A flame storage combined frequency modulation model predictive control method based on digital twinning

Abstract

The application discloses a firelight energy storage combined frequency modulation model prediction control method based on digital twinning, and relates to the technical field of power system frequency modulation control, wherein the method comprises the following steps: constructing a thermal power-photovoltaic-energy storage combined frequency modulation dynamic model and a system frequency dynamic response overall model; constructing a digital twinning prediction model based on the combined frequency modulation dynamic model, and fusing multi-source data to generate future frequency modulation power demand; then, combining the frequency dynamic response model to construct a combined frequency modulation optimization model, and solving a multi-source power distribution optimal solution; the solution is sent to each system as an instruction, and the running state is fed back to the digital twinning model to realize closed-loop rolling optimization. The application realizes multi-source collaborative robust optimization by explicitly considering the thermal power and energy storage constraints, and solves the problem of insufficient feasibility of frequency modulation strategies.

Description

Technical Field

[0001] This invention relates to the field of power system frequency regulation control technology, and in particular to a predictive control method and device based on a digital twin-based combined thermal-solar-storage frequency regulation model. Background Technology

[0002] As the installed capacity and penetration rate of new energy sources such as photovoltaics in the power system continue to increase, the inherent volatility, intermittency, and randomness of their power output are becoming increasingly prominent, causing significant disturbances to the power grid's power balance and frequency regulation, posing a severe challenge to the safe and stable operation of the power grid. Traditional thermal power units are limited by the physical characteristics of equipment such as boilers and turbines, resulting in slow ramp-up rates and large adjustment inertia, making it impossible to quickly smooth out high-frequency and large-amplitude power fluctuations. Although energy storage systems have millisecond-level response capabilities, their adjustable capacity is strictly constrained by the state of charge (SOC), and frequent charging and discharging will exacerbate battery wear and lifespan degradation, making it difficult for them to independently undertake frequency regulation tasks for extended periods and under high intensity.

[0003] Existing frequency regulation control methods mostly employ traditional PID control or fixed logic rule control, resulting in relatively simple control structures and weak adaptive capabilities. These methods struggle to guarantee control effectiveness in grid-connected environments with high renewable energy penetration and complex disturbances. Some studies have attempted to introduce intelligent algorithms such as reinforcement learning, but these generally suffer from problems such as large training sample requirements, insufficient model generalization ability, and poor online deployment stability, failing to meet the needs of practical engineering applications. Therefore, there is an urgent need for a combined thermal-solar-storage frequency regulation control method that can integrate predictive information, consider multi-source operational constraints, and achieve rolling optimization to improve grid frequency stability and operational economy. Summary of the Invention

[0004] The main objective of this invention is to provide a predictive control method based on a digital twin-based frequency modulation model for combined thermal, solar, and energy storage systems.

[0005] Another objective of this invention is to propose a predictive control device based on a digital twin-based frequency modulation model for combined thermal, solar, and energy storage.

[0006] The third objective of this invention is to provide an electronic device.

[0007] The fourth objective of this invention is to provide a non-transitory computer-readable storage medium.

[0008] To achieve the above objectives, a first aspect of the present invention proposes a predictive control method based on a digital twin-based joint frequency modulation model of thermal-solar-storage system, comprising:

[0009] A joint frequency regulation dynamic model of a thermal power-photovoltaic-energy storage system is constructed to analyze the frequency response characteristics of thermal power units, photovoltaic power generation units and energy storage systems, and to establish an overall frequency dynamic response model of the thermal power-photovoltaic-energy storage joint system. Based on the aforementioned joint frequency regulation dynamic model, a digital twin prediction model of the combined thermal-solar-storage system is constructed, which integrates historical operating data and new energy prediction information to generate the frequency regulation power demand in the future prediction time domain. Based on the predicted frequency regulation power demand and combined with the overall frequency dynamic response model of the combined thermal-solar-storage system, a model-based predictive control combined frequency regulation power optimization model is constructed to solve the optimal solution of multi-source frequency regulation power allocation in the prediction time domain. The optimal solution for multi-source frequency regulation power allocation is used as the optimal frequency regulation power command and sent to thermal power units, photovoltaic power generation units and energy storage systems. The real-time operating status after execution is fed back to the digital twin prediction model to realize the closed-loop rolling optimization operation of the joint frequency regulation of thermal power, photovoltaic and energy storage.

