Equipment control strategy generation method and device based on large model, equipment and medium

Abstract

The invention discloses an equipment control strategy generation method and device based on a large model, equipment and a medium. The method comprises the following steps: acquiring a static information set corresponding to a target building, a current control state, a current control target and external factor characteristics; performing simulation prediction on the static information set, the current control state and the external factor characteristics based on a target generation type large model to obtain current prediction load information, and determining a to-be-selected control strategy set based on a preset strategy generation rule and the current prediction load information; performing simulation prediction on the to-be-selected control strategy in the to-be-selected control strategy set based on the target generation type large model to obtain an operation parameter simulation result; and performing strategy evaluation on the operation parameter simulation result based on the current control target to generate a strategy evaluation result, and determining a target control strategy corresponding to the target building based on the strategy evaluation result. According to the technical scheme, the control strategy of the electromechanical system can be accurately generated, and the operation efficiency of the electromechanical system is improved.

Description

Technical Field

[0001] This invention relates to the field of software engineering technology, and in particular to a method, apparatus, device, and medium for generating device control strategies based on large models. Background Technology

[0002] With the gradual development of software engineering technology, automatic equipment control technology is increasingly being applied to construction and municipal engineering projects. However, as construction and municipal engineering projects continue to expand, the number of electromechanical systems involved is also increasing. Therefore, how to automatically control a large number of electromechanical systems has become extremely important.

[0003] In existing technologies, traditional feedback control methods are commonly used to achieve automatic control of electromechanical systems. However, some electromechanical systems have significant physical inertia. For example, the air conditioning system control of large public buildings can have an inertia of 1 to 2 hours; another example is the temperature control system of municipal heating, which can have an inertia of 1 to 2 days. Relying on traditional feedback control cannot adequately meet the control requirements, resulting in reduced operating efficiency of the electromechanical system.

[0004] Therefore, how to accurately generate control strategies for electromechanical systems so that they can act in advance, achieve control objectives more precisely, and improve operating efficiency is an urgent problem to be solved. Summary of the Invention

[0005] This invention provides a method, apparatus, device, and medium for generating equipment control strategies based on a large model, which can solve the problem of low accuracy of control strategies for electromechanical systems in the prior art.

[0006] According to one aspect of the present invention, a method for generating device control strategies based on a large model is provided, comprising:

[0007] Acquire the set of static information, current control status, current control objectives, and external factor characteristics corresponding to the target building;

[0008] Based on the target generative large model, the static information set, current control state and external factor characteristics are simulated and predicted to obtain the current predicted load information corresponding to the target building, and the set of candidate control strategies corresponding to the target building is determined based on the preset strategy generation rules and the current predicted load information.

[0009] Based on the target generative large model, the candidate control strategies in the candidate control strategy set are simulated and predicted to obtain the simulation results of the operating parameters of the target building under the candidate control strategies.

[0010] Based on the current control objective, the simulation results of the operating parameters are evaluated to generate a strategy evaluation result. Based on the strategy evaluation result, the target control strategy corresponding to the target building is determined from the set of candidate control strategies.

[0011] According to another aspect of the present invention, a device control strategy generation apparatus based on a large model is provided, comprising:

[0012] The data acquisition module is used to acquire the static information set, current control status, current control target and external factor characteristics of the target building;

[0013] The strategy generation module is used to simulate and predict the static information set, current control state and external factor characteristics based on the target generative large model to obtain the current predicted load information corresponding to the target building, and to determine the set of candidate control strategies corresponding to the target building based on the preset strategy generation rules and the current predicted load information.

[0014] The strategy simulation module is used to simulate and predict the candidate control strategies in the candidate control strategy set based on the target generative large model, and obtain the simulation results of the operating parameters of the target building under the candidate control strategies.

[0015] The strategy evaluation module is used to evaluate the simulation results of the operating parameters based on the current control objective, generate strategy evaluation results, and determine the target control strategy corresponding to the target building from the set of candidate control strategies based on the strategy evaluation results.

[0016] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0017] At least one processor; and

[0018] A memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to execute the device control strategy generation method based on a large model as described in any embodiment of the present invention.

[0020] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the device control strategy generation method based on a large model as described in any embodiment of the present invention.

[0021] According to another aspect of the present invention, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the device control strategy generation method based on a large model as described in any embodiment of the present invention.

