, with the multiple processors coupled to each other via one or more buses. Optionally, the processor can also be other types of processors, etc., which are not limited in this embodiment.

[0077] The memory 24 can be used to store computer program instructions, as well as various types of computer program code, including program code for executing the scheme of this application. Optionally, the memory includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or compact disc read-only memory (CD-ROM), which is used for related instructions and data.

[0078] Input device 22 is used to input data and / or signals, and output device 23 is used to output data and / or signals. Output device 23 and input device 22 can be independent devices or an integrated device.

[0079] It is understood that in this embodiment of the application, the memory 24 can be used not only to store related instructions, but also the specific data stored in the memory is not limited.

[0080] Understandable, Figure 5 This illustration only shows a simplified design of an energy-saving control device for a central air conditioning system. In practical applications, the device may also include other necessary components, including, but not limited to, any number of input / output devices, processors, memories, etc., and all devices that can implement the embodiments of this application are within the protection scope of this application.

[0081] Thirdly, this application provides a terminal, which includes a touch screen, a memory, a transceiver, one or more processors, multiple applications, and one or more programs; wherein the one or more programs are stored in the memory, and the transceiver is used to transmit or receive wireless signals; when the one or more processors execute the one or more programs, the terminal enables the terminal to implement the energy-saving control method of the central air conditioning system as described above.

[0082] Fourthly, this application provides a computer-readable storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor of an electronic device, cause the processor to perform the energy-saving control method for a central air conditioning system as described above.

[0083] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0084] Those skilled in the art will readily understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. Those skilled in the art will also readily understand that the various embodiments of this application have different focuses, and for the sake of convenience and brevity, the same or similar parts may not be repeated in different embodiments. Therefore, parts not described or not described in detail in one embodiment can be referred to the descriptions in other embodiments.

[0085] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access, or a data storage device such as a server or data center that integrates one or more available media. The available medium can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., digital versatile discs (DVDs)), or semiconductor media (e.g., solid-state disks (SSDs)).

[0086] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. An energy-saving control method for a central air-conditioning system, characterized in that: The method comprises at least the following steps: The cooling load prediction model of the central air-conditioning system is constructed by using convolutional neural network to extract cooling load influencing factors and long short-term memory network to capture the long-term dependencies in historical cooling load data, and the cooling load of the next day is predicted hour by hour. Under the condition of meeting the hourly cooling load of the next day and minimizing the electricity cost, the cooling capacity within a fixed time period is allocated through a deep deterministic policy gradient algorithm; An energy efficiency model and a functional model of the central air-conditioning system are constructed, and with the goal of minimizing electricity costs, the operation strategy of the central air-conditioning system corresponding to the given cooling capacity is optimized to minimize the comprehensive energy consumption of the central air-conditioning system; wherein, the operation strategy adopts a combination of fuzzy logic control and model predictive control, matches the operating conditions of the chiller, chilled water pump, cooling water pump, and cooling tower for different load conditions under different wet-bulb temperatures, and determines the optimal operation mode combination through simulation combination and energy efficiency comparison.

2. The energy-saving control method of the central air-conditioning system according to claim 1, characterized in that: The method of using a convolutional neural network to extract cooling load influencing factors and a long short-term memory network to capture long-term dependencies in historical cooling load data to construct a cooling load prediction model for a central air-conditioning system includes: Obtain data related to the cooling load of the central air-conditioning system, including historical cooling load data, outdoor meteorological parameters, and indoor environmental parameters; Preprocess the acquired data, including cleaning and normalization, to make it meet the input requirements of the convolutional neural network-long short-term memory network model; The parameters of the convolutional neural network-long short-term memory network model are determined according to the characteristics of the acquired data, so that the features partially extracted by the convolutional neural network are input into the long short-term memory network to achieve feature fusion; the convolution kernel of the convolutional neural network can extract the local features and spatial information of the cooling load influencing factors in the input data, and the long short-term memory network can process the long-term dependencies in the historical cooling load data of the time series; The convolutional neural network-long short-term memory network model is trained using the preprocessed data, and the parameters of the convolutional neural network-long short-term memory network model are optimized through the back propagation algorithm and the optimization algorithm to minimize the error between the predicted value and the actual value.

3. The energy-saving control method of the central air-conditioning system according to claim 1, characterized in that: The method of allocating cooling capacity within a fixed time period by using a deep deterministic policy gradient algorithm under the condition of satisfying the hourly cooling load of the next day and minimizing the electricity cost includes: Through the deep deterministic policy gradient algorithm, the cooling supply allocation strategy is dynamically adjusted according to the real-time electricity price information and cooling load changes to achieve the goal of minimizing electricity costs; wherein, the allocation strategy includes the introduction of an ice storage system or a water storage system to store cooling capacity during periods of low electricity prices and release it during periods of peak electricity prices.

4. The energy-saving control method of the central air-conditioning system according to claim 1, characterized in that: The energy efficiency model and function model of the central air-conditioning system are constructed, and the operation strategy of the central air-conditioning system corresponding to the given cooling capacity is optimized with the goal of minimizing the electricity cost, including: In the energy efficiency model, a data-driven energy efficiency evaluation method is introduced, and the operation data is analyzed using a machine learning algorithm to predict the energy efficiency performance of the central air-conditioning system under different operating conditions. In the functional model, fuzzy logic control is used according to different load conditions under different wet-bulb temperatures. Based on actual operation and test data, fuzzy processing and full-condition modeling are performed on the chiller, chilled water pump, cooling water pump, and cooling tower to form a preliminary operation strategy. Through model predictive control, combined with the cold load forecast and electricity price changes in the future, the central air-conditioning system is optimized and controlled, and the optimal operation mode combination is determined through simulation combination and energy efficiency comparison.