[0010] Optionally, a joint frequency regulation dynamic model of the thermal power-photovoltaic-energy storage system is constructed, including: The overall operating characteristics of the thermal-solar-storage integrated system were analyzed to determine the functional positioning and regulation characteristics of thermal power units, photovoltaic power generation units and energy storage systems in joint frequency regulation. The dynamic response characteristics of each unit participating in frequency regulation are analyzed, and the regulation inertia and response lag characteristics of thermal power units, the regulation characteristics of photovoltaic power generation unit inverters, and the fast power response characteristics of energy storage systems are extracted. Based on the frequency response characteristics of each unit, a corresponding frequency modulation response sub-model is built, and the sub-model parameters and dynamic behaviors are integrated to form a complete joint frequency modulation dynamic model.

[0011] Optionally, establish an overall dynamic frequency response model for the combined thermal-solar-storage system, including: Based on the joint frequency regulation dynamic model, the coupling relationship and interaction characteristics of the frequency regulation output of each unit of thermal power, photovoltaic, and energy storage are determined. By introducing a system inertial center model and load damping characteristics, the dynamic influence relationship between load disturbance, unit frequency modulation output, and system frequency deviation is analyzed. A dynamic response equation is established between the system frequency deviation and the total power imbalance, forming an overall frequency dynamic response model that can characterize the overall frequency regulation dynamic behavior of the combined thermal-solar-storage system.

[0012] Optionally, based on the joint frequency regulation dynamic model, a digital twin prediction model of the combined thermal-solar-storage system is constructed, integrating historical operating data and new energy prediction information to generate the frequency regulation power demand in the future prediction time domain, including: Based on the joint frequency modulation dynamic model, the input variables and modeling basis of the digital twin prediction model are determined; Collect and integrate historical system operation data, real-time operation parameters, and new energy forecast information to complete data filtering, sorting, and standardization. Construct a digital twin prediction function to realize the rolling extrapolation of frequency modulation power demand in the future prediction time domain through mechanism and data fusion; The model input and operating status are updated in each control cycle, and the frequency modulation power demand in the future predicted time domain is output.

[0013] Optionally, based on the predicted frequency regulation power demand and combined with the overall frequency dynamic response model of the thermal-solar-storage integrated system, a model-based predictive control joint frequency regulation power optimization model is constructed, including: Using the predicted frequency modulation power demand as the feedforward input, the optimization objective and constraint boundary are determined by combining the overall frequency dynamic response model. Construct a multi-objective optimization function with the core objectives of minimizing system frequency deviation and minimizing the operating costs of thermal power units and energy storage systems; Based on the operating characteristics of each unit, constraints on the ramp-up of thermal power units, upper and lower limits of the state of charge of energy storage, and system power balance are set to form a complete constraint optimization system. A joint frequency modulation power optimization model is built based on the model predictive control algorithm. The dynamic characteristics of the overall frequency dynamic response model are integrated into the optimization process, so that the optimization process conforms to the dynamic operation law of the combined thermal-solar-storage system.

[0014] Optionally, the optimal solution for multi-source frequency modulation power allocation is solved in the prediction time domain, including: Within each control cycle, the input and constraints of the optimization problem are determined based on the generated frequency modulation power demand prediction results and the overall frequency dynamic response model. An optimization algorithm is used to solve the constrained optimization problem in a rolling manner, and an iterative solution is obtained for the multi-source frequency regulation power allocation scheme of thermal power, photovoltaic power and energy storage. Verify the feasibility and rationality of the allocation scheme, select the allocation result that satisfies all operational constraints and achieves the optimal dual objectives, and form the optimal solution for multi-source frequency modulation power allocation.

[0015] Optionally, the optimal solution for multi-source frequency regulation power allocation is used as the optimal frequency regulation power command and sent to thermal power units, photovoltaic power generation units, and energy storage systems. The real-time operating status after execution is fed back to the digital twin prediction model, including: The optimal solution for multi-source frequency regulation power allocation is converted into an executable optimal frequency regulation power command and sent to thermal power units, photovoltaic power generation units and energy storage systems; Real-time acquisition of operational status information after the execution of commands from each unit, including actual output of each unit, system frequency deviation, and energy storage state of charge; The collected status information is fed back to the digital twin prediction model to update the model input and correct subsequent prediction results; By cyclically executing prediction, optimization, command issuance, and status feedback, the closed-loop rolling optimization operation of the combined frequency regulation of thermal power, solar power, and energy storage is achieved.

