Optimal dispatching method and system considering global-local balance responsibility risk

Abstract

The present application relates to the technical field of optimization scheduling, and more particularly to an optimization scheduling method and system considering global-local balance responsibility risk, the method steps comprising: analyzing the unbalance mechanism of micro-scheduling balance area; quantifying the unbalanced operation risk considering power grid scheduling-actual output; establishing a risk coordination control model considering global balance responsibility and an economic scheduling model; and proposing an optimization scheduling model considering global-local balance responsibility risk. The present application solves the problem that the existing risk quantification evaluation only considers a single dimension, and lacks unified modeling and scheduling decision-making from the perspective of global-local balance responsibility of the system. The present application achieves the purpose of encouraging the subject to actively assume the system imbalance and obtaining the optimal risk and benefit level.

Description

Technical Field

[0001] This invention relates to the field of optimization scheduling technology, and in particular to an optimization scheduling method and system that considers the global-local balance responsibility risk. Background Technology

[0002] With the rapid growth of new energy installed capacity and the continuous increase in the penetration rate of distributed photovoltaic power, the uncertainty on both the source and load sides of the power system has significantly increased, and the difficulty of system power balance has continued to increase. Renewable energy sources such as wind power and photovoltaic power are characterized by strong output fluctuations and large prediction errors. Their power deviations exhibit significant dynamic changes in both time and space dimensions. Coupled with factors such as random load fluctuations and uncertainties in electricity market prices, system imbalances occur frequently, posing a severe challenge to the safe and stable operation of the power grid.

[0003] Meanwhile, due to the large number of power grid topologies and main types in the system, the task is quite difficult and the feasibility verification is challenging. Therefore, it is crucial to consider developing an optimized scheduling strategy for a micro-schedule balancing zone with balancing potential. Summary of the Invention

[0004] This invention provides an optimized scheduling method and system that considers global-local balance responsibility risk, which can effectively solve the problems in the background art.

[0005] This invention provides an optimized scheduling method that considers global-local balance responsibility risk, comprising the following steps: Analyze the imbalance mechanism in the micro-scheduling equilibrium zone and determine the composition and distribution characteristics of the imbalance quantity; Based on the composition and distribution characteristics of imbalance, the risk of unbalanced operation between power grid dispatch and actual output is quantitatively considered. With the goal of minimizing the risk of unbalanced operation of the system on the same day, and in combination with the main body's scheduling output constraints, a risk collaborative control model considering global balance responsibility is constructed; at the same time, with the goal of maximizing the main body's additional revenue on the same day after optimized scheduling, and in combination with the main body's actual output constraints, an economic scheduling model considering global balance responsibility is constructed. By integrating the risk collaborative control model and the economic dispatch model into a unified solution framework, an optimized dispatch model considering global-local balance responsibility risk is formed. At the global level, the model uses the risk collaborative control model to calculate multiple system risk thresholds, and solves multiple benefit-risk trade-off dispatch schemes through these thresholds, and selects the dispatch scheme based on Pareto optimality. At the local level, the entire day is divided into multiple local time blocks, and rolling optimization dispatch is performed to output the optimal grid dispatch load and main body actual output scheme.

[0006] Furthermore, the operational risks of imbalance between power grid dispatch and actual output include the operational risks of system regulation capacity imbalance, the operational risks of deviation in the balance responsibility borne by the main system entity, and the operational risks of system imbalance.

