Output prediction method for real-time energy management of integrated energy system and related device

Abstract

The invention discloses an output prediction method for real-time energy management of an integrated energy system and a related device. Obtaining a corrected predicted value with a confidence interval; taking the corrected predicted value with the confidence interval as the input, taking the lowest operation cost of the integrated energy system and keeping the balance of supply and demand as the target, constructing a target function, solving the target function, and obtaining a fault-tolerant energy scheduling scheme. And natural coupling of an error correction mechanism and a scheduling model is realized.

Description

Technical Field

[0001] This invention belongs to the field of energy management of integrated energy systems, and relates to a method and related apparatus for real-time energy management of integrated energy systems to predict output. Background Technology

[0002] With the advancement of the "dual-carbon" strategy and the deepening of energy structure transformation, Integrated Energy Systems (IES), as a key form of synergistic utilization of multiple energy sources, are becoming an important support for green and low-carbon energy systems. A typical IES structure may include photovoltaic power generation systems, battery energy storage, hot water storage equipment, water electrolysis hydrogen production devices, solid oxide fuel cells (SOFC), air source heat pumps, and other equipment, and carry out multi-energy dispatch optimization for multi-energy coupling demand scenarios (such as industrial parks, large buildings, and small town microgrids).

[0003] In such systems, the Energy Management System (EMS) undertakes the unified scheduling and control of various types of equipment. Its core objective is to achieve real-time supply and demand balance of multiple energy forms such as electricity, heat, and hydrogen, minimize operating costs, delay equipment aging, and improve system self-sufficiency. However, due to the strong volatility and unpredictability of renewable energy sources, such as photovoltaics, traditional EMS solutions face significant challenges posed by operational uncertainties.

[0004] In traditional approaches, Model Predictive Control (MPC) is widely used for real-time scheduling of integrated energy systems, employing rolling optimization strategies to achieve dynamic feedback adjustments to the system state. However, the performance of MPC control is highly dependent on the predicted results of parameters such as future illumination and load. Most current methods formulate optimization schemes based on deterministic predictions, failing to fully reflect the impact of uncertainty on scheduling outcomes. In actual operation, prediction biases can easily lead to problems such as battery overcharging, insufficient heat load, or energy waste.

[0005] To overcome these shortcomings, some studies have attempted to introduce scene trees or probabilistic models to construct scene sets and solve different possible scenarios through stochastic optimization. However, such methods face limitations in high-frequency rolling scheduling, such as an excessive number of scenes, model dimensionality expansion, and low solution speed, making it difficult to meet real-time requirements. Other studies employ robust optimization or distributed robust optimization strategies, designing scheduling solutions based on worst-case scenarios given a known set of uncertainties. While this ensures scheduling feasibility, it often suffers from conservative optimization results, redundant energy consumption, and increased operating costs. Meanwhile, some deep learning-based prediction correction methods attempt to improve prediction accuracy, such as using LSTM networks for illumination error regression. However, these methods require a large amount of high-quality training data, are difficult to generalize to scenarios with drastic fluctuations and weak predictive model timeliness, and have high training and deployment costs.

[0006] In summary, existing methods generally lack the ability to dynamically identify and correct prediction errors online, or introduce excessive computational complexity into integrated scheduling control, resulting in an inability to balance optimization accuracy and real-time response. Therefore, there is an urgent need for an overall framework that balances prediction error modeling with optimized scheduling stability, possesses real-time response capabilities, and has a simple and deployable algorithm structure. This framework should be able to naturally couple the error correction mechanism with the scheduling model to improve the practicality and robustness of integrated energy systems in photovoltaic-dominated scenarios. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method and related device for real-time energy management of integrated energy systems. This method and related device can take into account measurement errors to determine energy dispatch schemes and achieve natural coupling between error correction mechanisms and dispatch models.

[0008] To achieve the above objectives, this invention discloses an output prediction method for real-time energy management of an integrated energy system, comprising: Obtain the corrected predicted value with confidence interval; Using the corrected prediction value with confidence interval as input, and aiming to minimize the operating cost of the integrated energy system while maintaining supply and demand balance, an objective function is constructed, and the objective function is solved to obtain a fault-tolerant energy dispatch scheme.

[0009] Furthermore, the process of obtaining the corrected predicted value with confidence interval is as follows: Predicted solar radiation values ​​for a predetermined time in the future; The predicted solar radiation values ​​at a predetermined future time and their corresponding measured values ​​are input as an observation sequence into a Hidden Markov Model (HMM) to obtain corrected prediction values ​​with confidence intervals.