[0016] To achieve the above objectives, a second aspect of the present invention provides a predictive control device for a combined frequency modulation model of thermal power, solar power, and energy storage based on digital twins, comprising: The multi-source modeling module is used to construct a joint frequency regulation dynamic model of a thermal power-photovoltaic-energy storage system, analyze the frequency response characteristics of thermal power units, photovoltaic power generation units and energy storage systems, and establish an overall frequency dynamic response model of the thermal power-photovoltaic-energy storage joint system. The demand forecasting module is used to construct a digital twin forecasting model of the combined thermal-solar-storage system based on the joint frequency regulation dynamic model, integrate historical operating data and new energy forecasting information, and generate frequency regulation power demand in the future forecast time domain. The power optimization module is used to construct a model-based predictive control joint frequency regulation power optimization model based on the predicted frequency regulation power demand and combined with the overall frequency dynamic response model of the thermal-solar-storage integrated system, and solve the optimal solution for multi-source frequency regulation power allocation in the prediction time domain. The closed-loop feedback module is used to take the optimal solution of the multi-source frequency regulation power allocation as the optimal frequency regulation power command, send it to the thermal power unit, photovoltaic power generation unit and energy storage system, and feed back the real-time operating status after execution to the digital twin prediction model to realize the closed-loop rolling optimization operation of the joint frequency regulation of thermal power, photovoltaic and energy storage.

[0017] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0018] To achieve the above objectives, a third aspect of this application provides an electronic device, including a processor and a memory; wherein the processor reads executable program code stored in the memory to run a program corresponding to the executable program code, for implementing a predictive control method based on a digital twin-driven joint frequency modulation model of thermal power, solar power, and energy storage as described in the first aspect embodiment.

[0019] To achieve the above objectives, the fourth aspect of this application proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a predictive control method for a combined frequency modulation model of fire-solar-storage based on digital twins as described in the first aspect embodiment.

[0020] The embodiments of the present invention have the following beneficial effects: 1. Improved frequency stability under complex disturbance conditions. This invention uses a digital twin predictive model to anticipate future frequency regulation needs and employs MPC rolling optimization control to effectively reduce grid frequency deviation.

[0021] 2. Robust optimization of multi-source coordinated frequency regulation is achieved. This invention explicitly considers thermal power ramp-up constraints and energy storage SOC constraints during the optimization process, improving the feasibility and robustness of the frequency regulation control strategy.

[0022] 3. Reduced frequency regulation operating costs and improved system economy. By incorporating thermal power wear costs and energy storage lifespan loss costs into the objective function, the system's total life-cycle operating costs are minimized. Attached Figure Description

[0023] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 A flowchart of a predictive control method based on a digital twin-driven joint frequency modulation model for thermal, solar, and energy storage is provided in this embodiment of the invention. Figure 2 An overall framework diagram of a predictive control method based on a digital twin-driven joint frequency modulation model for thermal-solar-storage systems is provided in an embodiment of the present invention. Figure 3 This is a block diagram of a dynamic model for combined frequency modulation of thermal power, solar power, and energy storage provided in an embodiment of the present invention. Figure 4 This is a structural diagram of a predictive control device based on a digital twin-based frequency modulation model for a combined thermal-solar-storage system, provided in an embodiment of the present invention. Detailed Implementation

[0024] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0025] 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 of the present invention. 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.

[0026] The following description, with reference to the accompanying drawings, describes a predictive control method and apparatus based on a digital twin-based combined frequency modulation model of thermal, solar, and energy storage, according to an embodiment of the present invention.

[0027] Example 1 This invention provides a predictive control method based on a digital twin-based joint frequency modulation model for thermal-solar-storage systems. Figure 1 This is a flowchart illustrating a predictive control method based on a digital twin for a combined frequency modulation model of thermal power, solar power, and energy storage, provided in an embodiment of the present invention. Figure 1 , Figure 2 As shown, the method includes the following steps: Step S1: Construct a joint frequency regulation dynamic model of the thermal power-photovoltaic-energy storage system, analyze the frequency response characteristics of the thermal power unit, photovoltaic power generation unit and energy storage system, and establish an overall frequency dynamic response model of the thermal power-photovoltaic-energy storage joint system.