[0022] The technical solution of this invention involves acquiring a set of static information, current control state, current control objective, and external factor characteristics corresponding to a target building. Then, based on a target-generative large-scale model, simulation and prediction are performed on the static information set, current control state, and external factor characteristics to obtain the current predicted load information corresponding to the target building. Based on preset strategy generation rules and the current predicted load information, a set of candidate control strategies corresponding to the target building is determined. Further, based on the target-generative large-scale model, simulation and prediction are performed on the candidate control strategies in the set of candidate control strategies to obtain the simulation results of the operating parameters corresponding to the target building under the candidate control strategies. Finally, based on the current control objective, strategy evaluation is performed on the simulation results of the operating parameters to generate strategy evaluation results. Based on the strategy evaluation results, the target control strategy corresponding to the target building is determined from the set of candidate control strategies. By applying a generative large-scale model to the future state simulation stage, the robustness of the entire simulation and optimization control process is enhanced. It can better accommodate various situations where data is missing in the project, solves the problem of low accuracy of control strategies for electromechanical systems in the prior art, and can accurately generate control strategies for electromechanical systems, enabling the electromechanical system to act in advance and more accurately achieve the control objective, thereby improving the operating efficiency of the electromechanical system.

[0023] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart of a device control strategy generation method based on a large model according to Embodiment 1 of the present invention;

[0026] Figure 2 This is a flowchart of a device control strategy generation method based on a large model according to Embodiment 2 of the present invention;

[0027] Figure 3This is a schematic diagram of a device control strategy generation device based on a large model according to Embodiment 3 of the present invention;

[0028] Figure 4 This is a schematic diagram of the structure of an electronic device that implements the device control strategy generation method based on a large model according to an embodiment of the present invention. Detailed Implementation

[0029] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0030] It should be noted that the terms "target," "option," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0031] Example 1

[0032] Figure 1 This is a flowchart of a device control strategy generation method based on a large model, provided in Embodiment 1 of the present invention. This embodiment is applicable to situations requiring advance control of electromechanical systems. The method can be executed by a device control strategy generation device based on a large model. This device can be implemented in hardware and / or software and can be configured in electronic devices, for example, in computer devices. Figure 1 As shown, the method includes:

[0033] S110. Obtain the set of static information, current control status, current control target and external factor characteristics corresponding to the target building.

[0034] The target building can refer to a building containing the electromechanical system for which the selected equipment control strategy is determined. For example, the target building can be a house, building, or factory. Typically, the electromechanical system requiring control strategy determination is determined based on actual application needs, and the building containing that electromechanical system is designated as the target building. Static information can refer to the basic attribute data of a building that is relatively fixed and does not frequently change with time or usage status. The static information set can refer to the collection of various static information corresponding to the same target building. Optionally, the static information set can include: building information, spatial information, equipment information, and connection relationship information. Building information can refer to the attribute, status, and characteristic data of the physical components or systems themselves within the building. For example, building information can include function, area, city, climate zone, shape coefficient, and window-to-wall ratio. Spatial information can refer to information about defined and managed areas, rooms, or locations within the building. For example, spatial information can include function, area, orientation, internal and external zones, and daily routines. Equipment information can refer to a detailed data set of various electromechanical equipment, instruments, fixtures, and key assets installed within the building. For example, equipment information may include type, rated power, rated flow rate, and rated temperature. Connection information may refer to how different components, equipment, systems, and spaces within a building are interconnected, related, dependent on, or influence each other. For example, connection information may include equipment piping connections and equipment spatial connections.

[0035] In this context, "control state" refers to the operating mode or stage of an electromechanical system under specific control logic. Typically, control state describes how the controller makes decisions and outputs instructions to influence the behavior of controlled objects, such as motors, valves, and air conditioning units. For example, control state may include the number of equipment turned on, the number of equipment turned off, the water temperature and flow rate of the turned-on equipment, control results, and control instructions. "Current control state" refers to the control state of the electromechanical system within the target building during the current time period. "Control objective" refers to the evaluation criteria used to assess the operating state of the electromechanical system. "Current control objective" refers to the control objective corresponding to the current time period. Typically, the current control objective can be adjusted according to actual application requirements. Optionally, the current control objective may include boundary constraints, cost constraints, and deviation constraints. Boundary constraints refer to conditions that impose limit values ​​on the operating parameters of the electromechanical system. For example, boundary constraints may include maximum temperature limits, minimum temperature limits, high temperature duration, and low temperature duration. Cost constraints refer to conditions that constrain the operating cost of the electromechanical system. For example, cost constraints may include minimum energy consumption, minimum cost, and minimum carbon emissions. Deviation constraints refer to conditions used to constrain the degree to which the operating results of an electromechanical system deviate from boundary constraints. For example, a deviation constraint could be: when energy consumption, cost, or carbon emissions are roughly the same, the lower the deviation from the boundary constraints, the better. External factor characteristics refer to the attributes and manifestations of various conditions that significantly affect a building but originate outside its boundaries and are not directly controlled by it. For example, external factor characteristics could include outdoor temperature, outdoor humidity, solar radiation intensity, electricity price, and carbon price.

