A Parallel Control Method and Device for a Smart Heating System Based on ACP Theory

Abstract

The present invention discloses a parallel control method for an intelligent heating system based on the ACP theory, including: perceiving data information of heat sources, heat substations, heat networks, and heat users through an actual intelligent heating system; establishing an artificial intelligent heating system by using the data information, physical entity structure, and parameters of the actual intelligent heating system, and adopting mechanism description, knowledge representation, and machine learning methods; obtaining an optimized intelligent heating control strategy through computational experiments, and inputting it into a parallel intelligent heating system formed by combining the actual intelligent heating system and the artificial intelligent heating system for parallel execution, continuously correcting the model structure and parameters of the artificial intelligent heating system to make the artificial intelligent heating system approach the actual intelligent heating system for the cultivation of the artificial intelligent heating system; and through parallel control, continuously correcting the intelligent heating control strategy to make the intelligent heating control strategy of the actual intelligent heating system approach the optimized control strategy of the artificial intelligent heating system.

Description

Technical Field

[0001] The present invention belongs to the technical field of smart heating, and specifically relates to a parallel control method and device for a smart heating system based on ACP theory. Background Art

[0002] The heating system has complex characteristics such as uncertainty, strong coupling and large delay. With the increasing demand for heat from users and the increasing awareness of heating companies on optimizing the control of heating systems, these characteristics of heating systems are becoming increasingly complex, and with this comes the continuous improvement of various control requirements for heating systems.

[0003] At present, advanced control algorithms such as fuzzy control, predictive control, neural networks, etc. are constantly being proposed. Although these control strategies have shown their own advantages, it is difficult to achieve real-time and accurate control for complex systems with strong uncertainty such as the operation process of the heating system. The heating system process is a complex dynamic process. Its operating conditions are not constant. In addition, there will be various internal and external disturbances during the operation process, and even sudden situations such as local failures. The above problems make the design of its control system face great challenges.

[0004] Based on the above technical problems, it is necessary to design a new parallel control method and device for intelligent heating system based on ACP theory. Summary of the invention

[0005] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a parallel control method and device for an intelligent heating system based on the ACP theory, which can realize the parallel execution of the actual intelligent heating system and the artificial intelligence heating system during the operation of the heating system, and make real-time corrections to the artificial intelligence heating system model and parameters, as well as the heating control strategy, to ensure the stable operation of the heating system and meet the system operation evaluation requirements. The parallel control method has smaller overshoot and less adjustment time during the control process of the heating system, and is not prone to model mismatch problems. It has significant control performance advantages and is highly feasible.

[0006] In order to solve the above technical problems, the technical solution of the present invention is:

[0007] The present invention provides a smart heating parallel control method based on ACP theory, and the smart heating parallel control method comprises:

[0008] The data information of heat sources, heating stations, heating networks and heat users is sensed through the data acquisition devices set up in the actual smart heating system;

[0009] Utilize the data information, physical entity structure, and parameters of the actual intelligent heating system, and adopt mechanism description, knowledge representation, and machine learning methods to establish an artificial intelligent heating system;

[0010] Obtain an optimized intelligent heating control strategy through computational experiments, and input it into the parallel intelligent heating system formed by the combination of the actual intelligent heating system and the artificial intelligent heating system for parallel execution. Continuously correct the model structure and parameters of the artificial intelligent heating system to make the artificial intelligent heating system approach the actual intelligent heating system for the cultivation of the artificial intelligent heating system;

[0011] Simultaneously input the optimized intelligent heating control strategy obtained through computational experiments into the cultivated artificial intelligent heating system and the actual intelligent heating system for parallel control. The artificial intelligent heating system performs synchronous computational experiments according to the state data of the actual intelligent heating system to obtain the state data of the artificial intelligent heating system. Through data comparison and feedback error, continuously correct the intelligent heating control strategy to make the intelligent heating control strategy of the actual intelligent heating system approach the optimized control strategy of the artificial intelligent heating system.

[0012] Furthermore, the data information of the heat source, heat substation, heat network, and heat users is sensed through the data acquisition devices set in the actual intelligent heating system, including:

[0013] Set a first actual space, a second actual space, and a third actual space in the actual intelligent heating system. Obtain the opening degree of the primary valve, primary supply water temperature, secondary supply water temperature, secondary supply water temperature, secondary return water temperature, circulating pump frequency, building entrance supply water temperature, and user room temperature related to the heat source, heat substation, heat network, and heat users through the temperature acquisition device, pressure acquisition device, flow acquisition device, and pump valve data acquisition device set in the first actual space; obtain meteorological data including outdoor temperature, humidity, wind speed, sunshine, rainfall, and snowfall through the second actual space; transmit the data information of the first actual space and the second actual space through the third actual space.