[0007] Furthermore, the specific quantitative formula for considering the balance risk of the system's regulatory capacity is as follows: ; ; In the formula, Let be the lower bound for the equilibrium of the nth type of subject in time period t; The number of entities in the nth category; and For the nth type and i-th subject, the actual output and dispatched output during the t-th time period; To support the maximum adjustment capacity in time period t, which is the time period t. To support the balance ratio coefficient of the i-th subject of the n-th category in the t-th time period; The directional balance coefficient is the coefficient for time period t. To consider the risk of the nth type of subject in time period t, taking into account the system's adjustment capacity; The specific quantitative formula for considering the operational risk of the system's main body's deviation in balancing responsibility is as follows: ; In the formula, To account for the risk in time period t of the nth type of entity that bears the imbalance of responsibility for the system's main body; and For the nth type and i-th subject, during the t-th time period, the actual output-dispatch deviation and the initial output-dispatch deviation are respectively. To assume balancing responsibility for the i-th subject of the n-th category during the t-th time period; The specific quantitative formula for considering the operational risk of system imbalance is as follows: ; In the formula, To consider the risk of the nth type of subject in the t-th time period for system imbalance.

[0008] Furthermore, the risk collaborative control model is specifically as follows: ; In the formula, N represents the total number of system imbalance categories; T represents the total number of categories over a given time period.

[0009] Furthermore, the constraints of the risk collaborative control model include upper and lower limits of the main dispatch output, the magnitude of the main dispatch output, the constant system dispatch load output before and after dispatch, and the ramp-up constraint of the main dispatch output.

[0010] Furthermore, the economic scheduling model is specifically as follows: ; In the formula, N represents the total number of system imbalance categories; T represents the total number of categories over a given time period. , , , These represent the new spot market revenue, balanced responsibility incentive revenue, penalty expenses, and cost expenses for the nth category and the tth time period, respectively.

[0011] Furthermore, the additional spot market revenue is specifically as follows: ; The specific benefits of balancing responsibility incentives are as follows: ; ; The specific penalty fees are as follows: ; The specific costs and expenses are as follows: ; In the formula, , , For the t-th time period, the spot electricity price, the balancing responsibility incentive electricity price, and the assessment penalty electricity price; The initial actual output of the i-th subject in the n-th category during the t-th time period; , Let α be the optimized scheduling cost and initial cost of the i-th subject in the n-th category during the t-th time period; α be the penalty coefficient for assuming balancing responsibility; and Δt be the time step.

[0012] Furthermore, the constraints of the economic dispatch model include upper and lower limits of the actual output of the main body, upper and lower limits of the output of new energy sources, main body ramp-up constraints, and balance risk constraints of the system's regulation capacity.

[0013] Furthermore, the optimized scheduling model considering global-local balance of responsibility and risk is specifically as follows at the global level: The daily system imbalance risk obtained from the risk collaborative control model is denoted as R; the daily increase in main entity revenue after optimized scheduling obtained from the economic scheduling model is denoted as C. Calculate the system risk threshold: ; and ; In the formula, Let the system risk threshold be the j-th iteration. , e represents the lower limit of the system risk threshold; e represents the adjusted system risk threshold interval. Under different system risk thresholds, the risk collaborative control model and economic scheduling model are solved to obtain a set of benefit-risk trade-off scheduling schemes; For each set of payoff-risk tradeoff scheduling schemes, a comprehensive evaluation function is used to evaluate them. The comprehensive evaluation function is as follows: F = (1-θ)×R + θ×C; In the formula, θ is the weighting coefficient of the economic scheduling objective function; All scheduling schemes are evaluated and then selected to obtain a scheduling scheme that satisfies the Pareto optimality.

[0014] Furthermore, the optimal scheduling model considering the global-local balance of responsibility and risk is specifically as follows at the local level: The entire day is divided into local time blocks of varying lengths, and the local risk thresholds are scaled proportionally: ; ; In the formula, Define the system risk threshold for the j-th local time period; , This is the lower limit of the system risk threshold; The system risk threshold interval is adjusted; Let x be the length of the xth local time interval; The length of a day; The rolling optimization scheduling is specifically as follows: In the future The goal is to maximize the system's operational benefits within a given time period. This is combined with local risk margins to optimize the actual output behavior of the main entities and the grid dispatch load. The constraints of the rolling optimization dispatch are consistent with those of the risk collaborative control model and the economic dispatch model. The optimal economic and safe power system dispatch result is obtained by iteratively solving the problem using the particle swarm optimization algorithm.

