An optimized control method for ice storage central air conditioning based on load prediction and optimized allocation

Abstract

The present invention discloses an optimized control method for a chilled water central air conditioning system based on load prediction and optimized allocation, which relates to the technical field of load control. The method includes obtaining the operation data of the central air conditioning system for preprocessing, and predicting the load of the central air conditioning system through an intelligent algorithm; forming an optimized load scheme for the central air conditioning system according to the predicted load, and generating an optimized operation state scheme using a genetic algorithm; regulating the central air conditioning system according to the optimized load scheme and the optimized operation state scheme; displaying the operation state and load data of the central air conditioning system in real time, and recording and storing the load data in a database. By collecting and preprocessing the operation data of the central air conditioning system, predicting the load of the central air conditioning system, extracting the load of the air conditioning terminal, forming an optimized load scheme according to the operation data of the central air conditioning system, and combining with the optimized operation state scheme to regulate the operation of the central air conditioning system, the present invention effectively improves the accuracy of the operation control of the central air conditioning system, optimizes the energy consumption of the central air conditioning system, and improves the load utilization efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of load control, and in particular to an optimized control method for a chilled water central air conditioning system based on load prediction and optimized allocation. Background Art

[0002] With the popularization of the concepts of intelligent buildings and smart homes, the energy efficiency and operation management of central air conditioning systems have become core issues in building energy management. In recent years, the control strategies of central air conditioning systems have gradually developed towards a more intelligent and energy-saving direction from traditional time-based control and simple temperature control modes. Early air conditioning control systems mainly relied on preset operation schedules and manual adjustment, lacking the ability to sense real-time load changes and make dynamic adjustments. With the development of sensing technology, data acquisition, and processing technology, central air conditioning load prediction and intelligent scheduling schemes based on real-time data have gradually become a research hotspot. For example, many studies have proposed using input information such as environmental sensors and equipment operation status monitoring data, combined with machine learning algorithms to predict air conditioning loads, and using optimization algorithms to perform load allocation and operation mode adjustment. These methods can dynamically adjust the operation status of the air conditioning system according to real-time data, thereby improving the operation efficiency and energy-saving effect of the system. However, there are still defects in the existing technologies. Existing algorithms have problems of high computational complexity and slow convergence speed when dealing with complex systems. Especially in multi-device and large-scale air conditioning systems, traditional optimization methods often struggle to meet real-time requirements and lack adaptability and accuracy in complex environments. Summary of the Invention

[0003] In view of the problems existing in the above-mentioned existing optimized control methods for chilled water central air conditioning systems based on load prediction and optimized allocation, the present invention is proposed.

[0004] Therefore, the problem to be solved by the present invention is that the existing technologies are difficult to meet real-time requirements and lack adaptability and accuracy in complex environments.

[0005] To solve the above technical problems, the present invention provides the following technical solution: An optimized control method for a chilled water central air conditioning system based on load prediction and optimized allocation, which includes obtaining central air conditioning operation data for preprocessing, and predicting the central air conditioning load through an intelligent algorithm; forming an optimized central air conditioning load plan according to the predicted load, and generating an optimized operation status plan using a genetic algorithm; regulating the central air conditioning according to the load optimized plan and the operation status optimized plan; displaying the operation status and load data of the central air conditioning in real time, and recording and storing the load data in a database.

[0006] As a preferred embodiment of the optimized control method for the chilled central air-conditioning based on load prediction and optimized allocation of the present invention, wherein: the acquisition of the operating data of the central air-conditioning for preprocessing refers to collecting the operating data of the central air-conditioning through installed sensors, wirelessly connecting the sensors to form a sensor network and connecting the sensor network to the data center. The data center cleans the operating data of the central air-conditioning collected by the sensor network and performs denoising processing using Kalman filtering, and normalizes the denoised operating data of the central air-conditioning.