[0014] Furthermore, the utilization of the data information, physical entity structure, and parameters of the actual intelligent heating system, and the establishment of an artificial intelligent heating system by adopting mechanism description, knowledge representation, and machine learning methods include:

[0015] Set a first artificial space in the artificial intelligent heating system for parallel information execution with the third actual space, and obtain the data information of the actual intelligent heating system through the first artificial space;

[0016] Using the data information of the actual intelligent heating system, combining the physical entity structures and parameters of heat sources, heat stations, heat networks, and heat users, and aiming at the changes in the operating state of the actual intelligent heating system, adopt mechanism description, knowledge representation, and machine learning methods. Based on the operating mechanism model of the system itself, combined with the knowledge graph model accumulated by experts and the simulation model of digital twins, construct a software-defined computable, reconfigurable, and programmable artificial intelligence heating system model, including at least a heat source model, a heat station model, a heat network model, a heat user model, an outdoor meteorological model, a hydraulic condition model, and a correlation relationship library of each model;

[0017] Using the established artificial intelligence heating system model, by changing the model parameters and relationship weights defined by software, describe different operating states of the system, and use the multi-agent method to describe, model, and evaluate the changes in different operating states, and establish a virtual artificial intelligence heating system that executes in parallel with the actual intelligent heating system.

[0018] Furthermore, after establishing the artificial intelligence heating system, it also includes:

[0019] Using the real data tested by the actual intelligent heating system to verify the credibility of the models in the artificial intelligence heating system: use the same heat source output, heat station control strategy, heat network parameters, and meteorological data for the actual intelligent heating system and the artificial intelligence heating system, measure and calculate the change states of the primary supply water temperature, secondary supply water temperature, secondary supply water temperature, secondary return water temperature, building entrance supply water temperature, and user room temperature. By comparing the real measurement data and the model output data calculated manually, correct and optimize the models in the artificial intelligence heating system until the data is relatively consistent, then pass the credibility verification, indicating that this artificial intelligence heating system can be used for subsequent computational experiments and parallel execution processes.

[0020] Furthermore, the use of computational experiments to obtain optimized intelligent heating control strategies includes:

[0021] Design the heat source model, heat station model, heat network model, heat user model, outdoor meteorological model, and hydraulic condition model in the established artificial intelligence heating system as different intelligent agents. Through the combined interaction rules of each intelligent agent, establish different behavioral characteristic models according to the behavior of a single intelligent agent and the behavior of the group of intelligent agents, establish behavioral intelligent agents, and through reinforcement learning, each intelligent agent continuously interacts with the controlled system object in the closed-loop system to generate different system operation scenarios. After simulation, simulation, and calculation, obtain the experimental test data under different operation scenarios;

[0022] Using data mining, feature extraction, machine learning methods, intelligent optimization algorithms, and reinforcement learning algorithms to train and learn the experimental test data, and obtaining optimized intelligent heating control strategies under different system operation scenarios, including heat source load optimization distribution strategies, pump valve regulation strategies, heat user load prediction, and heating parameter optimization strategies;

[0023] Among them, the process of computational experiment includes: determining the start time, step size, duration, initial conditions, and constraint conditions of the experiment; setting corresponding intelligent heating control strategies for different experimental scenarios; inputting the experimental conditions and the set intelligent heating control strategies into the model in the artificial intelligent heating system to execute the computational experiment process, and recording the experimental results of the system model parameters, system operation status, and the change process of the hydraulic condition during the experiment; after the computational experiment is completed, the artificial intelligent heating system transmits the experimental results to the analysis and evaluation module to analyze and evaluate the safety, stability, economy, energy conservation and environmental protection, and user thermal comfort of the system. If the evaluation results meet the preset requirements, the intelligent heating control strategy of the experimental process is used as the optimization result; otherwise, the intelligent heating control strategy is corrected, and then enters the next round of computational experiment process. Through multiple cyclic experiments, an optimized intelligent heating control strategy is obtained.

[0024] Furthermore, the cultivation of the artificial intelligent heating system includes:

[0025] Inputting the intelligent heating control strategy and the state data of the initial intelligent heating system into the parallel intelligent heating system formed by combining the actual intelligent heating system and the artificial intelligent heating system. Through the parallel execution of multiple artificial intelligent heating systems and the actual intelligent heating system, obtaining the computational experiment results of multiple artificial intelligent heating systems, comparing the computational experiment results of different artificial intelligent heating systems with the results of the actual intelligent heating system to obtain the state errors of different artificial intelligent heating systems, and then feeding back the state errors to the corresponding artificial intelligent heating systems to correct the model structure and parameters of the artificial intelligent heating systems. The corrected artificial intelligent heating systems continue to operate to form a closed-loop control, making the computational experiment results of the artificial intelligent heating systems continuously approach the results of the actual intelligent heating system.

[0026] Furthermore, inputting the optimized intelligent heating control strategy obtained from the computational experiment into the cultivated artificial intelligent heating system and the actual intelligent heating system simultaneously for parallel control. The artificial intelligent heating system performs synchronous computational experiments according to the state data of the actual intelligent heating system to obtain the state data of the artificial intelligent heating system. Through data comparison and feedback error, continuously correct the intelligent heating control strategy to make the intelligent heating control strategy of the actual intelligent heating system approach the optimized control strategy of the artificial intelligent heating system, including:

[0027] Apply the optimized intelligent heating control strategy and the state data of the initial intelligent heating system to both the actual intelligent heating system and the cultivated artificial intelligent heating system simultaneously for parallel control. The artificial intelligent heating system performs synchronous calculation experiments according to the state data of the actual intelligent heating system to obtain the state data of the artificial intelligent heating system. Compare the state data of the actual intelligent heating system with the calculation experiment results of the artificial intelligent heating system, and feedback the error through the virtual-real interaction method to continuously correct the intelligent heating control strategy. Combine the original optimized control strategy and interference information to obtain a more optimized intelligent heating control strategy. Through multiple feedback corrections by parallel control, the control strategy of the actual intelligent heating system approaches the optimized control strategy of the artificial intelligent heating system, achieving the optimized control of the actual intelligent heating system.