[0028] In this embodiment, the overall operating characteristics of the combined thermal-solar-storage system are first analyzed to determine the functional positioning and regulation characteristics of the thermal power unit, photovoltaic power generation unit, and energy storage system in joint frequency regulation. The dynamic response patterns of each unit participating in frequency regulation are analyzed, extracting the regulation inertia and response lag characteristics of the thermal power unit, the inverter regulation characteristics of the photovoltaic power generation unit, and the fast power response characteristics of the energy storage system. Based on the frequency response characteristics of each unit, corresponding frequency regulation response sub-models are built. The sub-model parameters and dynamic behaviors are integrated to form a complete joint frequency regulation dynamic model, the structure of which is as follows: Figure 3 As shown.

[0029] The dynamic model mainly consists of five parts: load disturbance module, frequency regulation response sub-model module for each unit, power imbalance calculation module, grid frequency dynamic link module, and measurement feedback module. The modules work together to complete the frequency regulation dynamic response process of the thermal-solar-storage integrated system.

[0030] Specifically, the frequency regulation dynamic characteristics of a thermal power unit, considering both the governor and the turbine reheat circuit, are expressed by the following transfer function: ,in The governor time constant represents the response delay time for the governor to complete the action adjustment after receiving the frequency deviation signal; The time constant of the reheater reflects the lag characteristics of the turbine reheating process in energy transfer and conversion. is the high-pressure cylinder power proportionality coefficient, representing the proportion of the high-pressure cylinder output power to the total turbine output power; s is the Laplace operator, used for frequency domain analysis of the dynamic system. In the frequency regulation process, the photovoltaic power generation system participates in power regulation through the inverter. Its frequency response characteristics can be equivalent to a first-order inertial element, and its transfer function is expressed as: ,in The frequency modulation gain of a photovoltaic inverter represents the amplification ratio of the output power of a photovoltaic power generation system as the frequency deviation signal changes. Let be the response time constant of the photovoltaic system, reflecting the combined response delay of the inverter control loop and power output stage; s is the Laplace operator. The energy storage system is used for rapid frequency regulation, and its power response characteristics can also be expressed as a first-order inertial element, with the transfer function expressed as: ,in The frequency modulation gain of the energy storage system characterizes the amplification ratio of the charging and discharging power of the energy storage system as a function of the frequency deviation signal. is the response time constant of the energy storage system, reflecting the combined response delay of the energy storage converter and the battery body; s is the Laplace operator.

[0031] Based on this, and using a joint frequency regulation dynamic model, the coupling relationship and interaction characteristics of the frequency regulation output of each unit (thermal power, photovoltaic, and energy storage) are determined. A system inertial center model and load damping characteristics are introduced to analyze load disturbances. The dynamic influence relationship between unit frequency modulation output and system frequency deviation Δf is established, and the dynamic response equation of system frequency deviation is constructed: H is the system inertia constant, which reflects the sum of the overall rotational inertia of the combined thermal-solar-storage system and determines the system's resistance to power imbalance changes in frequency; D is the load damping coefficient, which characterizes the sensitivity of load power to changes in system frequency and reflects the damping effect of the load on frequency deviation. This represents the increase in the system's mechanical power, corresponding to the total frequency regulation output of the thermal power unit, photovoltaic system, and energy storage system. The system electrical power increment corresponds to the load disturbance. Applying a Laplace transform to this equation yields the dynamic response equation in the frequency domain: , where s is the Laplace operator, This is the Laplace transform of the system frequency deviation, representing the difference between the actual operating frequency and the rated frequency of the system. The Laplace transform of the total power imbalance of the system is, i.e. and The difference is expressed as: ,in This represents the change in frequency regulation output of thermal power units. This represents the change in frequency regulation output of the photovoltaic power generation unit. This refers to the change in frequency regulation output of the energy storage system. This represents the load disturbance variation of the regional power grid, thus forming a comprehensive frequency dynamic response model that can characterize the overall frequency regulation dynamic behavior of the combined thermal-solar-storage system, such as... Figure 3 As shown, load disturbance Thermal power unit frequency regulation model Photovoltaic inverter frequency modulation model Frequency regulation model of energy storage system The combined effect creates a power imbalance The frequency deviation output by the power grid frequency dynamic link It also performs measurement feedback to form a complete frequency modulation dynamic closed loop.

[0032] Step S2: Based on the joint frequency regulation dynamic model, construct a digital twin prediction model of the combined thermal-solar-storage system, integrate historical operating data and new energy prediction information, and generate the frequency regulation power demand in the future prediction time domain.