[0015] This invention also provides an optimized scheduling system that considers global-local balance responsibility risk, comprising: The analysis module is used to analyze the imbalance mechanism in the micro-scheduling equilibrium zone and determine the composition and distribution characteristics of the imbalance. The quantification module is used to quantify the operational risk of imbalance between power grid dispatch and actual output based on the composition and distribution characteristics of the imbalance. The modeling module is used to construct a risk collaborative control model that considers global balance responsibility, with the goal of minimizing the risk of unbalanced operation of the system on the same day and in combination with the main body's scheduling output constraints; at the same time, with the goal of maximizing the main body's additional revenue on the same day after optimized scheduling, an economic scheduling model that considers global balance responsibility is constructed in combination with the main body's actual output constraints. The solution module integrates the risk collaborative control model and the economic dispatch model into a unified solution framework, forming an optimized dispatch model that considers global-local balance responsibility risks. At the global level, the model uses the risk collaborative control model to calculate multiple system risk thresholds, solves multiple benefit-risk trade-off dispatch schemes through these thresholds, and selects the dispatch scheme based on Pareto optimality. At the local level, the entire day is divided into multiple local time blocks, and rolling optimization dispatch is performed to output the optimal grid dispatch load and main body actual output scheme.

[0016] The technical solution of this invention can achieve the following technical effects: This invention, through quantitative analysis considering the system's regulation capacity, system imbalance, and the risk of deviation in the balance responsibility borne by the main body, constructs an economic dispatch model that considers the risk of global balance responsibility and collaborative control. It introduces an incremental system risk threshold and uses particle swarm optimization algorithm and Pareto optimality method to solve the optimal power grid dispatch scheme. At the same time, it divides local time periods for rolling optimization dispatch, so as to incentivize the main body to actively assume the system imbalance and obtain the optimal risk and benefit level. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a framework diagram of the optimal scheduling solution considering global-local balance responsibility risk in this invention; Figure 2 This is a standardized output diagram of each power generation entity under the original conditions of this invention; Figure 3 This is a spatiotemporal distribution diagram of the risks of each entity in the system under Scheme 1 of the present invention; Figure 4 This is a diagram illustrating the system risk-benefit situation of adjusting the risk threshold under Scheme 2 of the present invention; Figure 5 This is a risk-return diagram of the differentiated time-segmented block system for adjusting risk thresholds under Scheme 3 of the present invention. Detailed Implementation

[0019] 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.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0021] This invention relates to an optimized scheduling method that considers global-local balance responsibility risk, and the steps and main contents include: S1: Analyze the imbalance mechanism in the micro-scheduling equilibrium zone and determine the composition and distribution characteristics of the imbalance quantity; S2: Based on the composition and distribution characteristics of the imbalance, quantitatively consider the operational risk of imbalance between power grid dispatch and actual output; S3: With the goal of minimizing the risk of unbalanced operation of the system on the same day, and in combination with the main body's scheduling output constraints, a risk collaborative control model considering global balance responsibility is constructed; at the same time, with the goal of maximizing the main body's new revenue on the same day after optimized scheduling, and in combination with the main body's actual output constraints, an economic scheduling model considering global balance responsibility is constructed. S4: The risk collaborative control model and the economic dispatch model are integrated into a unified solution framework to form an optimized dispatch model that considers global-local balance responsibility risk. At the global level, the model uses the risk collaborative control model to calculate multiple system risk thresholds, solves multiple benefit-risk trade-off dispatch schemes through multiple system risk thresholds, and selects dispatch schemes based on Pareto optimality. At the local level, the entire day is divided into multiple local time blocks, and rolling optimization dispatch is performed to output the optimal grid dispatch load and main body actual output scheme.

[0022] In order to achieve optimized scheduling from the perspective of the imbalance mechanism of the micro-scheduling balance region, this embodiment first analyzes the imbalance mechanism of the micro-scheduling balance region.