[0028] Furthermore, a fourth actual space is also set up in the actual intelligent heating system to obtain the social behavior information of heat users and the heating company, including user activity behavior information, metering and charging information of the heating company, and attribute information of the buildings of heating users; a second artificial space is also set up in the artificial intelligent heating system to build a virtual space corresponding to the fourth actual space, imitate the social behavior of heat users and the heating company, and perform parallel execution.

[0029] Furthermore, the parallel execution between the third actual space and the first artificial space is to use the data obtained from the calculation experiment and the intelligent heating control strategy to guide the third actual space. Through the connection between the third actual space and the first artificial space in parallel execution, apply the calculation experiment results in the first artificial space to the third actual space and the first artificial space, and optimize and adjust the first artificial space according to the feedback information formed by the operation results and target deviation of the third actual space and the first artificial space, realizing the parallel execution of virtual-real interaction;

[0030] A parallel blockchain is also set up in the parallel intelligent heating system formed by the combination of the actual intelligent heating system and the artificial intelligent heating system to securely store, verify, and securely transmit the data information transmitted by the parallel intelligent heating system; the parallel blockchain is an artificial blockchain based on the construction of a multi-mapping of the real blockchain. Analyze and evaluate the evolution law of the real blockchain through the calculation experiment on the artificial blockchain, and apply the data and strategies obtained through the calculation experiment to the real blockchain and the artificial blockchain by the parallel execution method to manage and control the real blockchain.

[0031] The present invention also provides an intelligent heating system parallel control device based on the ACP theory. The intelligent heating system parallel control device includes:

[0032] An actual system data acquisition unit, configured to sense data information of heat sources, heat stations, heat networks, and heat users through data acquisition devices provided in the actual intelligent heating system;

[0033] An artificial system establishment unit, which uses the data information, physical entity structure, and parameters of the actual intelligent heating system, and adopts mechanism description, knowledge representation, and machine learning methods to establish an artificial intelligent heating system;

[0034] A computational experiment unit, configured to obtain an optimized intelligent heating control strategy through computational experiments;

[0035] A parallel execution unit, configured to input the optimized intelligent heating control strategy into a parallel intelligent heating system formed by combining the actual intelligent heating system and the artificial intelligent heating system for parallel execution, continuously correcting the model structure and parameters of the artificial intelligent heating system to make the artificial intelligent heating system approach the actual intelligent heating system, and cultivating the artificial intelligent heating system; it is also configured to input the optimized intelligent heating control strategy obtained through computational experiments into the cultivated artificial intelligent heating system and the actual intelligent heating system simultaneously for parallel control. The artificial intelligent heating system performs synchronous computational experiments according to the state data of the actual intelligent heating system to obtain the state data of the artificial intelligent heating system. Through data comparison and feedback error, the intelligent heating control strategy is continuously corrected to make the intelligent heating control strategy of the actual intelligent heating system approach the optimized control strategy of the artificial intelligent heating system.

[0036] The beneficial effects of the present invention are as follows:

[0037] By proposing an intelligent heating parallel control method based on the ACP theory and applying it to the intelligent heating system, the present invention constructs an artificial intelligent heating system equivalent to the actual intelligent heating system on the basis of the ACP theory. Through computational experiments, the optimal artificial intelligent heating system model and its corresponding heating control strategy are selected to guide and evaluate the operation of the actual intelligent heating system, making it gradually approach the artificial intelligent heating system; and realizing the parallel execution of the actual intelligent heating system and the artificial intelligent heating system during the operation of the heating system, and real-time correcting the model and parameters of the artificial intelligent heating system and the heating control strategy to ensure the stable operation of the heating system and meet the system operation evaluation requirements. This parallel control method has a smaller overshoot and less adjustment time during the control process of the heating system, and is not prone to model mismatch problems, with significant control performance advantages and great feasibility.

[0038] Other features and advantages will be described in the subsequent description, and some of them will become obvious from the description, or be understood by implementing the present invention. The objectives and other advantages of the present invention are achieved and obtained through the structures specifically pointed out in the description and the drawings.

[0039] In order to make the above objects, features and advantages of the present invention more obvious and understandable, the following specifically gives preferred embodiments and, in conjunction with the accompanying drawings, the detailed description is as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required to be used in the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0041] Figure 1 Schematic diagram of the process of a smart heating parallel control method based on the ACP theory of the present invention;

[0042] Figure 2 Schematic diagram of the structure of the parallel smart heating system of the present invention;

[0043] Figure 3 Schematic diagram of the cultivation process of the artificial intelligence heating system by the parallel execution of the artificial intelligence heating system and the actual smart heating system of the present invention;

[0044] Figure 4 Schematic diagram of the process of optimizing the heating control strategy and realizing parallel control by the parallel execution of the artificial intelligence heating system and the actual smart heating system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0045] In order to make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the drawings. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0046] Embodiment 1

[0047] Figure 1 It is a schematic diagram of the process of a smart heating parallel control method based on the ACP theory involved in the present invention.