[0010] Furthermore, the objective function is:

[0011] in, For grid interaction costs; Costs related to battery aging; / Punishment for wasting heat / hydrogen energy; Penalty for changes in equipment handling; To constrain the over-boundary relaxation penalty, T represents the predictive control time domain.

[0012] Furthermore, the constraints corresponding to the objective function include energy storage power constraints and energy state constraints.

[0013] Furthermore, the energy storage power constraint is as follows:

[0014] The energy state constraint (SoE) is: , in, This is the power / energy uncertainty estimate converted from the prediction error. As slack variables, slight out-of-bounds errors are allowed, and the penalty weights are applied. control.

[0015] Furthermore, the process of solving the objective function to obtain a fault-tolerant energy scheduling scheme is as follows: The confidence interval is taken as the uncertainty range; By tightening constraints within the uncertain range, a fault-tolerant energy scheduling scheme is obtained.

[0016] This invention discloses an output prediction system for real-time energy management of an integrated energy system, comprising: The acquisition module is used to acquire corrected prediction values ​​with confidence intervals; The solution module is used to take the corrected prediction value with confidence interval as input, and with the goal of minimizing the operating cost of the integrated energy system and maintaining supply and demand balance, to construct an objective function, solve the objective function, and obtain a fault-tolerant energy scheduling scheme.

[0017] Furthermore, the process of obtaining the corrected predicted value with confidence interval is as follows: Predicted solar radiation values ​​for a predetermined time in the future; The predicted solar radiation values ​​at a predetermined future time and their corresponding measured values ​​are input as an observation sequence into a Hidden Markov Model (HMM) to obtain corrected prediction values ​​with confidence intervals.

[0018] This invention discloses a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of an output prediction method for real-time energy management of the integrated energy system.

[0019] The present invention discloses a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the steps of the output prediction method for real-time energy management of the integrated energy system are implemented.

[0020] The present invention has the following beneficial effects: The output prediction method and related devices for real-time energy management of integrated energy systems described in this invention, in specific operation, dynamically sense and correct photovoltaic output prediction errors by constructing a prediction error identification and correction unit based on a Hidden Markov Model, forming reliable interval-based prediction results. Furthermore, it introduces constraint tightening and soft constraint relaxation mechanisms into Model Predictive Control (MPC) scheduling to achieve embedded buffering of uncertainties and improve control robustness. Compared to traditional scenario optimization, robust optimization, or deep learning methods, this invention does not require a large number of scenario generation or complex distribution assumptions, and has the advantages of simple computational structure, high solution efficiency, and fast rolling response. It is particularly suitable for integrated energy systems, microgrids, and smart buildings with high proportions of renewable energy access, providing a stable, flexible, and economical unified solution for scheduling control in actual energy systems. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of the method of the present invention; Figure 2 A graph showing the statistical comparison of errors under different prediction correction methods; Figure 3 A scatter plot comparing different correction methods. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] In the description of this invention, it should be understood that the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0025] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0026] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Additionally, the character " / " in this invention generally indicates that the preceding and following objects have an "or" relationship.

[0027] It should be understood that although terms such as first, second, third, etc., may be used in the embodiments of the present invention to describe the preset range, these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from one another. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0028] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0029] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0030] The accompanying drawings illustrate various structural schematic diagrams according to embodiments disclosed in this invention. These drawings are not to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0031] Example 1 The output prediction method for real-time energy management of integrated energy systems according to the present invention includes the following steps: 1) Prediction and correction unit (based on hidden Markov model) This invention uses a Hidden Markov Model (HMM) to model the solar irradiance prediction error, constructs an "observation-state" mapping relationship using historical data, and performs interval estimation of the prediction error range for future periods, outputting a corrected prediction value with confidence interval.

[0032] 11) State definition; Observations ,in, v(t) For the current observation value, v max (t) This is the upper limit of solar radiation outside the sun on that day.

[0033] Hidden state variables ,in, r(t) The actual observed value represents the prediction error ratio.

[0034] By V o (t) and V h (t)Discretize the data into N and M levels with equal probability density, respectively, and construct the state space S = {s1, ... s2}. M} and the observation space O={o1,…,o N}

[0035] 12) Parameter training (model training based on frequency statistics methods or the Baum-Welch algorithm); Construct the transition probability matrix ; Construct the emission probability matrix ; 13) Predictive correction (online phase); The Viterbi algorithm is used to determine the most likely hidden state path. ; The future state is corrected exponentially based on the latest error value S(t):

[0036] in, , is a time-weighted function; Map the corrected state back to the predicted correction factor: , The upper and lower bounds of the prediction interval are It outputs a corrected prediction sequence with an error range, providing uncertainty input for scheduling optimization.