[0007] As a preferred embodiment of the optimized control method for the chilled central air-conditioning based on load prediction and optimized allocation of the present invention, wherein: the prediction of the load through intelligent learning refers to constructing a prediction model through an LSTM model, including an input gate, a forget gate, and an output gate. The input of the prediction model is set as the operating data of the central air-conditioning, and the output is the predicted load of the central air-conditioning. The prediction model is trained using a training data set, and the mean square error is set as the loss function. The gradient of the loss function is calculated through backpropagation, and the parameters of the prediction model are updated using an optimization algorithm until the loss function is minimized. The parameters of the trained prediction model are used to optimize the prediction model;

[0008] The operating parameters of the central air-conditioning are input into the optimized prediction model to obtain the future load of the central air-conditioning, and a reinforcement learning mechanism is introduced, and the Q-learning algorithm is used to continuously optimize the prediction model.

[0009] As a preferred embodiment of the optimized control method for the chilled central air-conditioning based on load prediction and optimized allocation of the present invention, wherein: the formation of the optimized load scheme for the central air-conditioning according to the predicted load includes,

[0010] Each air-conditioning terminal in the central air-conditioning is defined as an independent agent i, and the load of each air-conditioning terminal is extracted from the future load of the central air-conditioning , and the operating time of each air-conditioning terminal is obtained according to the operating data of the central air-conditioning, and the ratio of the operating time of each air-conditioning terminal in a day is calculated as the working time factor of the agent i ;

[0011] Query the energy efficiency curve of the air-conditioning terminal to obtain the energy efficiency ratio of the air-conditioning terminal under the load ; ;

[0012] Statistical data on the load change of the air-conditioning terminal during the operating time, and fitting the load change curve through regression analysis. The absolute value of the slope of the load change curve of the air-conditioning terminal at the load is obtained as the load change factor ;

[0013] Comprehensively correct the working time factor through a non-standard interaction term 、Coefficient of Performance and load change factor :

[0014]

[0015] Among them is the non-standard interaction correction term of agent i, and are the coupling parameters of the non-standard interaction correction term;

[0016] Pair the agents in pairs. According to the running time of each air-conditioning terminal in a day, record the synchronous running duration of each pair of agents in a day, and calculate the ratio of the synchronous running duration to the maximum synchronous running duration and the ratio of the synchronous running duration to the running duration of agent i , and calculate the sum of all agent pairs containing agent i as the cooperation coefficient of agent i , and take the sum of all agent pairs containing agent i as the running coefficient of agent i ;

[0017] Define the time correlation term and running coefficient of agent i: :

[0018]

[0019] Calculate the cooperation strength between agent i and other agents according to the time correlation term : :

[0020]

[0021] Among them is the number of agents;

[0022] Define the total utility function U of the central air-conditioning through the cooperation strength as:

[0023]

[0024] Among them is the load weight of agent i, is the weight of the non-standard interaction correction term of agent i;

[0025] Use the gradient descent method to optimize and update the load of each agent i, and calculate through the updated loads of agent i and agent j :

[0026]

[0027] where is the ratio of the synchronized running duration of the updated agents i and j to the running duration of agent i, and are the loads of the updated agents i and j respectively;

[0028] Based on recalculate the cooperation coefficient and running coefficient of agent i, and recalculate the total utility function value of the central air conditioner according to the loads of the updated agents. Iteratively adjust the agent loads until the total utility function value of the central air conditioner reaches the maximum and then stop. Output the agent loads that maximize the total utility function value of the central air conditioner as the loads of each air conditioner terminal to form an optimized central air conditioner load scheme.

[0029] As a preferred embodiment of the optimized control method for a chilled water central air conditioner based on load prediction and optimized allocation according to the present invention, wherein: the use of the genetic algorithm to generate an optimized operation scheme means randomly initializing the operation states and operation times when the air conditioner terminals reach the set temperature in the central air conditioner operation data to generate an initial population. Each air conditioner terminal represents an individual in the initial population. Optimize the initial population through the genetic algorithm, calculate the fitness of the individuals in the initial population, select individuals for crossover and mutation operations to generate new individuals, calculate the fitness of the new individuals for iteration, and stop the iteration after the fitness converges. Use the operation states and operation times of the air conditioner terminals in the individual with the highest fitness in the iterated population as the optimized operation scheme.

[0030] As a preferred embodiment of the optimized control method for a chilled water central air conditioner based on load prediction and optimized allocation according to the present invention, wherein: regulating the central air conditioner according to the optimized load scheme and the optimized operation scheme means optimizing the control of the air conditioner terminals of the central air conditioner according to the generated optimized load scheme and the optimized operation scheme, and continuously monitoring the central air conditioner operation data to continuously generate an optimized load scheme and an optimized operation scheme for central air conditioner operation regulation.