[0048] Figure 2 It is a schematic diagram of the structure of the parallel smart heating system involved in the present invention.

[0049] Figure 3 It is a schematic diagram of the cultivation process of the artificial intelligence heating system by the parallel execution of the artificial intelligence heating system and the actual smart heating system involved in the present invention.

[0050] Figure 4 Schematic diagram of the process for optimizing the heating control strategy and implementing parallel control for the parallel execution of the artificial intelligence heating system and the actual intelligent heating system involved in the present invention.

[0051] As Figures 1-4 shown, Embodiment 1 of the present invention provides an intelligent heating parallel control method based on the ACP theory. The intelligent heating parallel control method includes:

[0052] Perceiving data information of heat sources, heat stations, heat networks, and heat users through data acquisition devices set in the actual intelligent heating system;

[0053] Using the data information, physical entity structure, and parameters of the actual intelligent heating system, and adopting mechanism description, knowledge representation, and machine learning methods to establish an artificial intelligence heating system;

[0054] Obtaining an optimized intelligent heating control strategy through computational experiments, and inputting it into the parallel intelligent heating system formed by combining the actual intelligent heating system and the artificial intelligence heating system for parallel execution, continuously correcting the model structure and parameters of the artificial intelligence heating system to make the artificial intelligence heating system approach the actual intelligent heating system for the cultivation of the artificial intelligence heating system;

[0055] Simultaneously inputting the optimized intelligent heating control strategy obtained through computational experiments into the cultivated artificial intelligence heating system and the actual intelligent heating system for parallel control. The artificial intelligence heating system performs synchronous computational experiments according to the state data of the actual intelligent heating system to obtain the state data of the artificial intelligence heating system. Through data comparison and feedback error, continuously correct the intelligent heating control strategy to make the intelligent heating control strategy of the actual intelligent heating system approach the optimized control strategy of the artificial intelligence heating system.

[0056] In this embodiment, the perceiving of data information of heat sources, heat stations, heat networks, and heat users through data acquisition devices set in the actual intelligent heating system includes:

[0057] Setting a first actual space, a second actual space, and a third actual space in the actual intelligent heating system. Obtaining the opening degree of the primary valve, the primary supply water temperature, the secondary supply water temperature, the secondary supply water temperature, the secondary return water temperature, the circulating pump frequency, the building entrance supply water temperature, and the user room temperature related to heat sources, heat stations, heat networks, and heat users through the temperature acquisition device, pressure acquisition device, flow acquisition device, and pump valve data acquisition device set in the first actual space; obtaining meteorological data including outdoor temperature, humidity, wind speed, sunshine, rainfall, and snowfall through the second actual space; transmitting the data information of the first actual space and the second actual space through the third actual space.

[0058] In this embodiment, by using the data information, physical entity structure and parameters of the actual intelligent heating system, and adopting mechanism description, knowledge representation and machine learning methods, an artificial intelligent heating system is established, including:

[0059] In the artificial intelligent heating system, a first artificial space for parallel information execution with the third actual space is set, and the data information of the actual intelligent heating system is obtained through the first artificial space;

[0060] Using the data information of the actual intelligent heating system, combining the physical entity structure and parameters of heat sources, heat stations, heat networks and heat users, and aiming at the change of the operation state of the actual intelligent heating system, adopting mechanism description, knowledge representation and machine learning methods, based on the operation mechanism model of the system itself, combining the knowledge graph model accumulated by experts and the simulation model of digital twins, a software-defined computable, reconfigurable and programmable artificial intelligent heating system model is constructed, at least including a heat source model, a heat station model, a heat network model, a heat user model, an outdoor weather model, a hydraulic condition model and a correlation relationship library of each model;

[0061] Using the established artificial intelligent heating system model, by changing the model parameters and relationship weights defined by software, different operation states of the system are described, and the multi-agent method is used to describe, model and evaluate the changes of different operation states, and a virtual artificial intelligent heating system parallel to the actual intelligent heating system is established.

[0062] In this embodiment, after the artificial intelligent heating system is established, it further includes:

[0063] Using the real data tested by the actual intelligent heating system to verify the credibility of the models in the artificial intelligent heating system: using the same heat source output, heat station control strategy, heat network parameters and meteorological data for the actual intelligent heating system and the artificial intelligent heating system, measuring and calculating the change states of the primary supply water temperature, secondary supply water temperature, secondary return water temperature, building entrance supply water temperature and user room temperature, and by comparing the real measurement data and the model output data calculated artificially, the models in the artificial intelligent heating system are corrected and optimized until the data are relatively consistent, then the credibility verification is passed, indicating that the artificial intelligent heating system can be used for subsequent computational experiments and parallel execution processes.