[0023] When grid dispatching results are translated from planned instructions into actual generator operation, power deviations inevitably occur. These deviations stem not only from minor fluctuations in active power output of traditional thermal power units due to equipment status and fuel characteristics, but also from the strong uncertainties and instantaneous power changes of renewable energy sources such as wind and solar power influenced by complex factors like weather and cloud cover, as well as unpredictable random variations on the load side. Therefore, the combined effect of these factors results in a persistent and significant power difference between actual power generation and real-time dispatching instructions—a system imbalance.

[0024] To address the aforementioned power imbalance problem and improve system regulation efficiency on a smaller spatial and shorter time scale, the concept of a "micro-dispatch balancing zone" is proposed. A "micro-dispatch balancing zone" refers to a region with abundant distributed photovoltaic and flexible load resources, sufficient flexible regulation capabilities, and the capacity and potential for regional power balancing.

[0025] In this method, the imbalance mechanism of the micro-dispatch balance zone is used to quantitatively consider the operational risk of imbalance between power grid dispatch and actual output.

[0026] Because different entities exhibit significantly different actual load output characteristics under varying external environments and behavioral traits, the resulting system security and stability risks also differ. However, entities of the same type often share high similarities in production patterns, load characteristics, and scheduling responses. To facilitate systematic analysis, this embodiment categorizes entities and uses a unified modeling and quantitative assessment of the potential system risks they may trigger based on typical categorized characteristics.

[0027] In theory, the functional relationship between different types of subjects is not unique. Considering that a further optimized scheduling model needs to be established, a simplified model is considered, and it is assumed that the overall system risk Rt at time t is the linear sum of the risks of each subject type.

[0028] ; In the formula, N is the number of main types, where n=1 represents thermal power, n=2 represents wind power, and n=3 represents photovoltaic power; The contribution of the nth type of entity to system risk; Let n be the risk of the nth subject at time t.

[0029] Furthermore, the system risks within the micro-dispatch balancing zone primarily stem from the discrepancy between grid dispatch results and actual power output. This discrepancy is influenced by various factors, including market grid connection assessment rules and the volatility of renewable energy output. The accumulation of these discrepancies at the system level can weaken the system's real-time balancing capability and pose a potential threat to safe and stable operation. Therefore, this method, from a system operation perspective, focuses on the imbalance risks caused by the inconsistency between grid dispatch and actual power output. Specifically, this includes: balancing risks considering the system's regulation capacity; operational risks considering the deviation in the system's balancing responsibility; and operational risks considering the amount of system imbalance.

[0030] 1. Consider the balance risk of system regulation capacity. The smaller the system's positive regulation capacity, the greater the downward fluctuation trend of new energy sources, the larger the negative error of grid dispatch versus actual output, and the greater the balance risk within the region. Based on the risk quantification method of the absorbable domain, combined with the definition of the regulation capacity of the regulating entity, this paper proposes a risk quantification method. Let t be the lower bound for the equilibrium of the nth type of agent. Further, a risk quantification method considering the system's adaptability is defined as follows: ; ; In the formula, Let be the lower bound for the equilibrium of the nth type of subject in time period t; The number of entities in the nth category; and For the nth type and i-th subject, the actual output and dispatched output during the t-th time period; To support the maximum adjustment capacity in time period t, which is the time period t. To support the balance ratio coefficient of the i-th subject of the n-th category in the t-th time period; is the directional balance coefficient for time period t, used to indicate the direction of system imbalance. It is set to 1 when the system imbalance direction is positive and -1 when the system imbalance direction is negative. Considering the risk of the nth type of subject in time period t, which is related to the system's adjustment capability.

[0031] Among them are: ; ; ; In the formula, Let t be the system imbalance quantity at time t; , Let be the maximum rise and fall rate of the i-th subject in the n-th class.