[0031] As a preferred embodiment of the optimized control method for a chilled water central air conditioner based on load prediction and optimized allocation according to the present invention, wherein: the real-time display of the central air conditioner operation state and load data means that the data center updates and displays the monitored central air conditioner operation state and load data in real time, synchronously compares and displays the obtained optimized central air conditioner load scheme with the central air conditioner operation data, and forms a data record according to the time stamp.

[0032] As a preferred solution of the optimized control method for ice storage central air-conditioning based on load prediction and optimal allocation of the present invention, wherein: the step of recording and storing load data in a database means that while the data center displays the load data, it stores the load data and data records in the database, and the database regularly detects the integrity and security of the stored data and uploads it to the cloud for backup.

[0033] A computer device includes: a memory and a processor; the memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned optimized control method for ice storage central air-conditioning based on load prediction and optimal allocation are implemented.

[0034] A computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, the steps of the above-mentioned optimized control method for ice storage central air-conditioning based on load prediction and optimal allocation are implemented.

[0035] The beneficial effects of the present invention are as follows: By collecting and preprocessing the operation data of the central air-conditioning, predicting the load of the central air-conditioning, extracting the load of the air-conditioning terminal, forming a load optimization plan based on the operation data of the central air-conditioning, and combining the operation status optimization plan to control the operation of the central air-conditioning, the accuracy of the operation control of the central air-conditioning is effectively improved, the energy consumption of the central air-conditioning is optimized, and the load utilization efficiency is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0037] Figure 1 It is a schematic flow chart of an optimized control method for ice storage central air-conditioning based on load prediction and optimal allocation. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0038] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and understandable, the following will describe the specific embodiments of the present invention in detail with reference to the drawings of the specification.

[0039] Many specific details are set forth in the following description in order to fully understand the present invention, but the present invention may also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the connotation of the present invention, so the present invention is not limited by the specific embodiments disclosed below.

[0040] Second, the "one embodiment" or "embodiment" referred to herein means a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it an embodiment that is separate from or mutually exclusive of other embodiments selectively.

[0041] Embodiment 1

[0042] Referring to Figure 1 , which is the first embodiment of the present invention. This embodiment provides an optimized control method for a chilled water central air conditioning system based on load prediction and optimal allocation. The optimized control method for a chilled water central air conditioning system based on load prediction and optimal allocation includes

[0043] S1. Obtain the operation data of the central air conditioning system for preprocessing, and predict the load of the central air conditioning system through an intelligent algorithm;

[0044] Specifically, obtaining the operation data of the central air conditioning system for preprocessing means collecting the operation data of the central air conditioning system through installed sensors, including load power, startup status, temperature, etc. The central air conditioning system includes a central host and air conditioning terminals. The central host is responsible for the overall operation of the central air conditioning system, and the air conditioning terminals are used for independent heating or cooling. The sensors are wirelessly connected to form a sensor network, and the sensor network is connected to a data center. The data center cleans the operation data of the central air conditioning system collected by the sensor network and performs denoising processing using Kalman filtering, and standardizes the denoised operation data of the central air conditioning system.

[0045] The introduction of the sensor network greatly improves the efficiency and accuracy of data collection in the air conditioning system. Installing sensors and transmitting data to the data center through a wireless network not only eliminates the complexity of traditional wired connections but also realizes remote collection and processing of real-time data. After preprocessing and cleaning the operation data of the central air conditioning system, the system can accurately predict future load demands. This load prediction can not only improve the energy efficiency of the air conditioning system but also dynamically adjust the system according to the predicted load demands. Due to the combination of intelligent algorithms and real-time data monitoring, the air conditioning system can make adaptive adjustments according to changes in the external environment and equipment status. This flexibility enables the system to effectively cope with load fluctuations under different environmental conditions and ensures the stability and efficient operation of the air conditioning system under various working conditions.