[0064] In this embodiment, the use of computational experiments to obtain optimized intelligent heating control strategies includes:

[0065] Design the heat source model, heat substation model, heat network model, heat user model, outdoor meteorological model, and hydraulic condition model in the established artificial intelligence heating system as different intelligent agents. Based on the combination and interaction rules of each intelligent agent, establish different behavioral feature models according to the behavior of individual intelligent agents and the behavior of the group of intelligent agents, and establish behavioral intelligent agents. Through reinforcement learning, each intelligent agent continuously interacts with the controlled system object in the closed-loop system to generate different system operation scenarios. After simulation, simulation, and calculation, obtain the experimental test data under different operation scenarios;

[0066] Use data mining, feature extraction, machine learning methods, intelligent optimization algorithms, and reinforcement learning algorithms to train and learn the experimental test data to obtain optimized intelligent heating control strategies under different system operation scenarios, including heat source load optimal distribution strategies, pump valve regulation strategies, heat user load prediction, and heating parameter optimization strategies;

[0067] Among them, the process of the computational experiment includes: determining the start time, step size, duration, initial conditions, and constraint conditions of the experiment; setting the corresponding intelligent heating control strategies for different experimental scenarios; inputting the experimental conditions and the set intelligent heating control strategies into the models in the artificial intelligence heating system, executing the computational experiment process, and recording the experimental results of the system model parameters, system operation status, and the change process of the hydraulic condition during the experiment; after the computational experiment is completed, the artificial intelligence heating system transmits the experimental results to the analysis and evaluation module to analyze and evaluate the safety, stability, economy, energy conservation and environmental protection, and user thermal comfort of the system. If the evaluation results meet the preset requirements, then use the intelligent heating control strategy of the experimental process as the optimization result; otherwise, correct the intelligent heating control strategy, and then enter the next round of computational experiment process. Through multiple loop experiments, obtain the optimized intelligent heating control strategy.

[0068] In this embodiment, the cultivation of the artificial intelligence heating system includes:

[0069] Input the intelligent heating control strategy and the state data of the initial intelligent heating system into the parallel intelligent heating system formed by combining the actual intelligent heating system and the artificial intelligence heating system. Through the parallel execution of multiple artificial intelligence heating systems and the actual intelligent heating system, obtain the computational experiment results of multiple artificial intelligence heating systems. Compare the computational experiment results of different artificial intelligence heating systems with the results of the actual intelligent heating system to obtain the state errors of different artificial intelligence heating systems, and then feedback the state errors to the corresponding artificial intelligence heating system to correct the model structure and parameters of the artificial intelligence heating system. The corrected artificial intelligence heating system continues to run to form a closed-loop control, so that the computational experiment results of the artificial intelligence heating system continuously approach the results of the actual intelligent heating system.

[0070] In this embodiment, the optimized intelligent heating control strategy obtained from the computational experiment is simultaneously input into the cultivated artificial intelligent heating system and the actual intelligent heating system for parallel control. The artificial intelligent heating system performs synchronous computational experiments according to the state data of the actual intelligent heating system to obtain the state data of the artificial intelligent heating system. Through data comparison and feedback error, the intelligent heating control strategy is continuously corrected, so that the intelligent heating control strategy of the actual intelligent heating system approaches the optimized control strategy of the artificial intelligent heating system, including:

[0071] The optimized intelligent heating control strategy and the state data of the initial intelligent heating system are simultaneously applied to the actual intelligent heating system and the cultivated artificial intelligent heating system for parallel control. The artificial intelligent heating system performs synchronous computational experiments according to the state data of the actual intelligent heating system to obtain the state data of the artificial intelligent heating system. Compare the state data of the actual intelligent heating system with the computational experiment results of the artificial intelligent heating system, feedback the error through the virtual-real interaction method, continuously correct the intelligent heating control strategy, combine the original optimized control strategy and interference information to obtain a more optimized intelligent heating control strategy. Through multiple feedback corrections in parallel control, the control strategy of the actual intelligent heating system approaches the optimized control strategy of the artificial intelligent heating system, realizing the optimized control of the actual intelligent heating system.

[0072] It should be noted that the ACP theory composed of Artificial societies, Computational experiments, and Parallel execution plays an important role in the modeling and regulation of complex systems. The ACP method is not only an organic combination of artificial societies, computational experiments, and parallel execution, but also an integration of psychology, information, intelligence, simulation, decision-making, and execution. Based on the core idea of transforming the "virtual" and "soft" parts of a complex system into a decomposable, computable, and executable process, first, the actual physical system is modeled through artificial organization, then the system is analyzed and evaluated through computational experiments, and finally, parallel execution is used to control and manage the physical system, making it a main control process and gradually approaching the artificial system, thus achieving a relatively ideal state.