[0036] In this embodiment of the invention, the target building can typically be determined based on a pre-built data platform, and the static information set, current control status, current control target and external factor characteristics corresponding to the target building can be obtained. This embodiment of the invention does not impose specific limitations on this.

[0037] S120. Based on the target-generating large model, the static information set, the current control state and external factor characteristics are simulated and predicted to obtain the current predicted load information corresponding to the target building, and the set of candidate control strategies corresponding to the target building is determined based on the preset strategy generation rules and the current predicted load information.

[0038] In this context, a generative large model can refer to a giant artificial intelligence model capable of autonomously creating new content. A target generative large model can refer to a pre-trained, final generative large model. Typically, a target generative large model can be used to implement data simulation. It is worth noting that in this embodiment of the invention, the training process of the target generative large model is not limited. Load information can refer to quantified data on the energy, power, or service demands required by an electromechanical system within a specific time period. For example, load information can be equipment operating power and energy consumption. Predicted load information can refer to load information for future time periods generated after simulating and predicting input data using the target generative large model. For example, predicted load information can be load information every 15 minutes within the next 24 hours. Current predicted load information can refer to the predicted load information corresponding to the current time period.

[0039] The preset strategy generation rule refers to a pre-defined rule used to limit the generation process of equipment control strategies. For example, the preset strategy generation rule may include a specific equipment control strategy generation process. A control strategy refers to a scheme used to control and adjust parameters such as the on / off time, number of units, initial activation frequency, and outlet water temperature setpoint of equipment. Typically, control strategies are not real-time but rather forward-looking, such as strategy schemes every 15 minutes for the next 4 hours. Candidate control strategies refer to initially generated equipment control strategies that have not yet been evaluated or verified. The set of candidate control strategies refers to a collection of various candidate control strategies corresponding to the same target building.

[0040] S130. Based on the target generative large model, simulate and predict the candidate control strategies in the candidate control strategy set to obtain the simulation results of the operating parameters corresponding to the target building under the candidate control strategies.

[0041] The simulation results of operating parameters can refer to the predicted equipment operating status generated by simulating and predicting the static information set, current control state, and external factor characteristics under the selected control strategy using a target-generative large model. For example, the simulation results of operating parameters can include the energy consumption, cost, and carbon emissions required for equipment operation under the selected control strategy.

[0042] S140. Based on the current control objective, perform strategy evaluation on the simulation results of the operating parameters, generate strategy evaluation results, and determine the target control strategy corresponding to the target building from the set of candidate control strategies based on the strategy evaluation results.

[0043] The strategy evaluation result can refer to the judgment result generated after judging the simulation results of operating parameters using the current control objective. The target control strategy can refer to the equipment control strategy ultimately selected to implement control over the electromechanical system.

[0044] The technical solution of this invention involves acquiring a set of static information, current control state, current control objective, and external factor characteristics corresponding to a target building. Then, based on a target-generative large-scale model, simulation and prediction are performed on the static information set, current control state, and external factor characteristics to obtain the current predicted load information corresponding to the target building. Based on preset strategy generation rules and the current predicted load information, a set of candidate control strategies corresponding to the target building is determined. Further, based on the target-generative large-scale model, simulation and prediction are performed on the candidate control strategies in the set of candidate control strategies to obtain the simulation results of the operating parameters corresponding to the target building under the candidate control strategies. Finally, based on the current control objective, strategy evaluation is performed on the simulation results of the operating parameters to generate strategy evaluation results. Based on the strategy evaluation results, the target control strategy corresponding to the target building is determined from the set of candidate control strategies. By applying a generative large-scale model to the future state simulation stage, the robustness of the entire simulation and optimization control process is enhanced. It can better accommodate various situations where data is missing in the project, solves the problem of low accuracy of control strategies for electromechanical systems in the prior art, and can accurately generate control strategies for electromechanical systems, enabling the electromechanical system to act in advance and more accurately achieve the control objective, thereby improving the operating efficiency of the electromechanical system.

[0045] Example 2

[0046] Figure 2 This is a flowchart of a method for generating equipment control strategies based on a large model, provided in Embodiment 2 of the present invention. This embodiment is a refinement of the above embodiment. Specifically, it refines the operation of determining the set of candidate control strategies corresponding to the target building based on preset strategy generation rules and the current predicted load information. Specifically, it may include: obtaining historical predicted load information corresponding to the target building, numerically comparing the historical predicted load information and the current predicted load information to generate a numerical comparison result; determining the master control strategy corresponding to the target building based on the numerical comparison result and a preset master equipment adjustment strategy; determining the slave equipment control strategy corresponding to the target building based on a preset slave equipment adjustment strategy and the master control strategy; and combining the master control strategy and the slave equipment control strategy to obtain the set of candidate control strategies corresponding to the target building. Figure 2 As shown, the method includes:

[0047] S210. Obtain the set of static information, current control status, current control target and external factor characteristics corresponding to the target building.