[0032] 2. Consider the operational risk of deviations in the balancing responsibilities borne by the main system entities. In the micro-scheduling balancing zone environment, considering the balancing responsibilities borne by each entity and the imbalances of the entities before and after optimization, this deviation directly reflects the operational risk borne by the entities, as follows: ; ; ; In the formula, To account for the risk in time period t of the nth type of entity that bears the imbalance of responsibility for the system's main body; and For the nth type and i-th subject, during the t-th time period, the actual output-dispatch deviation and the initial output-dispatch deviation are respectively. To assume balancing responsibility for the i-th subject of the n-th category during the t-th time period.

[0033] The first step of this paper is to predict the market imbalance of the micro-schedule balance zone based on the market clearing load and actual output characteristics of the previous N days and T time periods, and to use typical time series forecasting models (such as LSTM models) to predict the market imbalance of the next one or more time periods. At the same time, a sliding window is adopted to further predict the system imbalance of the next day's 96 time periods or even multiple days. The second step is to select typical load characteristics such as the imbalance of each entity, installed capacity, and actual output to calculate the imbalance responsibility allocated to each entity.

[0034] ; Among them, new energy sources do not bear the responsibility of balancing upward adjustment.

[0035] ; In the formula, ω1, ω2, and ω3 are the balance responsibility weight coefficients that take into account the main imbalance, installed capacity, and actual output. This represents the predicted system imbalance.

[0036] 3. Consider the operational risks associated with system imbalances. In a micro-scheduling equilibrium environment, various entities may deviate from scheduling instructions during operation. This deviation directly reflects the operational risks borne by the entity, as follows: ; In the formula, To consider the risk of the nth type of subject in the t-th time period for the system imbalance.

[0037] Based on the method for quantifying the operational risk of imbalance between power grid dispatch and actual output, and combined with constraints, this model collaboratively considers the system's regulation capacity, system imbalance, and operational risk related to the deviation of the system's main body's balancing responsibility. The risk collaborative control model, which considers the overall balancing responsibility, takes minimizing the daily system operational risk R as its objective function. ; In the formula, N represents the total number of system imbalance categories; T represents the total number of categories over a given time period.

[0038] The constraints of the risk collaborative control model include upper and lower limits of the main dispatch output, the magnitude of the main dispatch output, the constant system dispatch load output before and after dispatch, and the ramp-up constraint of the main dispatch output.

[0039] 1. Main Dispatch Output Limits. Each participant's planned power generation must remain within its actual power generation limits. For thermal power, this limit is from installed capacity to minimum technical output; for renewable energy, the lower limit is 0.

[0040] ; ; ; In the formula, , , For the maximum load, minimum load, and minimum technical output ratio of the i-th entity in the n-th category.

[0041] 2. Main Dispatch Output Range. Based on the initial dispatch output, there are also upper and lower limits: ; In the formula, ε1 and ε2 are the upper and lower limits of the main dispatch output.

[0042] 3. The system's dispatched load output remains constant before and after dispatch. Based on the assumption that the system load-side output is constant, the dispatched output on the generation side before and after dispatch is also constant.

[0043] .

[0044] 4. Main Dispatch Output Ramp-up Constraints. The main dispatch output has upper and lower limits on its rate of change between adjacent time periods: .

[0045] This method further incorporates the differences in cost models among different types of entities, as well as constraints, and considers an economic scheduling model with global balancing responsibility, aiming to maximize the daily additional revenue C of the entity after optimization. ; In the formula, N represents the total number of system imbalance categories; T represents the total number of categories over a given time period. , , , These represent the new spot market revenue, balanced responsibility incentive revenue, penalty expenses, and cost expenses for the nth category and the tth time period, respectively.