[0073] The core of the present invention is that the core of the parallel system is the ACP method, which is mainly completed by three parts: ① Artificial Society (A): Driven by the data of the actual system, and by means of system mechanism, knowledge representation, machine learning, etc., for various elements and problems in the actual system, construct computable, reconfigurable, programmable software-defined objects, software-defined processes, software-defined relationships, etc., and then combine these objects, processes, relationships, etc. into a software-defined artificial system, and use the artificial system to model complex system problems; ② Computational Experiment (C): Based on this "computational laboratory" of the artificial system, use computational experiments to design the combination and interaction rules of various agents, generate various scenarios, run to generate complete scenario data, and by means of machine learning, parallel dynamic programming, data mining, etc., analyze the data to obtain the optimal strategies under various scenarios; ③ Parallel Execution (P): Simultaneously promote the artificial system and the actual system, and through a certain way, conduct virtual-real interaction to guide and manage the actual system with parallel execution. In terms of the process, the parallel system completes the closed-loop processing process through data acquisition, artificial system modeling, computational experiment scenario deduction, experimental analysis and prediction, control decision optimization and implementation, real-virtual system real-time feedback, and implementation effect real-time evaluation. Computational experiments usually include learning and training, experiment and evaluation, prediction and control, deeply understand the operation state change process of the intelligent heating system under various complex working conditions, discover the change rules of the system operation state, and evaluate the performance of the intelligent heating system under different control and management strategies. The processes of learning and training, experiment and evaluation, prediction and control are carried out simultaneously in the actual intelligent heating system and the artificial intelligent heating system, which is called parallel execution. Through parallel execution, the model of the artificial intelligent heating system can be further improved to guide the prediction and control process of the actual intelligent heating system.

[0074] In this embodiment, a fourth actual space is further set in the actual intelligent heating system to obtain the social behavior information of heat users and the heating company, including user activity behavior information, heating company metering and charging information, and the attribute information of the buildings of heating users; a second artificial space is further set in the artificial intelligent heating system to build a virtual space corresponding to the fourth actual space, imitate the social behavior of heat users and the heating company, and conduct parallel execution.

[0075] In this embodiment, the parallel execution between the third actual space and the first artificial space is to guide the third actual space by using the data obtained from computational experiments and the intelligent heating control strategy. Through the connection between the third actual space and the first artificial space in parallel execution, the computational experiment results in the first artificial space are applied to both the third actual space and the first artificial space, and the first artificial space is optimized and adjusted according to the feedback information composed of the operation results and target deviations of the third actual space and the first artificial space, so as to achieve the parallel execution of virtual-real interaction;

[0076] In the parallel intelligent heating system formed by combining the actual intelligent heating system and the artificial intelligent heating system, a parallel blockchain is also provided to securely store, verify, and securely transmit the data information transmitted by the parallel intelligent heating system; the parallel blockchain is an artificial blockchain based on constructing a multi-mapping of the real blockchain. Through computational experiments on the artificial blockchain, the evolution law of the real blockchain is analyzed and evaluated. The data and strategies obtained through computational experiments are applied to the real blockchain and the artificial blockchain by using the parallel execution method to manage and control the real blockchain.

[0077] It should be noted that the parallel execution between the third actual space and the first artificial space is equivalent to the parallel network of the parallel intelligent heating system. The third actual space is equivalent to the real network space, and the first artificial space is equivalent to the artificial network space. The parallel network is an important foundation for supporting the artificial network space and the real network space, and can achieve global control of the network space and self-adaptive optimization of the network space structure and parameters, etc. In the parallel network, the real network space feeds back the real operation data to the artificial network space in real time through the perception ability of the parallel network. The artificial network space constructs and improves its own structure and parameters based on the real data, realizing the multi-scenario mapping from the real network space to the artificial network space. Based on the artificial network space, through a large number of repeated computational experiment simulations of the real network, rich experimental data are obtained. Combining the real data to evaluate, analyze, and optimize the network control and management strategies, the real-time optimization of the real network security strategy is realized. Through the parallel execution method, the real network and the artificial network achieve virtual-real interaction and real-time linkage of parameter optimization.

[0078] Embodiment 2

[0079] This Embodiment 2 provides a parallel control device for an intelligent heating system based on the ACP theory. The parallel control device for the intelligent heating system includes:

[0080] An actual system data acquisition unit, which is used to sense the data information of the heat source, heat substation, heat network, and heat users through the data acquisition device set in the actual intelligent heating system;

[0081] An artificial system establishment unit, which uses the data information, physical entity structure and parameters of an actual intelligent heating system, and adopts mechanism description, knowledge representation and machine learning methods to establish an artificial intelligent heating system;

[0082] A computational experiment unit, which is used to obtain an optimized intelligent heating control strategy by means of computational experiments;

[0083] A parallel execution unit, which is used to input the optimized intelligent heating control strategy into a parallel intelligent heating system formed by combining the actual intelligent heating system and the artificial intelligent heating system for parallel execution, continuously correct the model structure and parameters of the artificial intelligent heating system, make the artificial intelligent heating system approach the actual intelligent heating system, and cultivate the artificial intelligent heating system; it is also used to input the optimized intelligent heating control strategy obtained by computational experiments into the cultivated artificial intelligent heating system and the actual intelligent heating system at the same time for parallel control. The artificial intelligent heating system performs synchronous computational experiments according to the state data of the actual intelligent heating system to obtain the state data of the artificial intelligent heating system. Through data comparison and feedback error, the intelligent heating control strategy is continuously corrected to make the intelligent heating control strategy of the actual intelligent heating system approach the optimized control strategy of the artificial intelligent heating system.

[0084] In several embodiments provided in the present application, it should be understood that the disclosed systems and methods can also be implemented in other ways. The system embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions and operations of systems, methods and computer program products according to multiple embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0085] In addition, the functional modules in each embodiment of the present invention can be integrated together to form an independent part, or each module can exist alone, or two or more modules can be integrated to form an independent part.

[0086] If a function is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this 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 of the 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 (ROMs), random access memories (RAMs), magnetic disks, or optical discs.