[0048] S220. Based on the target generative large model, the static information set, current control state and external factor characteristics are simulated and predicted to obtain the current predicted load information corresponding to the target building.

[0049] Specifically, after obtaining the static information set, current control state, and external factor characteristics corresponding to the target building, the static information set, current control state, and external factor characteristics can be input into the target generative large model. The target generative large model can then simulate and predict the input data to obtain the current predicted load information of each electromechanical system in the target building.

[0050] S230. Obtain the historical predicted load information corresponding to the target building, and numerically compare the historical predicted load information with the current predicted load information to generate a numerical comparison result.

[0051] Historical predicted load information refers to load information predicted for the target building within a historical time period. For example, historical predicted load information could be the predicted load information corresponding to the previous data collection time, such as the load information predicted for the target building fifteen minutes ago. Numerical comparison refers to the operation of calculating the difference between historical predicted load information and current predicted load information. The numerical comparison result refers to the numerical result obtained after comparing historical and current predicted load information. For example, the numerical comparison result can be the difference between historical and current predicted load information.

[0052] S240. Based on the numerical comparison results and the preset master equipment adjustment strategy, determine the host control strategy corresponding to the target building.

[0053] The preset master device adjustment strategy can refer to a pre-set strategy for adjusting the number of master devices. For example, the preset master device adjustment strategy could be based on the current number of master devices, allowing for a fluctuation of one unit up or down, and when switching master strategies, the number of master devices should not skip levels; that is, two master devices cannot become four master devices in the next moment, but can become three master devices. The master device control strategy can refer to a strategy used to control the master devices in the electromechanical system. For example, the master device control strategy can be the number of master devices that can operate in the electromechanical system.

[0054] In an optional implementation, determining the host control strategy corresponding to the target building based on the numerical comparison results and the preset host equipment adjustment strategy includes:

[0055] Step a1: If the numerical comparison result meets the preset load threshold, then obtain the number of historical hosts corresponding to the historical predicted load information, and adjust the number of historical hosts based on the preset master device adjustment strategy to obtain the host control strategy corresponding to the target building.

[0056] The preset load threshold can refer to a pre-set value used to evaluate the numerical comparison results. For example, the preset load threshold could be ±30%. The number of hosts can refer to the number of operational host devices. The historical number of hosts can refer to the number of hosts corresponding to historical predicted load information.

[0057] Step a2: If the numerical comparison result exceeds the preset load threshold, then the current predicted load information is calculated based on the static information set to determine the number of baseline hosts corresponding to the target building, and the number of baseline hosts is adjusted based on the preset master equipment adjustment strategy to obtain the host control strategy corresponding to the target building.

[0058] The baseline number of hosts can refer to the number of hosts calculated using the device information and the current predicted load information in the static information set. For example, if the rated power of the hosts in the device information is 600kW and the current predicted load information is 1500kW, then the current predicted load information is used to perform an up-rounding division on the rated power of the hosts, ultimately resulting in a baseline number of 3 hosts.

[0059] Specifically, if the current predicted load information does not differ significantly from historical predicted load information, the current number of operating hosts can be used as a baseline, and the number of hosts can be adjusted by fluctuating up or down by one unit according to a preset master equipment adjustment strategy, thus generating a master control strategy. Conversely, if the current predicted load information differs significantly from historical predicted load information, the required baseline number of hosts can be calculated based on the current predicted load information, and the number of hosts can be adjusted by fluctuating up or down by one unit according to a preset master equipment adjustment strategy, thus generating a master control strategy. This master control strategy provides a valid basis for subsequent operations.

[0060] It is worth noting that when generating host control policies, the duration of each policy must not be too short, and the duration should not be shorter than a certain set time. That is, adjustments should only be made after a policy has reached relative stability. Therefore, it is necessary to ensure that the time interval between two consecutive changes to the host control policy is greater than one hour. This embodiment of the invention will not elaborate further on this.

[0061] S250. Determine the slave device control strategy corresponding to the target building based on the preset slave device adjustment strategy and the host control strategy.