[0033] In this embodiment, the dynamic model of joint frequency modulation of thermal power, solar power, and energy storage obtained in S1 is used as the core mechanism to determine the input variables and modeling boundaries of the digital twin prediction model. The input and output relationship of this model is as follows: Figure 2 As shown, the input variables specifically include the load disturbance sequence. Photovoltaic prediction error The system outputs a sequence of frequency regulation power demand in the future time domain, based on multi-dimensional information such as energy storage SOC status and real-time system operating status, providing feedforward support for subsequent model predictive control.

[0034] This application's embodiments first collect and fuse multi-source operational data, including historical system frequency regulation operational data, real-time collected operational parameters, and new energy prediction information. The collected data undergoes standardization processing such as filtering, noise reduction, and normalization to eliminate dimensional differences and outlier interference from different data sources, ensuring data quality and modeling reliability. Based on this, a digital twin prediction function is constructed: ,in It is a digital twin prediction function, constructed by fusing mechanistic models and data-driven methods, which can accurately map the dynamic relationship between system state, control variables and disturbance variables and frequency modulation power demand; Let k be the system state variable at time k. The control quantity at time k. Let k be the disturbance at time k.

[0035] This application's embodiments achieve future... Rolling projection of cadence power demand, among which The preset prediction time domain length represents the number of control cycles the model can predict in advance. Within each control cycle, the model input and system operating state are updated in real time, and the future is re-inferred based on the latest acquired state variables, control variables, and disturbance variables. The frequency modulation power requirement of the step, output frequency modulation power requirement sequence (i=1,2,…, This sequence can accurately reflect the changing trend of system frequency modulation demand in the future time domain, providing forward-looking feedforward information for subsequent joint frequency modulation power optimization based on MPC, effectively improving the predictive ability and dynamic response accuracy of frequency modulation control, and reducing the risk of frequency fluctuations.

[0036] Step S3: Based on the predicted frequency regulation power demand and combined with the overall frequency dynamic response model of the combined thermal-solar-storage system, construct a model-based predictive control combined frequency regulation power optimization model, and solve the optimal solution for multi-source frequency regulation power allocation in the prediction time domain.

[0037] In this embodiment of the application, the frequency regulation power demand prediction result obtained in S2 is used as the feedforward input. Combined with the overall frequency dynamic response model obtained in S1, the optimization objectives and constraint boundaries are clarified, and an MPC optimization framework that fits the operating characteristics of the combined thermal-solar-storage system is constructed.

[0038] This application first constructs a multi-objective optimization function with the core objectives of minimizing system frequency deviation and minimizing the operating costs of thermal power units and energy storage systems. The expression is as follows:

[0039] in, The predicted time domain length represents the number of future control cycles covered by the optimization calculation; The preset weighting coefficients can be flexibly adjusted according to the needs of the frequency regulation scenario, respectively representing the priority of frequency deviation, thermal power unit operating cost, and energy storage system operating cost in the optimization objectives; The system frequency deviation at time k+i reflects the degree to which the system frequency deviates from the rated value in the future time domain; The operating cost of the thermal power unit at time k+i specifically includes coal fuel cost, unit regulation loss cost, start-up and shutdown auxiliary cost, etc. The operating cost of the energy storage system at time k+i specifically includes the cost of charging and discharging power loss, the cost of battery cycle life degradation, and maintenance costs.

[0040] Based on this, the embodiments of this application, combined with the operating characteristics of each unit, set three types of core constraints to ensure the feasibility of the optimization scheme and the safety of the equipment: First, the ramp-up constraint for thermal power units, expressed as follows: ,in Let be the change in power output of the thermal power unit at time k+i. This refers to the maximum permissible ramp rate for thermal power units. The duration of a single control cycle is used to limit the rate of change of the output of thermal power units, avoid excessively rapid adjustments that could cause mechanical damage to core components such as boilers and turbines, and ensure the stable operation of thermal power units.

[0041] Second, the energy storage SOC constraint, expressed as follows: ,in This is the lower limit of the safe state of charge for energy storage systems. This is the upper limit of the safe state of charge of the energy storage system. This constraint is used to ensure that the energy storage system is always in a reasonable state of charge range during frequency regulation, to avoid overcharging or over-discharging from damaging the battery life, and to ensure that the energy storage system has the ability to continuously participate in frequency regulation.

[0042] Thirdly, there is the power balance constraint, expressed as follows: ,in The frequency modulation power demand is predicted by S2. , , These are the frequency regulation outputs of thermal power units, photovoltaic power generation units, and energy storage systems, respectively. This constraint is used to ensure that the sum of the frequency regulation outputs of each unit on the power generation side is fully matched with the predicted frequency regulation power demand, maintain system power balance, and provide a basic guarantee for frequency stability.