[0087] Taking the ideal embodiments of the present invention described above as an inspiration, through the above description, relevant staff can, without departing from the technical idea of this invention, make various changes and modifications. The technical scope of this invention is not limited to the content in the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A parallel control method for intelligent heating based on the ACP theory, characterized in that The described intelligent heating parallel control method includes: Perceiving the data information of heat sources, heat stations, heat networks, and heat users through data acquisition devices set in the actual intelligent heating system; Utilizing the data information, physical entity structure, and parameters of the actual intelligent heating system, and adopting mechanism description, knowledge representation, and machine learning methods to establish an artificial intelligent heating system; Obtaining an optimized intelligent heating control strategy through computational experiments, and inputting it into the parallel intelligent heating system formed by the combination of the actual intelligent heating system and the artificial intelligent heating system for parallel execution, continuously correcting the model structure and parameters of the artificial intelligent heating system to make the artificial intelligent heating system approach the actual intelligent heating system for the cultivation of the artificial intelligent heating system; Acting on both the actual intelligent heating system and the cultivated artificial intelligent heating system simultaneously with the optimized intelligent heating control strategy obtained through computational experiments and the state data of the initial intelligent heating system for parallel control. The artificial intelligent heating system performs synchronous computational experiments according to the state data of the actual intelligent heating system to obtain the state data of the artificial intelligent heating system. Comparing the state data of the actual intelligent heating system with the computational experiment results of the artificial intelligent heating system, feedback the error through the virtual-real interaction method, continuously correct the intelligent heating control strategy, combine the original optimized control strategy and interference information to obtain a more optimized intelligent heating control strategy. Through multiple feedback corrections by parallel control, the control strategy of the actual intelligent heating system approaches the optimized control strategy of the artificial intelligent heating system, realizing the optimized control of the actual intelligent heating system.

2. The intelligent heating parallel control method according to claim 1, wherein The perceiving the data information of heat sources, heat stations, heat networks, and heat users through data acquisition devices set in the actual intelligent heating system includes: Setting a first actual space, a second actual space, and a third actual space in the actual intelligent heating system. Obtaining the opening degree of the primary valve, primary supply water temperature, primary return water temperature, secondary supply water temperature, secondary return water temperature, circulating pump frequency, building entrance supply water temperature, and user room temperature related to heat sources, heat stations, heat networks, and heat users through the temperature acquisition device, pressure acquisition device, flow rate acquisition device, and pump valve data acquisition device set in the first actual space; obtaining meteorological data including outdoor temperature, humidity, wind speed, sunshine, rainfall, and snowfall through the second actual space; and transmitting the data information of the first actual space and the second actual space through the third actual space.

3. The intelligent heating parallel control method according to claim 2, wherein, The utilizing the data information, physical entity structure, and parameters of the actual intelligent heating system, and adopting mechanism description, knowledge representation, and machine learning methods to establish an artificial intelligent heating system includes: Setting a first artificial space in the artificial intelligent heating system for parallel information execution with the third actual space, and obtaining the data information of the actual intelligent heating system through the first artificial space; Using the data information of the actual intelligent heating system, combining the physical entity structures and parameters of heat sources, heat substations, heat networks and heat users, and aiming at the changes in the operating status of the actual intelligent heating system, adopting mechanism description, knowledge representation and machine learning methods, based on the operating mechanism model of the system itself, combining the knowledge graph model accumulated by experts and the simulation model of digital twins, constructing a software-defined computable, reconfigurable and programmable artificial intelligence heating system model, at least including a heat source model, a heat substation model, a heat network model, a heat user model, an outdoor meteorological model, a hydraulic condition model and a library of association relationships between models; Using the established artificial intelligence heating system model, by changing the model parameters and relationship weights defined by software, describing different operating states of the system, and using the multi-agent method to describe, model and evaluate the changes in different operating states, establishing a virtual artificial intelligence heating system that executes in parallel with the actual intelligent heating system.

4. The intelligent heating parallel control method according to claim 3, wherein After establishing the artificial intelligence heating system, it further includes: Using the real data tested by the actual intelligent heating system to verify the credibility of the models in the artificial intelligence heating system: using the same heat source output, heat substation control strategy, heat network parameters and meteorological data for the actual intelligent heating system and the artificial intelligence heating system, measuring and calculating the change states of the primary supply water temperature, primary return water temperature, secondary supply water temperature, secondary return water temperature, building entrance supply water temperature and user room temperature, and through comparing the real measurement data and the model output data calculated manually, correcting and optimizing the models in the artificial intelligence heating system until the data are relatively consistent, then passing the credibility verification, indicating that the artificial intelligence heating system can be used for subsequent computational experiments and parallel execution processes.