[0062] The preset slave device adjustment strategy refers to a pre-defined strategy for adjusting the number of slave devices. For example, the preset slave device adjustment strategy can be to use a floating set number of units according to the host control strategy as the number of slave devices. Specifically, taking a chiller as the host and water pumps as the slave devices as an example, the preset slave device adjustment strategy can be as shown in the table below:

[0063] Table 1 Preset Adjustment Strategies from Devices

[0064] Current refrigeration strategy test1 test2 test3 M large 0 small +1 large pump +2 large pumps +3 large pumps 0 large N small +1 small pump +2 small pumps M large N small +1 large pump +1 large pump and 1 small pump +2 large pumps and 1 small pump

[0065] Here, the current chiller strategy can represent the host control strategy. N can represent the number of low-power chillers, M can represent the number of high-power chillers, and test1, test2, and test3 can represent different fluctuation scenarios.

[0066] Here, the slave device control strategy refers to the strategy used to control slave devices in an electromechanical system. For example, the slave device control strategy can be the number of operational slave devices in the electromechanical system. Specifically, continuing the example above, if the master control strategy is M large 0 small, the corresponding slave device control strategy can be M+1 large pumps, M+2 large pumps, or M+3 large pumps. Similarly, if the master control strategy is 0 large N small, the corresponding slave device control strategy can be N+1 small pumps or N+2 small pumps. If the master control strategy is M large N small, the corresponding slave device control strategy can be M+1 large pumps, M+1 large pumps N+1 small pumps, or M+2 large pumps N+1 small pumps.

[0067] S260. Combine the host control strategy and the slave device control strategy to obtain a set of candidate control strategies corresponding to the target building.

[0068] Specifically, after obtaining the current predicted load information for the target building, the current predicted load information can be numerically compared with the historical predicted load information for the target building to generate a numerical comparison result. Then, based on the numerical comparison result and a preset master equipment adjustment strategy, the master control strategy for the target building is determined. Further, based on a preset slave equipment adjustment strategy and the master control strategy, the slave equipment control strategy for the target building is determined. Finally, the master control strategy and the corresponding slave equipment control strategy are combined and processed. This yields a set of candidate control strategies for the target building, providing a valid basis for subsequent operations.

[0069] Typically, at each time point, such as hourly or every 15 minutes, 3 to 5 sets of slave device control strategies can be generated. In this embodiment of the invention, it is preferable to generate a set of candidate control strategies every 15 minutes. Furthermore, based on the above embodiments, this embodiment of the invention can also add the initial operating state of the slave devices to the set of candidate control strategies to ensure the smooth operation of subsequent work. Specifically, taking a chiller as the master device in the same electromechanical system, and air conditioners, chillers, chilled water pumps, cooling pumps, and cooling towers as examples. The strategy parameters for the host can be the host control strategy; the strategy parameters for the air conditioner can be the air conditioner demand time (i.e., slave device control strategy) and the indoor temperature setpoint (i.e., initial operating state); the strategy parameters for the chiller can be the chiller strategy (i.e., slave device control strategy) and the outlet water temperature setpoint (i.e., initial operating state); the strategy parameters for the chilled pump can be the chilled pump strategy (i.e., slave device control strategy) and the initial switching frequency (i.e., initial operating state); the strategy parameters for the cooling pump can be the cooling pump strategy (i.e., slave device control strategy) and the initial switching frequency (i.e., initial operating state); and the strategy parameters for the cooling tower can be the cooling tower strategy (i.e., slave device control strategy), the initial switching frequency, the initial switching gear, and the initial number of units switched (i.e., initial operating state). The device control strategies are sequentially stored, generating one data point every 15 minutes. The final set of candidate control strategies is shown in the table below.

[0070] Table 2 Set of Control Strategies to be Selected

[0071]

[0072] S270. Based on the target generative large model, simulate and predict the candidate control strategies in the candidate control strategy set to obtain the simulation results of the operating parameters corresponding to the target building under the candidate control strategies.

[0073] Specifically, after obtaining the set of candidate control strategies for the target building, each candidate control strategy in the set of candidate control strategies, along with the set of static information, the current control state, and the characteristics of external factors, are input into the target generative large model. The target generative large model is then used for simulation and prediction to obtain the simulation results of the operating parameters of the target building under each candidate control strategy.

[0074] S280. Based on the current control objective, perform a strategy evaluation on the simulation results of the operating parameters and generate a strategy evaluation result.

[0075] Specifically, after obtaining the simulation results of the operating parameters, the current control objectives can be used to evaluate the simulation results. For example, boundary constraints can be used to determine whether the simulation results meet the maximum temperature limit, minimum temperature limit, high temperature duration, and low temperature duration specified in the boundary constraints. Cost constraints can be used to determine whether the simulation results represent the lowest energy consumption, lowest cost, and lowest carbon emissions. Simultaneously, deviation constraints can be used to assess the degree of deviation of the simulation results from the boundary constraints.