[0037] 2) Scheduling optimization unit (based on MPC and constraint tightening) The scheduling optimization part adopts the rolling model predictive control (MPC) strategy, combined with the upper-level correction prediction input, to optimize the power and energy status of equipment such as batteries, hot water tanks, and hydrogen storage tanks in each scheduling cycle, with the goal of minimizing operating costs and maintaining supply and demand balance.

[0038] 21) Optimize the objective function; In the predictive control time domain T, the objective function is:

[0039] Among them, power grid interaction costs Battery aging costs Penalties for wasting heat / hydrogen energy Equipment processing change penalty ; Lax penalties for exceeding boundaries

[0040] 22) Constraint tightening strategy; For equipment operating states within a range of uncertainty, this invention improves robustness by dynamically tightening constraints and introducing a buffer zone.

[0041] For the energy storage power constraint:

[0042] The energy state constraint (SoE) is as follows: , in, This is the power / energy uncertainty estimate converted from the prediction error. As slack variables, slight out-of-bounds errors are allowed, and the penalty weights are applied. control.

[0043] 3) System Coupling The two-layer correction framework structurally presents a coupled relationship of "upper-layer perception and correction + lower-layer optimization and execution," and in terms of its operational mechanism, it adopts a time-closed-loop control logic of "rolling prediction-correction-scheduling-execution" to ensure that the system can operate stably in dynamically changing environments. Specific description: 31) Module collaboration process (time-closed-loop logic); In each scheduling cycle (e.g., every 15 minutes), the system completes a full prediction-scheduling closed-loop operation in the following order: 311) Obtain the current system status: Collect real-time operating data including measured solar irradiance, electrical / thermal / hydrogen load, SoE status of energy storage devices, and power purchase status from the grid.

[0044] 312) Input forecast data: Call the solar irradiance forecast values ​​for the next few hours generated by the meteorological agency or forecasting module. v(t) Construct the predicted sequence.

[0045] The prediction correction unit (HMM module) is activated: It inputs the predicted and measured values ​​over a past period as an observation sequence into the HMM model, infers the most probable error state path, and performs exponential decay correction based on the most recent actual error, ultimately outputting a sequence of predicted values ​​with upper and lower bounds. .

[0046] Input the corrected prediction into the scheduling unit (MPC module): Use the corrected... Given the error interval, construct a scheduling optimization problem within the uncertainty range. By tightening the internal constraints, a fault-tolerant energy scheduling scheme is obtained.

[0047] Solve and execute the first step of control action: perform rolling solution on the predicted time domain (e.g., 24 hours) to obtain the control input sequence, and execute only the equipment instructions at the first moment (i.e. the current moment), such as battery discharge power, heat pump start intensity, hydrogen storage tank charging and discharging amount, etc.

[0048] Waiting to enter the next time step, the loop repeats. This control framework has feedforward + feedback fusion characteristics, which can continuously correct the impact of prediction errors on scheduling accuracy, ensuring that the system has dynamic correction and real-time optimization capabilities. 32) Design of data coupling and decoupling interfaces between modules; The various functional modules interact with each other through interface functions or data channels: The HMM module outputs two types of data: center prediction value. Uncertainty upper and lower limits ; After receiving this information, the MPC module performs two tasks: constructing the multi-energy flow cost term in the objective function; and dynamically adjusting the device operation constraint boundaries to accommodate errors.

[0049] The entire process can be encapsulated as the following function chain relationship (where... For the current control decision, state refers to the device and system state parameters:

[0050] Example 2 refer to Figure 1 This invention comprises two main modules (upper layer + lower layer), which operate within each real-time scheduling cycle (e.g., 15 minutes), forming a closed-loop control process. The specific process is as follows: 1) Input data preparation. Obtain the predicted solar irradiance sequence for the future forecast period. v(t) The data comes from external weather services or forecasting modules; it also collects forecast values ​​and corresponding measured irradiance values ​​over a past period. r(t) Historical data used for error identification; current system operating status (such as battery SoC, thermal load, grid power, equipment output limits, etc.).

[0051] 2) Construct the input for the HMM model. Calculate the observed variables of the historical observation sequence. V o and hidden state variables V h Discretize the two variables separately (with equal probability density partitioning) to form a state set and an observation set; construct an HMM model (state transition matrix A, emission matrix B).

[0052] 3) Error State Path Inference. The Viterbi algorithm is used to decode the error state path of the current historical observation sequence to obtain the most likely hidden error state sequence corresponding to the predicted value. .