5. The intelligent heating parallel control method according to claim 1, wherein, The use of computational experiments to obtain optimized intelligent heating control strategies includes: Designing the heat source model, heat substation model, heat network model, heat user model, outdoor meteorological model and hydraulic condition model in the established artificial intelligence heating system as different intelligent agents, establishing different behavioral feature models according to the combination interaction rules of each intelligent agent, based on the behavior of a single intelligent agent and the behavior of a group of intelligent agents, establishing behavioral intelligent agents, and through reinforcement learning, each intelligent agent continuously interacts with the controlled system object in a closed-loop system to generate different system operation scenarios, and after simulation, simulation and calculation, obtaining experimental test data under different operation scenarios; Adopting data mining, feature extraction and machine learning methods, intelligent optimization algorithms and reinforcement learning algorithms to train and learn the experimental test data, obtaining optimized intelligent heating control strategies under different system operation scenarios, including heat source load optimal distribution strategies, pump valve control strategies, heat user load prediction, and heating parameter optimization strategies; Among them, the process of the computational experiment includes: determining the start time, step size, duration, initial conditions, and constraint conditions of the experiment; setting corresponding intelligent heating control strategies for different experimental scenarios; inputting the experimental conditions and the set intelligent heating control strategies into the model in the artificial intelligent heating system to execute the computational experiment process, and recording the experimental results of the system model parameters, system operation status, and the change process of the hydraulic conditions during the experiment; after the computational experiment is completed, the artificial intelligent heating system transmits the experimental results to the analysis and evaluation module to analyze and evaluate the safety, stability, economy, energy conservation, environmental protection, and user thermal comfort of the system. If the evaluation results meet the preset requirements, the intelligent heating control strategy of the experimental process is used as the optimization result; otherwise, the intelligent heating control strategy is corrected, and then it enters the next round of computational experiment process. Through multiple cyclic experiments, an optimized intelligent heating control strategy is obtained.

6. The intelligent heating parallel control method according to claim 1, characterized in that The cultivation of the artificial intelligent heating system includes: Inputting the intelligent heating control strategy and the state data of the initial intelligent heating system into the parallel intelligent heating system formed by the combination of the actual intelligent heating system and the artificial intelligent heating system. Through the parallel execution of multiple artificial intelligent heating systems and the actual intelligent heating system, obtaining the computational experiment results of multiple artificial intelligent heating systems, comparing the computational experiment results of different artificial intelligent heating systems with the results of the actual intelligent heating system to obtain the state errors of different artificial intelligent heating systems, and then feeding back the state errors to the corresponding artificial intelligent heating systems to correct the model structure and parameters of the artificial intelligent heating systems. The corrected artificial intelligent heating systems continue to operate to form a closed-loop control, making the computational experiment results of the artificial intelligent heating systems continuously approach the results of the actual intelligent heating system.

7. The intelligent heating parallel control method according to claim 2, wherein A fourth actual space is also set in the actual intelligent heating system to obtain the social behavior information of heat users and heating companies, including user activity behavior information, heating company metering and charging information, and the attribute information of the buildings of heating users; a second artificial space is also set in the artificial intelligent heating system to build a virtual space corresponding to the fourth actual space, imitate the social behavior of heat users and heating companies, and perform parallel execution.

8. The intelligent heating parallel control method according to claim 3, wherein The parallel execution between the third actual space and the first artificial space is to use the data obtained from the computational experiment and the intelligent heating control strategy to guide the third actual space. Through the parallel execution of the connection between the third actual space and the first artificial space, applying the computational experiment results in the first artificial space to the third actual space and the first artificial space, and optimizing and adjusting the first artificial space according to the feedback information composed of the operation results of the third actual space and the first artificial space and the target deviation, so as to achieve the parallel execution of virtual-real interaction; In the parallel intelligent heating system formed by the combination of the actual intelligent heating system and the artificial intelligent heating system, a parallel blockchain is also set up to securely store, verify, and securely transmit the data information transmitted by the parallel intelligent heating system; the parallel blockchain is an artificial blockchain based on the construction of a multi-mapping of the real blockchain. Through computational experiments on the artificial blockchain, the evolution law of the real blockchain is analyzed and evaluated. The data and strategies obtained through computational experiments are applied to the real blockchain and the artificial blockchain using the parallel execution method to manage and control the real blockchain.

9. A parallel control device for an intelligent heating system based on the ACP theory, characterized in that, The parallel control device of the intelligent heating system includes: An actual system data acquisition unit, which is used to sense the data information of heat sources, heat stations, heat networks, and heat users through the data acquisition devices set in the actual intelligent heating system; An artificial system establishment unit, which uses the data information, physical entity structure, and parameters of the actual intelligent heating system, and adopts mechanism description, knowledge representation, and machine learning methods to establish an artificial intelligent heating system; A computational experiment unit, which is used to obtain optimized intelligent heating control strategies through computational experiments; A parallel execution unit, which is used to simultaneously apply the optimized intelligent heating control strategies obtained through computational experiments and the state data of the initial intelligent heating system to the actual intelligent heating system and the cultivated artificial intelligent heating system for parallel control. The artificial intelligent heating system performs synchronous computational experiments according to the state data of the actual intelligent heating system to obtain the state data of the artificial intelligent heating system. The state data of the actual intelligent heating system and the computational experiment results of the artificial intelligent heating system are compared, and the error is fed back through the virtual-real interaction method to continuously correct the intelligent heating control strategy. Combining the original optimized control strategy and interference information, a more optimized intelligent heating control strategy is obtained. Through multiple feedback corrections by parallel control, the control strategy of the actual intelligent heating system approaches the optimized control strategy of the artificial intelligent heating system, realizing the optimized control of the actual intelligent heating system.

Images