[0076] S290. If the strategy evaluation result is that the simulation result of the operating parameters meets the current control objective, then the candidate control strategy corresponding to the simulation result of the operating parameters shall be used as the target control strategy corresponding to the target building.

[0077] Specifically, if the strategy evaluation result of the simulation of the operating parameters satisfies the boundary constraints, cost constraints, and deviation constraints, then the candidate control strategy corresponding to the simulation result of the operating parameters can be used as the target control strategy for the target building.

[0078] S2100: The target control strategy is sent to the control execution system based on a preset time period, so that the control execution system executes the corresponding control action according to the target control strategy.

[0079] The preset time period can refer to a pre-defined time value used to limit the data transmission cycle. For example, the preset time period can be thirty minutes. The control execution system can refer to a computer system used to implement the target control strategy.

[0080] Specifically, after obtaining the target control strategy corresponding to the target building, the target control strategy can be sent to the control execution system at preset time intervals. The control execution system then converts the target control strategy into control commands and interfaces them with the specific control hardware, thus completing the equipment control process of the electromechanical system.

[0081] The technical solution of this invention involves acquiring a set of static information, current control state, current control objective, and external factor characteristics corresponding to a target building. Then, based on a target-generative large-scale model, simulation and prediction are performed on the static information set, current control state, and external factor characteristics to obtain the current predicted load information corresponding to the target building. Further, historical predicted load information corresponding to the target building is acquired, and the historical predicted load information and current predicted load information are numerically compared to generate a numerical comparison result. Based on the numerical comparison result and a preset master equipment adjustment strategy, a master control strategy corresponding to the target building is determined. Based on a preset slave equipment adjustment strategy and the master control strategy, a slave equipment control strategy corresponding to the target building is determined. The master control strategy and slave equipment control strategy are then combined to obtain a set of candidate control strategies corresponding to the target building. Further, based on the target-generative large-scale model, simulation and prediction are performed on the candidate control strategies in the set of candidate control strategies to obtain the simulation results of the operating parameters corresponding to the target building under the candidate control strategies. Based on the current control objective, the simulation results of the operating parameters are evaluated to generate a strategy evaluation result. If the strategy evaluation result indicates that the simulation results of the operating parameters meet the current control objective, then the candidate control strategy corresponding to the simulation results of the operating parameters is adopted as the target control strategy for the target building. Finally, the target control strategy is sent to the control execution system based on a preset time period, so that the control execution system can execute the corresponding control actions according to the target control strategy. By applying a generative large model to the future state simulation stage, the robustness of the entire simulation and optimization control process is enhanced. It can better accommodate various situations where data is missing in the project, and solves the problem of low accuracy of control strategies for electromechanical systems in existing technologies. It can accurately generate control strategies for electromechanical systems, enabling them to act in advance and achieve control objectives more precisely, thereby improving the operating efficiency of the electromechanical system.

[0082] Example 3

[0083] Figure 3 This is a schematic diagram of a device control strategy generation device based on a large model, provided in Embodiment 3 of the present invention. Figure 3 As shown, the device includes: a data acquisition module 310, a strategy generation module 320, a strategy simulation module 330, and a strategy evaluation module 340;

[0084] Among them, the data acquisition module 310 is used to acquire the static information set, current control status, current control target and external factor characteristics corresponding to the target building;

[0085] The strategy generation module 320 is used to simulate and predict the static information set, current control state and external factor characteristics based on the target generative large model, to obtain the current predicted load information corresponding to the target building, and to determine the set of candidate control strategies corresponding to the target building based on the preset strategy generation rules and the current predicted load information.

[0086] The strategy simulation module 330 is used to simulate and predict the candidate control strategies in the candidate control strategy set based on the target generative large model, and obtain the simulation results of the operating parameters of the target building under the candidate control strategy.

[0087] The strategy evaluation module 340 is used to evaluate the simulation results of the operating parameters based on the current control objective, generate strategy evaluation results, and determine the target control strategy corresponding to the target building from the set of candidate control strategies based on the strategy evaluation results.