[0053] 4) Dynamic error correction. This is based on the most recent measured error. The predicted error state path is dynamically corrected using an exponential decay function: The corrected state is then mapped back to the predicted value range: At the same time, upper and lower confidence intervals (representing the range of uncertainty) are determined.

[0054] 5) Predicted and corrected output. Output center predicted value sequence. Output the uncertainty interval consisting of the upper and lower limits of the corresponding error. , which serves as the input parameter for the scheduling module.

[0055] 6) Construct the MPC rolling scheduling model. Set the predictive control time domain (e.g., the next 24 hours); construct the scheduling objective function, considering multi-energy loads (electricity, heat, hydrogen) and cost factors, including: grid purchase cost, energy storage aging penalty, heat / hydrogen waste penalty, output change penalty, and slack variable penalty (soft constraint treatment).

[0056] 7) Constraint tightening. Based on the prediction error range, the operational boundaries of key equipment (batteries, heating tanks, hydrogen tanks) are dynamically tightened: , If the tightening is too tight, a slack variable will be automatically introduced. And write it into the penalty term of the objective function.

[0057] 8) Solve the optimization model. Input the corrected predictions and tightened constraints into the MPC scheduling model; use the solver to complete one optimization in the control time domain; extract only the first control action at the current time. implement.

[0058] 9) Execute scheduling and enter the next cycle. Send scheduling instructions (such as battery discharge, SOFC start / stop, heat pump adjustment, etc.) to the controller; wait to enter the next cycle, and repeat steps 1) to 9) to form a real-time rolling closed-loop control system.

[0059] Figure 2 This figure shows a statistical comparison of solar irradiance prediction errors under different prediction correction methods. The horizontal axis represents different estimation methods (original prediction sequence, uncorrected HMM estimation, HMM-based correction, DES-based correction, and LSTM-based correction), and the vertical axis represents the irradiance prediction error (unit: W / m²). Figure 2 As can be seen, the error distribution of the uncorrected prediction sequence is relatively dispersed, while after applying the present invention, the prediction error is significantly narrowed, the median is closer to zero, and the error dispersion is significantly reduced, demonstrating superior stability and accuracy compared to methods such as DES and LSTM. This indicates that the present invention can effectively improve prediction accuracy and reduce scheduling deviations caused by uncertainty while maintaining real-time performance.

[0060] Figure 3A scatter plot comparing predicted / estimated irradiance values ​​with measured irradiance values ​​under different correction methods is shown, with the red dashed line representing the ideal equation (predicted value equals measured value). The results show that the scatter plot distribution based on the HMM correction unit is closest to the equation, and its overall fit is higher than that based on DES and LSTM. This further demonstrates that the present invention has stronger adaptability and accuracy in capturing the dynamic characteristics of irradiance prediction errors, and can better support the real-time scheduling of integrated energy systems in actual operation.

[0061] In practical applications, this invention... Figure 2 The comparison of error distributions shown verifies the effectiveness of the prediction correction unit. The results indicate that after adopting the HMM-based two-layer correction framework, the median error of irradiation prediction is significantly reduced, and the error distribution range is significantly narrowed. Meanwhile, from... Figure 3 A comparison of the scattered points reveals that the corrected predictions closely match the measured values, with the overall point cloud distribution more concentrated near the equation line. This high-precision prediction input further enhances the robustness of the lower-level MPC scheduling optimization unit, enabling the system to maintain a safe, stable, and efficient operating state when facing photovoltaic fluctuations. Therefore, this invention not only proposes a two-layer framework combining prediction correction and constraint tightening at the theoretical level, but also demonstrates its significant advantages in accuracy and robustness in empirical results, providing a feasible and efficient solution for real-time energy management of integrated energy systems in scenarios with high proportions of photovoltaic access.

[0062] Example 3 The output prediction system for real-time energy management of the integrated energy system of the present invention includes: The acquisition module is used to acquire corrected prediction values ​​with confidence intervals; The solution module is used to take the corrected prediction value with confidence interval as input, and with the goal of minimizing the operating cost of the integrated energy system and maintaining supply and demand balance, to construct an objective function, solve the objective function, and obtain a fault-tolerant energy scheduling scheme.

[0063] In this embodiment, the process of obtaining the corrected predicted value with confidence interval is as follows: Predicted solar radiation values ​​for a predetermined time in the future; The predicted solar radiation values ​​at a predetermined future time and their corresponding measured values ​​are input as an observation sequence into a Hidden Markov Model (HMM) to obtain corrected prediction values ​​with confidence intervals.