[0046] Further, predicting the load through intelligent learning means constructing a prediction model through an LSTM model, including an input gate, a forget gate, and an output gate. Set the input of the prediction model as the operation data of the central air conditioner, and the output as the predicted load of the central air conditioner. Use the training data set to train the prediction model, and set the mean square error as the loss function. Calculate the gradient of the loss function through backpropagation, and use the optimization algorithm to update the parameters of the prediction model until the loss function is minimized. Use the trained prediction model parameters to optimize the prediction model;

[0047] Input the operation parameters of the central air conditioner into the optimized prediction model to obtain the future load of the central air conditioner, and introduce a reinforcement learning mechanism to continuously optimize the prediction model using the Q-learning algorithm.

[0048] LSTM can learn and remember the long-term dependencies in time series data through its unique gating mechanism, thus providing higher accuracy and robustness in the prediction of central air conditioner loads. By inputting the operation data of the central air conditioner into the LSTM model, the model can adaptively learn the load prediction rules during the training process and accurately predict the future load demand. Q-learning is a reinforcement learning method that allows the model to self-adjust and optimize according to historical prediction errors. When the LSTM model encounters load prediction errors in practical applications, Q-learning can adjust the learning strategy and parameters of the LSTM model according to the prediction error feedback, thereby gradually improving the prediction accuracy. Combining the LSTM model with the Q-learning algorithm forms a closed-loop optimization control system. By inputting the real-time operation data of the central air conditioner into the optimized prediction model, the dynamic prediction of future loads can be realized. This real-time prediction ability enables the air conditioning system to automatically adjust during operation to adapt to real-time load changes, avoiding the common energy efficiency waste problems in traditional air conditioning systems.

[0049] S2. Form an optimization plan for the central air conditioner load according to the predicted load, and use the genetic algorithm to generate an optimization plan for the operation state;

[0050] Specifically, forming an optimization plan for the central air conditioner load according to the predicted load includes:

[0051] Define each air conditioner terminal in the central air conditioner as an independent agent i, extract the load of each air conditioner terminal from the future load of the central air conditioner and obtain the operation time of each air conditioner terminal according to the operation data of the central air conditioner. Calculate the proportion of the operation time of each air conditioner terminal in a day as the working time factor of agent i The working time factor is used to describe the working proportion of each air conditioner terminal in a day and reflects the activity of the air conditioner terminal. By introducing this factor, the load distribution of air conditioner terminals with different working intensities can be made more refined;

[0052] Query the energy efficiency curve of the air - conditioning terminal to obtain the energy efficiency ratio of the air - conditioning terminal under the load . The energy efficiency ratio reflects the energy utilization efficiency of each air - conditioning terminal under a certain load. By querying the energy efficiency curve of each air - conditioning terminal, the energy efficiency data at different loads can be obtained in real - time, providing a basis for subsequent optimal allocation;

[0053] Statistical data on the load changes of the air - conditioning terminal during the operation time, and fitting the load change curve through regression analysis. Obtain the absolute value of the slope of the load change curve of the air - conditioning terminal at the load as the load change factor . The change in the load of the air - conditioning terminal is not only a static load value but is also closely related to the dynamic changes in the system operation. By quantifying the load change, the response of the system to load fluctuations can be better simulated, optimizing the load scheduling;

[0054] Comprehensively work time factor , energy efficiency ratio and load change factor through the non - standard interaction correction term:

[0055]

[0056] Among them is the non - standard interaction correction term of agent i, and are the coupling parameters of the non - standard interaction correction term, obtained through numerical experiments or historical data fitting, and obtained through device simulation or optimization calculation based on device characteristics and operating environment. Combining multiple influencing factors into one correction term can more accurately reflect the cooperation and energy efficiency optimization requirements of the air - conditioning terminal during actual operation;

[0057] Pair the agents pairwise. According to the operation time of each air - conditioning terminal in a day, record the synchronous operation duration of each pair of agents in a day, and calculate the ratio of the synchronous operation duration to the maximum synchronous operation duration (usually 24h) and the ratio of the synchronous operation duration to the operation duration of agent i . And calculate the sum of for all agent pairs containing agent i as the cooperation coefficient of agent i, and calculate the sum of for all agent pairs containing agent i as the operation coefficient of agent i;