[0088] The technical solution of this invention involves acquiring a set of static information, current control state, current control objective, and external factor characteristics corresponding to a target building. Then, based on a target-generative large-scale model, simulation and prediction are performed on the static information set, current control state, and external factor characteristics to obtain the current predicted load information corresponding to the target building. Based on preset strategy generation rules and the current predicted load information, a set of candidate control strategies corresponding to the target building is determined. Further, based on the target-generative large-scale model, simulation and prediction are performed on the candidate control strategies in the set of candidate control strategies to obtain the simulation results of the operating parameters corresponding to the target building under the candidate control strategies. Finally, based on the current control objective, strategy evaluation is performed on the simulation results of the operating parameters to generate strategy evaluation results. Based on the strategy evaluation results, the target control strategy corresponding to the target building is determined from the set of candidate control strategies. By applying a generative large-scale model to the future state simulation stage, the robustness of the entire simulation and optimization control process is enhanced. It can better accommodate various situations where data is missing in the project, solves the problem of low accuracy of control strategies for electromechanical systems in the prior art, and can accurately generate control strategies for electromechanical systems, enabling the electromechanical system to act in advance and more accurately achieve the control objective, thereby improving the operating efficiency of the electromechanical system.

[0089] Optional, static information set, including: building information, spatial information, equipment information, and connection relationship information.

[0090] Optionally, the policy generation module 320 can be used for:

[0091] Obtain the historical predicted load information corresponding to the target building, and numerically compare the historical predicted load information with the current predicted load information to generate a numerical comparison result;

[0092] Based on the numerical comparison results and the preset master equipment adjustment strategy, the host control strategy corresponding to the target building is determined.

[0093] The slave device control strategy corresponding to the target building is determined based on the preset slave device adjustment strategy and the host control strategy.

[0094] The host control strategy and the slave device control strategy are combined to obtain a set of candidate control strategies corresponding to the target building.

[0095] Optionally, the policy generation module 320 can be used for:

[0096] If the numerical comparison result meets the preset load threshold, then the number of historical hosts corresponding to the historical predicted load information is obtained, and the number of historical hosts is adjusted based on the preset master device adjustment strategy to obtain the host control strategy corresponding to the target building.

[0097] If the numerical comparison result exceeds the preset load threshold, then the current predicted load information is calculated based on the static information set to determine the number of baseline hosts corresponding to the target building, and the number of baseline hosts is adjusted based on the preset master equipment adjustment strategy to obtain the host control strategy corresponding to the target building.

[0098] Optional, the current control objectives include: boundary constraints, cost constraints, and deviation constraints.

[0099] Optionally, the strategy evaluation module 340 can be used to: if the strategy evaluation result is that the simulation result of the operating parameters meets the current control objective, then the candidate control strategy corresponding to the simulation result of the operating parameters is used as the target control strategy corresponding to the target building.

[0100] Optionally, the device control strategy generation apparatus based on a large model may further include: a strategy sending module, used to send the target control strategy to the control execution system based on a preset time period after the target control strategy corresponding to the target building is determined from the set of candidate control strategies based on the strategy evaluation results, so that the control execution system performs corresponding control actions according to the target control strategy.

[0101] The device control strategy generation apparatus based on a large model provided in the embodiments of the present invention can execute the device control strategy generation method based on a large model provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0102] Example 4

[0103] Figure 4A schematic diagram of an electronic device 410 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0104] like Figure 4 As shown, the electronic device 410 includes at least one processor 420 and a memory, such as a read-only memory (ROM) 430 or a random access memory (RAM) 440, communicatively connected to the at least one processor 420. The memory stores computer programs executable by the at least one processor. The processor 420 can perform various appropriate actions and processes based on the computer program stored in the ROM 430 or loaded into the RAM 440 from storage unit 490. The RAM 440 may also store various programs and data required for the operation of the electronic device 410. The processor 420, ROM 430, and RAM 440 are interconnected via a bus 450. An input / output (I / O) interface 460 is also connected to the bus 450.

[0105] Multiple components in electronic device 410 are connected to I / O interface 460, including: input unit 470, such as keyboard, mouse, etc.; output unit 480, such as various types of monitors, speakers, etc.; storage unit 490, such as disk, optical disk, etc.; and communication unit 4100, such as network card, modem, wireless transceiver, etc. Communication unit 4100 allows electronic device 410 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0106] Processor 420 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 420 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 420 performs the various methods and processes described above, such as methods for generating device control policies based on large models.

[0107] The method includes:

[0108] Acquire the set of static information, current control status, current control objectives, and external factor characteristics corresponding to the target building;

[0109] Based on the target generative large model, the static information set, current control state and external factor characteristics are simulated and predicted to obtain the current predicted load information corresponding to the target building, and the set of candidate control strategies corresponding to the target building is determined based on the preset strategy generation rules and the current predicted load information.

[0110] Based on the target generative large model, the candidate control strategies in the candidate control strategy set are simulated and predicted to obtain the simulation results of the operating parameters of the target building under the candidate control strategies.

[0111] Based on the current control objective, the simulation results of the operating parameters are evaluated to generate a strategy evaluation result. Based on the strategy evaluation result, the target control strategy corresponding to the target building is determined from the set of candidate control strategies.