[0043] Based on model predictive control algorithms, this embodiment of the application constructs a joint frequency regulation power optimization model, embedding the overall frequency dynamic response model obtained in S1 into the optimization solution process. This ensures that the optimization calculation fully considers dynamic characteristics such as system inertia and damping, guaranteeing that the optimization results conform to the actual operating laws of the combined thermal-solar-storage system. Within each control cycle, this embodiment updates the input variables and constraints of the optimization problem based on the latest generated frequency regulation power demand prediction results and the overall frequency dynamic response model. Efficient optimization algorithms such as sequential quadratic programming and interior-point methods are used to iteratively solve the constrained optimization problem, obtaining a multi-source frequency regulation output allocation scheme for thermal power, photovoltaics, and energy storage. Subsequently, the allocation scheme is feasibility-verified to check whether it meets all constraints. Allocation results that both meet equipment operating limitations and achieve multi-objective optimization are selected, ultimately forming the optimal solution for multi-source frequency regulation power allocation. This provides a precise basis for issuing subsequent frequency modulation commands.

[0044] Step S4: The optimal solution of the multi-source frequency regulation power allocation is used as the optimal frequency regulation power command and sent to the thermal power unit, photovoltaic power generation unit and energy storage system. The real-time operating status after execution is fed back to the digital twin prediction model to realize the closed-loop rolling optimization operation of the thermal-solar-storage joint frequency regulation.

[0045] In this embodiment of the application, the optimal solution for multi-source frequency modulation power allocation obtained in S3 is first processed. The command conversion process transforms the abstract optimized allocation results into optimal frequency regulation power commands that each unit can directly recognize and execute. This ensures that the command format matches the control interfaces of the thermal power unit, photovoltaic power generation unit, and energy storage system, avoiding deviations during command transmission and execution. Subsequently, the converted optimal frequency regulation power commands are sent to the corresponding devices. The entire command sending process is synchronized in real time, ensuring coordinated response from all units. Figure 2 As shown, this step is the core link connecting optimization decision-making and actual frequency regulation execution, and directly determines the implementation effect of frequency regulation control.

[0046] After the command is issued, this embodiment of the application collects the operating status information of each unit after the command is executed through the system real-time monitoring module. It focuses on collecting three types of key parameters: first, the actual frequency regulation output of each unit, including the actual output of thermal power units, photovoltaic power generation units, and energy storage systems, to verify the deviation between the actual output and the command output; second, the system frequency deviation Δf, to evaluate whether the frequency regulation effect has achieved the optimization target; and third, the energy storage state of charge (SOC), to monitor whether the operating status of the energy storage system meets the constraints. At the same time, the collected information also includes auxiliary parameters such as the operating temperature of each unit and equipment losses, so as to comprehensively grasp the operating status of the system.

[0047] After data collection, all real-time operational status information is organized and standardized, and then fed back to the digital twin prediction model in S2 to achieve real-time updates of the model input. Based on the actual operational data feedback, the digital twin prediction model corrects its prediction parameters and inference logic, compensates for possible deviations in the early prediction process, and further improves the accuracy of subsequent frequency regulation power demand prediction, forming a closed-loop iterative mechanism of "prediction-optimization-execution-feedback-correction".

[0048] This embodiment of the application, through the aforementioned cyclical execution process of prediction, optimization, command issuance, and status feedback, determines in real time whether the preset maximum cycle period has been reached. If the maximum cycle period has not been reached, it returns to S2, and based on the updated system status, it re-predicts the frequency regulation power demand, initiating a new round of optimization and execution. If the maximum cycle period has been reached, the current joint frequency regulation control process ends. Through this closed-loop rolling optimization mechanism, dynamic adaptive adjustment of the combined frequency regulation of thermal power, solar power, and energy storage is achieved. Under the premise of strictly meeting the ramp-up constraints of thermal power units, energy storage SOC constraints, and other equipment operation restrictions, it effectively reduces the fluctuation range of system frequency deviation Δf, improves system frequency stability, and reduces the operating costs of thermal power units and energy storage systems. Ultimately, it achieves a dual improvement in the stability and economy of the combined frequency regulation operation of thermal power, solar power, and energy storage, fully leveraging the frequency regulation advantages of each unit and solving the technical pain points of large fluctuations in new energy frequency regulation and lag in the response of thermal power units.