[0058] ​By calculating the ratio of the synchronous operation duration between the air-conditioning terminal and other devices to the maximum synchronous operation duration, a collaboration coefficient is formed, which reflects the degree of collaborative work between each air-conditioning terminal and other devices. The operation coefficient is the working proportion of the air-conditioning terminal in the whole system. These coefficients can be used to quantify the collaboration and participation of each agent;

[0059] According to the collaboration coefficient of agent i and the operation coefficient define the time correlation term :

[0060]

[0061] The time correlation term reflects the time synchronization relationship between devices through the ratio of operation durations between agents, considering the timing and collaborative effects between different air-conditioning terminals. It can dynamically adjust the relative contribution of each agent in the system and optimize the overall system efficiency through the collaboration intensity;

[0062] According to the time correlation term calculate the collaboration intensity between agent i and other agents :

[0063]

[0064] where is the number of agents;

[0065] Through the collaboration intensity define the total utility function U of the central air-conditioning as:

[0066]

[0067] where is the load weight of agent i, is the weight of the non-standard interaction correction term of agent i;

[0068] The total utility function is used to comprehensively consider the load weight, non-standard interaction correction term, and collaboration intensity of each agent, and finally measure the performance of the whole system. The introduction of this function in the present invention provides a clear quantitative goal for load optimization, ensuring that the overall system efficiency is maximized by optimizing the load of each agent;

[0069] Adopt the gradient descent method to optimize and update the load of each agent i:

[0070]

[0071] where is the load of the updated agent i, is the learning rate;

[0072] And the updated load of agent i and agent j is calculated :

[0073]

[0074] in is the ratio of the updated synchronous running time of agent i and agent j to the running time of agent i, and are the updated loads of agent i and agent j respectively;

[0075] based on The collaboration coefficient and operation coefficient of agent i are recalculated, and the total utility function value of the central air conditioner is recalculated according to the updated load of the agent. The agent load is iteratively adjusted until the total utility function value of the central air conditioner reaches the maximum and then stops. The agent load that maximizes the total utility function value of the central air conditioner is output as the load of each air-conditioning terminal to form a central air-conditioning load optimization plan.

[0076] By treating each air-conditioning terminal as an independent intelligent agent for modeling, local optimization in the air-conditioning system can be achieved, while taking into account the collaborative effect between devices, the introduction of non-standard interaction correction terms and collaborative coefficients, and the quantification of the collaborative relationship between air-conditioning terminals, the load of each device not only considers its own energy efficiency, but also comprehensively considers the collaborative effect with other devices. By introducing these correction terms, the load distribution of each device can be finely scheduled, so that the system can achieve the best energy utilization effect while meeting the comfort requirements. Through real-time optimization and dynamic adjustment of the load of each intelligent agent, the system can respond and make adjustments quickly to ensure that the equipment load is always in the optimal state.

[0077] Furthermore, using a genetic algorithm to generate an operation state optimization plan refers to randomly initializing the operation state and operation time of the air-conditioning terminal in the central air-conditioning operation data to reach the set temperature to generate an initial population, each air-conditioning terminal represents an individual in the initial population, the initial population is the input of the genetic algorithm, and represents the diversity of the initial air-conditioning terminal state. Each air-conditioning terminal, as an individual in the population, has a randomly initialized operation state and operation time. The genetic algorithm relies on the diversity in the population to explore the solution space, and ensures the extensiveness of the initial solution through random initialization, providing a higher global search capability for subsequent optimization. The initial population is optimized through the genetic algorithm, and the fitness of the individuals in the initial population is calculated to select individuals for crossover mutation operations to generate new individuals. The fitness of the new individuals is calculated for iteration, and the iteration is stopped after the fitness converges, and the operation state and operation time of the air-conditioning terminal in the individual with the highest fitness in the iterated population is used as the operation state optimization plan.

[0078] Through the iterative optimization of the genetic algorithm, the finally generated operating state and operating time represent the optimal air-conditioning scheduling strategy. These optimization schemes can maximize energy efficiency and reduce energy waste while ensuring comfort.

[0079] S3. Regulate the central air-conditioning according to the load optimization scheme and the operating state optimization scheme;

[0080] Specifically, regulating the central air-conditioning according to the load optimization scheme and the operating state optimization scheme means optimizing the control of the air-conditioning terminals of the central air-conditioning according to the generated load optimization scheme and operating state optimization scheme, and continuously generating the load optimization scheme and operating state optimization scheme by real-time monitoring of the central air-conditioning operation data for the operation regulation of the central air-conditioning.