[0112] In some embodiments, the large-model-based device control policy generation method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 490. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 410 via ROM 430 and / or communication unit 4100. When the computer program is loaded into RAM 440 and executed by processor 420, one or more steps of the large-model-based device control policy generation method described above may be performed. Alternatively, in other embodiments, processor 420 may be configured to execute the large-model-based device control policy generation method by any other suitable means (e.g., by means of firmware).

[0113] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0114] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0115] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0116] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0117] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0118] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0119] This application also discloses a computer program product, which includes a computer program that, when executed by a processor, implements the device control strategy generation method based on a large model provided in any embodiment of this application. This program product shares the same inventive concept as the device control strategy generation method based on a large model disclosed in the embodiments of this application, and therefore will not be described further here.

[0120] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0121] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for generating equipment control strategies based on a large model, characterized in that, include: Acquire the set of static information, current control status, current control objectives, and external factor characteristics corresponding to the target building; Based on the target generative large model, the static information set, current control state and external factor characteristics are simulated and predicted to obtain the current predicted load information corresponding to the target building, and the set of candidate control strategies corresponding to the target building is determined based on the preset strategy generation rules and the current predicted load information. Based on the target generative large model, the candidate control strategies in the candidate control strategy set are simulated and predicted to obtain the simulation results of the operating parameters of the target building under the candidate control strategies. Based on the current control objective, the simulation results of the operating parameters are evaluated to generate a strategy evaluation result. Based on the strategy evaluation result, the target control strategy corresponding to the target building is determined from the set of candidate control strategies.

2. The method according to claim 1, characterized in that, The static information set includes: building information, spatial information, equipment information, and connection relationship information.

3. The method according to claim 1, characterized in that, The process of determining the set of candidate control strategies for the target building based on preset strategy generation rules and the current predicted load information includes: Obtain the historical predicted load information corresponding to the target building, and numerically compare the historical predicted load information with the current predicted load information to generate a numerical comparison result; Based on the numerical comparison results and the preset master equipment adjustment strategy, the host control strategy corresponding to the target building is determined. The slave device control strategy corresponding to the target building is determined based on the preset slave device adjustment strategy and the host control strategy. The host control strategy and the slave device control strategy are combined to obtain a set of candidate control strategies corresponding to the target building.

4. The method according to claim 3, characterized in that, The step of determining the host control strategy corresponding to the target building based on the numerical comparison results and the preset host equipment adjustment strategy includes: If the numerical comparison result meets the preset load threshold, then the number of historical hosts corresponding to the historical predicted load information is obtained, and the number of historical hosts is adjusted based on the preset master device adjustment strategy to obtain the host control strategy corresponding to the target building. If the numerical comparison result exceeds the preset load threshold, then the current predicted load information is calculated based on the static information set to determine the number of baseline hosts corresponding to the target building, and the number of baseline hosts is adjusted based on the preset master equipment adjustment strategy to obtain the host control strategy corresponding to the target building.

5. The method according to claim 1, characterized in that, The current control objectives include: boundary constraints, cost constraints, and deviation constraints.

6. The method according to claim 1, characterized in that, The step of determining the target control strategy corresponding to the target building from the set of candidate control strategies based on the strategy evaluation results includes: If the strategy evaluation result indicates that the simulation result of the operating parameters meets the current control objective, then the candidate control strategy corresponding to the simulation result of the operating parameters will be used as the target control strategy corresponding to the target building.

7. The method according to claim 1, characterized in that, After determining the target control strategy corresponding to the target building from the set of candidate control strategies based on the strategy evaluation results, the method further includes: The target control strategy is sent to the control execution system based on a preset time period, so that the control execution system can perform corresponding control actions according to the target control strategy.

8. A device for generating equipment control strategies based on a large model, characterized in that, include: The data acquisition module is used to acquire the static information set, current control status, current control target and external factor characteristics of the target building; The strategy generation module is used to simulate and predict the static information set, current control state and external factor characteristics based on the target generative large model to obtain the current predicted load information corresponding to the target building, and to determine the set of candidate control strategies corresponding to the target building based on the preset strategy generation rules and the current predicted load information. The strategy simulation module is used to simulate and predict the candidate control strategies in the candidate control strategy set based on the target generative large model, and obtain the simulation results of the operating parameters of the target building under the candidate control strategies. The strategy evaluation module is used to evaluate the simulation results of the operating parameters based on the current control objective, generate strategy evaluation results, and determine the target control strategy corresponding to the target building from the set of candidate control strategies based on the strategy evaluation results.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the device control strategy generation method based on a large model as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the device control strategy generation method based on a large model as described in any one of claims 1-7.

Images