[0064] The module division in this embodiment is illustrative and represents only one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules in each embodiment of this application can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0065] Example 4 A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of an output prediction method for real-time energy management of an integrated energy system. For example, the steps include: obtaining a corrected prediction value with a confidence interval; using the corrected prediction value with the confidence interval as input, constructing an objective function with the goal of minimizing the operating cost of the integrated energy system and maintaining supply-demand balance; solving the objective function to obtain a fault-tolerant energy scheduling scheme. The memory may include main memory, such as high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device. The processor, network interface, and memory are interconnected via an internal bus, which can be an industry-standard architecture bus, a peripheral component interconnection standard bus, an extended industry-standard architecture bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The memory stores the program; specifically, the program may include program code, which includes computer operation instructions. The memory may include main memory and non-volatile memory, and provides instructions and data to the processor.

[0066] Example 5 A computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of an output prediction method for real-time energy management of an integrated energy system. For example, the method includes: obtaining a corrected prediction value with a confidence interval; using the corrected prediction value with the confidence interval as input, constructing an objective function with the goal of minimizing the operating cost of the integrated energy system and maintaining supply-demand balance; solving the objective function to obtain a fault-tolerant energy scheduling scheme. Specifically, the computer-readable storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. The volatile memory may include random access memory (RAM) and / or cache memory, etc. The non-volatile memory may include read-only memory (ROM), hard disk, flash memory, optical disk, magnetic disk, etc.

[0067] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0068] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0069] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0070] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0071] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and disclosure of the invention. This application is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0072] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

[0073] The above description is merely a preferred embodiment of the present invention and does not constitute any limitation on the present invention. Any simple modifications, alterations, or equivalent structural changes made to the above embodiments based on the technical essence of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for real-time energy management and output prediction in an integrated energy system, characterized in that, include; Obtain the corrected predicted value with confidence interval; Using the corrected prediction value with confidence interval as input, and aiming to minimize the operating cost of the integrated energy system while maintaining supply and demand balance, an objective function is constructed, and the objective function is solved to obtain a fault-tolerant energy dispatch scheme.

2. The output prediction method for real-time energy management of an integrated energy system according to claim 1, characterized in that, The process of obtaining the corrected predicted value with confidence interval is as follows: Predicted solar radiation values ​​for a predetermined time in the future; The predicted solar radiation values ​​at a predetermined future time and their corresponding measured values ​​are input as an observation sequence into a Hidden Markov Model (HMM) to obtain corrected prediction values ​​with confidence intervals.

3. The output prediction method for real-time energy management of an integrated energy system according to claim 1, characterized in that, The objective function is: in, For grid interaction costs; Costs related to battery aging; / Punishment for wasting heat / hydrogen energy; Penalty for changes in equipment handling; To constrain the over-boundary relaxation penalty, T represents the predictive control time domain.

4. The output prediction method for real-time energy management of an integrated energy system according to claim 1, characterized in that, The constraints corresponding to the objective function include energy storage power constraints and energy state constraints.

5. The output prediction method for real-time energy management of an integrated energy system according to claim 4, characterized in that, The energy storage power constraint is: The energy state constraint (SoE) is: ， in, This is the power / energy uncertainty estimate converted from the prediction error. As slack variables, slight out-of-bounds errors are allowed, and the penalty weights are applied. control.

6. The output prediction method for real-time energy management of an integrated energy system according to claim 1, characterized in that, The process of solving the objective function to obtain a fault-tolerant energy scheduling scheme is as follows: The confidence interval is taken as the uncertainty range; By tightening constraints within the uncertain range, a fault-tolerant energy scheduling scheme is obtained.

7. An output prediction system for real-time energy management of an integrated energy system, characterized in that, include; The acquisition module is used to acquire corrected prediction values ​​with confidence intervals; The solution module is used to take the corrected prediction value with confidence interval as input, and with the goal of minimizing the operating cost of the integrated energy system and maintaining supply and demand balance, to construct an objective function, solve the objective function, and obtain a fault-tolerant energy scheduling scheme.

8. The output prediction system for real-time energy management of an integrated energy system according to claim 7, characterized in that, The process of obtaining the corrected predicted value with confidence interval is as follows: Predicted solar radiation values ​​for a predetermined time in the future; The predicted solar radiation values ​​at a predetermined future time and their corresponding measured values ​​are input as an observation sequence into a Hidden Markov Model (HMM) to obtain corrected prediction values ​​with confidence intervals.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the output prediction method for real-time energy management of the integrated energy system as described in any one of claims 1-6.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the output prediction method for real-time energy management of the integrated energy system as described in any one of claims 1-6.

Images