[0081] Using the generated load optimization scheme and operating state optimization scheme, combined with the actual requirements and operating conditions of the air-conditioning terminals, dynamically scheduling the system not only realizes the accurate prediction and optimization of the load, but also can adjust according to the operating state of the air-conditioning terminals, so that the working load of each terminal always remains within the optimal range. Through the real-time control of the load and operating state, the energy waste of the air-conditioning system can be reduced, the overall energy efficiency of the system can be improved. Real-time monitoring of the central air-conditioning operation data and continuously generating the load optimization scheme and operating state optimization scheme can provide continuously optimized control strategies during the system operation. Through the real-time feedback mechanism, the air-conditioning system can continuously self-adjust during the actual operation to ensure that the load distribution and operating mode are always in the optimal state.

[0082] S4. Real-time display the operation state and load data of the central air-conditioning, and record and store the load data in the database;

[0083] Specifically, real-time display of the operation state and load data of the central air-conditioning means that the data center updates and displays the monitored operation state and load data of the central air-conditioning in real time, synchronously compares and displays the obtained load optimization scheme of the central air-conditioning with the central air-conditioning operation data, and forms a data record according to the time stamp.

[0084] Furthermore, recording and storing the load data in the database means that while the data center displays the load data, it stores the load data and the data record in the database. The database regularly detects the integrity and security of the stored data and uploads it to the cloud for backup.

[0085] Embodiment 2

[0086] When the above-mentioned functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0087] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch instructions from the instruction execution system, apparatus, or device and execute the instructions), or in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in combination with an instruction execution system, apparatus, or device.

[0088] More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection part with one or more wirings (electronic device), a portable computer disk cartridge (magnetic device), random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), fiber optic device, and portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other suitable processing as necessary, and then stored in a computer memory.

[0089] It should be understood that each part of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one of the following techniques known in the art or a combination thereof can be used: discrete logic circuits having logic gate circuits for implementing logic functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0090] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. An optimized control method for ice storage central air conditioning based on load prediction and optimized allocation, characterized in that: including, obtaining the operation data of the central air conditioner for preprocessing, and predicting the central air conditioner load through an intelligent algorithm; forming an optimization plan for the central air conditioner load according to the predicted load, and generating an optimization plan for the operation state using a genetic algorithm; regulating the central air conditioner according to the load optimization plan and the operation state optimization plan; real-time displaying the operation state and load data of the central air conditioner, and recording and storing the load data in a database; The predicting the load through intelligent learning refers to constructing a prediction model through an LSTM model, including an input gate, a forgetting gate, and an output gate. The input of the prediction model is set as the operation data of the central air conditioner, and the output is the predicted load of the central air conditioner. The prediction model is trained using a training data set, and the mean square error is set as the loss function. The gradient of the loss function is calculated through backpropagation, and the parameters of the prediction model are updated using an optimization algorithm until the loss function is minimized. The parameters of the trained prediction model are used to optimize the prediction model; Inputting the operation parameters of the central air conditioner into the optimized prediction model to obtain the future load of the central air conditioner, and introducing a reinforcement learning mechanism to continuously optimize the prediction model using the Q-learning algorithm; The forming the optimization plan for the central air conditioner load according to the predicted load includes, Define each air-conditioning terminal in the central air-conditioning system as an independent agent \(i\), and extract the load of each air-conditioning terminal from the future load of the central air-conditioning system , and obtain the operating time of each air-conditioning terminal according to the operating data of the central air-conditioning system, and calculate the proportion of the operating time of each air-conditioning terminal in a day as the working time factor of agent \(i\) ; Query the energy efficiency curve of the air-conditioning terminal to obtain the energy efficiency ratio of the air-conditioning terminal under the condition of load ; ; Statistically analyze the load change data of the air-conditioning terminal during the operation time, fit the load change curve through regression analysis, and obtain the absolute value of the slope of the load change curve of the air-conditioning terminal at the load as the load change factor ; Correction of the comprehensive working time factor through non-standard interactions , energy efficiency ratio and load change factor : Among them is the non-standard interaction correction term for agent i, and are the coupling parameters of the non-standard interaction correction term; Pair the agents in pairs, record the synchronous operation duration of each pair of agents within a day according to the operation time of each air-conditioning terminal in a day, and calculate the ratio of the synchronous operation duration to the maximum synchronous operation duration and the ratio of the synchronous operation duration to the operation duration of agent i , and calculate for all agent pairs containing agent i the sum as the cooperation coefficient of agent i , and calculate for all agent pairs containing agent i the sum as the operation coefficient of agent i ; According to the cooperation coefficient of agent i and the operation coefficient define the time correlation term : According to the time association item Calculate the collaboration intensity between agent i and other agents : wherein is the number of agents; Through the collaboration intensity Define the total utility function U of the central air conditioner as follows: wherein is the load weight of agent i, is the non-standard interaction correction term weight of agent i; Use the gradient descent method to optimize and update the load of each agent i and update the calculation through the loads of the updated agent i and agent j : Among them is the ratio of the synchronous running duration of the updated agent i and agent j to the running duration of agent i, and are the loads of the updated agent i and agent j respectively; Based on Recalculate the cooperation coefficient and operation coefficient of agent i, and recalculate the total utility function value of the central air conditioner according to the updated load of the agent. Iteratively adjust the agent load until the total utility function value of the central air conditioner reaches the maximum and then stop. Output the agent load that maximizes the total utility function value of the central air conditioner as the load of each air-conditioning terminal to form an optimization plan for the central air-conditioning load.