[0049] Example 2 This invention provides a predictive control device based on a digital twin-based joint frequency modulation model of thermal power, solar power, and energy storage. Figure 4 This is a schematic flowchart of a predictive control device based on a digital twin-based joint frequency modulation model for thermal, solar, and energy storage, provided in an embodiment of the present invention. Figure 4 As shown, the device includes: The multi-source modeling module 100 is used to construct a joint frequency regulation dynamic model of a thermal power-photovoltaic-energy storage system, analyze the frequency response characteristics of thermal power units, photovoltaic power generation units and energy storage systems, and establish an overall frequency dynamic response model of the thermal power-photovoltaic-energy storage joint system. Demand forecasting module 200 is used to construct a digital twin forecasting model of the combined thermal-solar-storage system based on the joint frequency regulation dynamic model, integrate historical operating data and new energy forecasting information, and generate frequency regulation power demand in the future forecast time domain. The power optimization module 300 is used to construct a model-based predictive control joint frequency regulation power optimization model based on the predicted frequency regulation power demand and combined with the overall frequency dynamic response model of the thermal-solar-storage integrated system, and solve the optimal solution of multi-source frequency regulation power allocation in the prediction time domain. The closed-loop feedback module 400 is used to take the optimal solution of the multi-source frequency regulation power allocation as the optimal frequency regulation power command, send it to the thermal power unit, photovoltaic power generation unit and energy storage system, and feed back the real-time operating status after execution to the digital twin prediction model to realize the closed-loop rolling optimization operation of the joint frequency regulation of thermal power, photovoltaic and energy storage.

[0050] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0051] Example 3 To implement the methods of the above embodiments, the present invention also provides an electronic device, which includes a memory and a processor; wherein the processor reads executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the various steps of the methods described above.

[0052] Example 4 To implement the above embodiments, this application also proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the method described in the foregoing embodiments.

[0053] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0054] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0055] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A predictive control method based on a digital twin-based joint frequency modulation model of thermal-solar-storage system, characterized in that, include: S1. Construct a joint frequency regulation dynamic model of the thermal power-photovoltaic-energy storage system, analyze the frequency response characteristics of thermal power units, photovoltaic power generation units and energy storage systems, and establish an overall frequency dynamic response model of the thermal power-photovoltaic-energy storage joint system. S2. Based on the joint frequency regulation dynamic model, construct a digital twin prediction model of the combined thermal-solar-storage system, integrate historical operating data and new energy prediction information, and generate the frequency regulation power demand in the future prediction time domain. S3. Based on the predicted frequency regulation power demand and combined with the overall frequency dynamic response model of the thermal-solar-storage integrated system, a model-predictive control-based joint frequency regulation power optimization model is constructed. This model solves for the optimal solution of multi-source frequency regulation power allocation within the prediction time domain. This includes: using the predicted frequency regulation power demand as feedforward input and combining it with the overall frequency dynamic response model to determine the optimization objective and constraint boundaries; constructing a multi-objective optimization function with the core objectives of minimizing system frequency deviation and minimizing the operating costs of thermal power units and energy storage systems; setting ramp-up constraints for thermal power units, upper and lower limits of energy storage charge state constraints, and system power balance constraints based on the operating characteristics of each unit, forming a complete constraint optimization system; building a joint frequency regulation power optimization model based on the model predictive control algorithm, integrating the dynamic characteristics of the overall frequency dynamic response model into the optimization process, making the optimization process conform to the dynamic operating laws of the thermal-solar-storage integrated system; determining the input and constraint conditions of the optimization problem within each control cycle based on the generated frequency regulation power demand prediction results and the overall frequency dynamic response model; and using an optimization algorithm to iteratively solve the constraint optimization problem to obtain a multi-source frequency regulation power allocation scheme for thermal power, photovoltaic, and energy storage. Verify the feasibility and rationality of the allocation scheme, select the allocation result that satisfies all operational constraints and achieves the optimal dual objectives, and form the optimal solution for multi-source frequency modulation power allocation; S4. The optimal solution of the multi-source frequency regulation power allocation is used as the optimal frequency regulation power command and sent to the thermal power unit, photovoltaic power generation unit and energy storage system. The real-time operating status after execution is fed back to the digital twin prediction model to realize the closed-loop rolling optimization operation of the joint frequency regulation of thermal power, photovoltaic and energy storage.