5. The energy-saving control method of the central air-conditioning system according to claim 1 or 4, characterized in that: The central air conditioning system is composed of a chiller, a chilled water pump, a cooling water pump, a cooling tower, an air handling unit, a fan coil, a water collector and a water distributor, which are output to different air outlets through air ducts; wherein, The input variables of the chiller energy efficiency model are cooling load, chilled water outlet temperature, cooling water return temperature, chilled water flow rate and cooling water flow rate, and the output variables are chiller power and number; The input variable of the energy efficiency model of the chilled water pump is the chilled water flow rate, and the output variable is the frequency / power and number of chilled water pumps; The input variable of the energy efficiency model of the cooling water pump is the cooling water flow rate, and the output power is the cooling water pump frequency / power and number of units; The input variables of the cooling tower energy efficiency model are the cooling tower heat dissipation and outdoor temperature and humidity, and the output variables are the cooling tower frequency / power and number of units.

6. The energy-saving control method of the central air-conditioning system according to claim 1, characterized in that: The comprehensive energy consumption of the central air-conditioning system is determined by the following formula: in, is the comprehensive energy consumption of the central air-conditioning system at time t, represents the real-time power of the chiller at time t, represents the real-time power of the chilled water pump at time t, represents the real-time power of the cooling water pump at time t, represents the real-time power of the cooling tower at time t, represents the penalty factor and for not meeting the constraints; represents the total heat dissipation of the cooling tower at time t, is the cooling load of the central air-conditioning system at time t, is the cooling load demand at time t; , , , , , They are the chilled water return temperature, chilled water outlet temperature, cooling water outlet temperature, cooling water return temperature, cooling water return approach temperature and wet bulb temperature at time t respectively; The time that the equipment has been started and stopped is the time that the equipment has been running. The penalty factor is added to the objective function given by the above formula to impose corresponding constraints to ensure the stable operation of the central air-conditioning system and the cooling capacity meets the demand.

7. An energy-saving control device for a central air-conditioning system, characterized in that: The device at least comprises: The prediction module is used to construct a cooling load prediction model for the central air-conditioning system by using a convolutional neural network to extract cooling load influencing factors and a long short-term memory network to capture long-term dependencies in historical cooling load data, and to predict the hourly cooling load for the next day; The first optimization module is used to allocate the cooling capacity within a fixed time period through a deep deterministic policy gradient algorithm under the condition of meeting the hourly cooling load of the next day and minimizing the electricity cost; The second optimization module is used to construct an energy efficiency model and a functional model of the central air-conditioning system, and optimizes the operation strategy of the central air-conditioning system corresponding to the given cooling capacity with the goal of minimizing electricity costs, so as to minimize the comprehensive energy consumption of the central air-conditioning system; wherein, the operation strategy adopts a combination of fuzzy logic control and model predictive control, matches the operating conditions of the chiller, chilled water pump, cooling water pump, and cooling tower for different load conditions under different wet-bulb temperatures, and determines the optimal operation mode combination through simulation combination and energy efficiency comparison.

8. The energy-saving control device for a central air-conditioning system according to claim 7, characterized in that: The prediction module comprises: An acquisition module is used to acquire data related to the cooling load of the central air-conditioning system, including historical cooling load data, outdoor meteorological parameters, and indoor environmental parameters; The preprocessing module is used to preprocess the acquired data, including cleaning and normalization, so that it meets the input requirements of the convolutional neural network-long short-term memory network model; A construction module is used to determine the parameters of the convolutional neural network-long short-term memory network model according to the characteristics of the acquired data, so as to input the features partially extracted by the convolutional neural network into the long short-term memory network to achieve feature fusion; wherein the convolution kernel of the convolutional neural network can extract the local features and spatial information of the cooling load influencing factors in the input data, and the long short-term memory network can process the long-term dependency in the historical cooling load data of the time series; The training module is used to train the convolutional neural network-long short-term memory network model using the preprocessed data, and optimize the parameters of the convolutional neural network-long short-term memory network model through the back propagation algorithm and the optimization algorithm to minimize the error between the predicted value and the actual value.

9. The energy-saving control device for a central air-conditioning system according to claim 7, characterized in that: The second optimization module is used to introduce a data-driven energy efficiency evaluation method into the energy efficiency model, analyze a large amount of operating data using a machine learning algorithm, and predict the energy efficiency performance of the central air-conditioning system under different operating conditions; In the functional model, fuzzy logic control is used to perform fuzzy processing and full-condition modeling of the chiller, chilled water pump, cooling water pump, and cooling tower according to different load conditions under different wet-bulb temperatures based on actual operation and prediction data to form a preliminary operation strategy. Through model predictive control, combined with the cold load forecast and electricity price changes for a period of time in the future, the central air-conditioning system is optimized and controlled, and the optimal operation mode combination is determined through simulation combination and energy efficiency comparison.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, which includes program instructions. When the program instructions are executed by a processor of an electronic device, the processor executes the energy-saving control method for a central air-conditioning system as described in any one of claims 1-6.

Images