2. The optimized control method for ice storage central air conditioning based on load prediction and optimized distribution according to claim 1, wherein: The obtaining the operation data of the central air conditioner for preprocessing refers to collecting the operation data of the central air conditioner by installing sensors, wirelessly connecting the sensors to form a sensor network and connecting the sensor network to a data center. The data center cleans the operation data of the central air conditioner collected by the sensor network and performs denoising processing using Kalman filtering, and normalizes the denoised operation data of the central air conditioner.

3. The optimized control method for a chilled water central air-conditioning based on load prediction and optimized distribution according to claim 2, wherein: The generating the optimization plan for the operation state using a genetic algorithm refers to randomly initializing the operation state and operation time when the air conditioner terminal reaches the set temperature in the operation data of the central air conditioner to generate an initial population. Each air conditioner terminal represents an individual in the initial population. The initial population is optimized through a genetic algorithm, and the fitness of the individuals in the initial population is calculated to select individuals for crossover and mutation operations to generate new individuals. The fitness of the new individuals is calculated for iteration, and the iteration stops after the fitness converges. The operation state and operation time of the air conditioner terminal in the individual with the highest fitness in the iterated population are used as the optimization plan for the operation state.

4. The optimized control method for ice storage central air conditioning based on load prediction and optimized allocation as claimed in claim 3, wherein: The regulating the central air conditioner according to the load optimization plan and the operation state optimization plan refers to optimizing the control of the air conditioner terminals of the central air conditioner according to the generated load optimization plan and operation state optimization plan, and continuously monitoring the operation data of the central air conditioner to continuously generate the load optimization plan and the operation state optimization plan for the operation regulation of the central air conditioner.

5. The optimized control method for ice storage central air conditioning based on load prediction and optimized distribution as described in claim 4, characterized in that: The real-time displaying the operation state and load data of the central air conditioner refers to the data center real-time updating and displaying the monitored operation state and load data of the central air conditioner, synchronously comparing and displaying the obtained load optimization plan of the central air conditioner with the operation data of the central air conditioner, and forming a data record according to the time stamp.

6. The optimized control method for ice storage central air conditioning based on load prediction and optimal allocation according to claim 5, characterized in that: The recording and storing the load data in a database refers to the data center storing the load data and the data record in the database while displaying the load data. The database regularly detects the integrity and security of the stored data and uploads it to the cloud for backup.

7. A computer device, comprising: Memory and processor; The memory stores a computer program, characterized in that: when the processor executes the computer program, the steps of the optimized control method for a chilled water central air conditioner based on load prediction and optimized allocation according to any one of claims 1 to 6 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the steps of the optimized control method for a chilled water central air conditioner based on load prediction and optimized allocation according to any one of claims 1 to 6 are implemented.

Images