2. The method according to claim 1, characterized in that, Constructing a joint frequency regulation dynamic model for a thermal power-photovoltaic-energy storage system, including: The overall operating characteristics of the thermal-solar-storage integrated system were analyzed to determine the functional positioning and regulation characteristics of thermal power units, photovoltaic power generation units and energy storage systems in joint frequency regulation. The dynamic response characteristics of each unit participating in frequency regulation are analyzed, and the regulation inertia and response lag characteristics of thermal power units, the regulation characteristics of photovoltaic power generation unit inverters, and the fast power response characteristics of energy storage systems are extracted. Based on the frequency response characteristics of each unit, a corresponding frequency modulation response sub-model is built, and the sub-model parameters and dynamic behaviors are integrated to form a complete joint frequency modulation dynamic model.

3. The method according to claim 2, characterized in that, Establish an overall dynamic frequency response model for a combined thermal-solar-storage system, including: Based on the joint frequency regulation dynamic model, the coupling relationship and interaction characteristics of the frequency regulation output of each unit of thermal power, photovoltaic, and energy storage are determined. By introducing a system inertial center model and load damping characteristics, the dynamic influence relationship between load disturbance, unit frequency modulation output, and system frequency deviation is analyzed. A dynamic response equation is established between the system frequency deviation and the total power imbalance, forming an overall frequency dynamic response model that can characterize the overall frequency regulation dynamic behavior of the combined thermal-solar-storage system.

4. The method according to claim 3, characterized in that, Based on the aforementioned joint frequency regulation dynamic model, a digital twin prediction model of the combined thermal-solar-storage system is constructed. This model integrates historical operating data with new energy forecasting information to generate the frequency regulation power demand in the future forecast time domain, including: Based on the joint frequency modulation dynamic model, the input variables and modeling basis of the digital twin prediction model are determined; Collect and integrate historical system operation data, real-time operation parameters, and new energy forecast information to complete data filtering, sorting, and standardization. Construct a digital twin prediction function to realize the rolling extrapolation of frequency modulation power demand in the future prediction time domain through mechanism and data fusion; The model input and operating status are updated in each control cycle, and the frequency modulation power demand in the future predicted time domain is output.

5. The method according to claim 1, characterized in that, The optimal solution for multi-source frequency regulation power allocation is used as the optimal frequency regulation power command and issued to thermal power units, photovoltaic power generation units, and energy storage systems. The real-time operating status after execution is fed back to the digital twin prediction model, including: The optimal solution for multi-source frequency regulation power allocation is converted into an executable optimal frequency regulation power command and sent to thermal power units, photovoltaic power generation units and energy storage systems; Real-time acquisition of operational status information after the execution of commands from each unit, including actual output of each unit, system frequency deviation, and energy storage state of charge; The collected status information is fed back to the digital twin prediction model to update the model input and correct subsequent prediction results; By cyclically executing prediction, optimization, command issuance, and status feedback, the closed-loop rolling optimization operation of the combined frequency regulation of thermal power, solar power, and energy storage is achieved.

6. A predictive control device based on a digital twin-based joint frequency modulation model of thermal power, solar power, and energy storage, employing the predictive control method based on a digital twin-based joint frequency modulation model of thermal power, solar power, and energy storage as described in any one of claims 1-5, characterized in that, include: The multi-source modeling module is used to construct a joint frequency regulation dynamic model of a thermal power-photovoltaic-energy storage system, analyze the frequency response characteristics of thermal power units, photovoltaic power generation units and energy storage systems, and establish an overall frequency dynamic response model of the thermal power-photovoltaic-energy storage joint system. The demand forecasting module is used to construct a digital twin forecasting model of the combined thermal-solar-storage system based on the joint frequency regulation dynamic model, integrate historical operating data and new energy forecasting information, and generate frequency regulation power demand in the future forecast time domain. The power optimization module is used to construct a model-based predictive control joint frequency regulation power optimization model based on the predicted frequency regulation power demand and combined with the overall frequency dynamic response model of the thermal-solar-storage integrated system, and solve the optimal solution for multi-source frequency regulation power allocation in the prediction time domain. The closed-loop feedback module is used to take the optimal solution of the multi-source frequency regulation power allocation as the optimal frequency regulation power command, send it to the thermal power unit, photovoltaic power generation unit and energy storage system, and feed back the real-time operating status after execution to the digital twin prediction model to realize the closed-loop rolling optimization operation of the joint frequency regulation of thermal power, photovoltaic and energy storage.

7. An electronic device, characterized in that, Including processor and memory; The processor reads executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the method as described in any one of claims 1-5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-5.

Images