[0046] The specific revenue from the new spot market is as follows: ; The specific benefits of balancing responsibility incentives are as follows: ; ; The specific penalty fees are as follows: ; The specific costs and expenses are as follows: ; In the formula, , , For the t-th time period, the spot electricity price, the balancing responsibility incentive electricity price, and the assessment penalty electricity price; The initial actual output of the i-th subject in the n-th category during the t-th time period; , Let α be the optimized scheduling cost and initial cost of the i-th subject in the n-th category during the t-th time period; α be the penalty coefficient for assuming balancing responsibility; and Δt be the time step.

[0047] The cost models differ somewhat for different types of entities, as detailed below: 1. Thermal power units: When thermal power units operate within their conventional output range, only conventional coal consumption costs exist, which are typically calculated using consumption characteristics and are quadratic functions: .

[0048] 2. In the process of wind and solar new energy power generation, its operating costs mainly include the costs of wind curtailment and solar curtailment.

[0049] ; In the formula, Forecast wind and solar power output for time period t; Electricity prices will be penalized for curtailing wind and solar power.

[0050] The constraints of the economic dispatch model include the upper and lower limits of the actual output of the main body, the upper and lower limits of the output of new energy sources, the main body's ramp-up constraint, and the balance risk constraint of the system's regulation capacity.

[0051] 1. Limits on the actual output of the main body. The actual output of the main body must be kept within its limits: .

[0052] 2. Upper and lower limits of renewable energy output. Based on the assumption that the initial actual output of renewable energy is the predicted output under the original conditions, i.e., no wind or solar curtailment, and considering that renewable energy does not have the ability to adjust upwards and has an upper limit on the curtailment rate, its actual output has upper and lower limits: ; ; In the formula, η is the upper limit of the new energy curtailment rate of the i-th subject in the n-th category.

[0053] 3. Main body ramp-up constraint. The actual output of the main body in adjacent time periods has upper and lower limits for the rate of change: .

[0054] 4. Risk constraints on balancing system regulation capacity. Sufficient regulation capacity is needed to maintain system balance in case of system imbalance. .

[0055] The optimal scheduling model considering global-local balance of responsibility and risk is specifically as follows at the global level: The daily system imbalance risk obtained from the risk collaborative control model is denoted as R; the daily increase in main entity revenue after optimized scheduling obtained from the economic scheduling model is denoted as C. Calculate the system risk threshold: ; and ; In the formula, Let the system risk threshold be the j-th iteration. , e represents the lower limit of the system risk threshold; e represents the adjusted system risk threshold interval. Under different system risk thresholds, the risk collaborative control model and economic scheduling model are solved to obtain a set of benefit-risk trade-off scheduling schemes; For each set of payoff-risk tradeoff scheduling schemes, a comprehensive evaluation function is used to evaluate them. The comprehensive evaluation function is as follows: F = (1-θ)×R + θ×C; In the formula, θ is the weighting coefficient of the economic scheduling objective function; All scheduling schemes are evaluated and then selected to obtain a scheduling scheme that satisfies the Pareto optimality.

[0056] The joint decision-making of scheduling load and actual output of the entities jointly affects the system's operational risk and the entities' revenue level. In the economic scheduling model C, the actual output of the entities is used as the decision variable to achieve revenue optimization; while in the risk control model R, the scheduling load is used as the decision variable to suppress system operational risk. However, since the models involve multi-type entities and multi-time period coupled decisions, and the objective function and constraints exhibit significant nonlinear and nonconvex characteristics, directly using intelligent optimization algorithms such as particle swarm optimization to uniformly solve the above multi-objective problem faces significant computational difficulties. In addition, there is an inherent conflict between economic objectives and risk control objectives. If a sequential optimization approach is adopted, performing risk optimization after obtaining the optimal economic benefit may lead to a significant decrease in entity revenue; and vice versa. Furthermore, this conflicting objective characteristic easily leads to a decrease in the convergence performance of intelligent optimization algorithms, and may even result in premature convergence or instability.

[0057] Furthermore, if economic scheduling objectives and risk control objectives are completely integrated into the same optimization model, the solution obtained is usually an overall compromise optimal solution in the sense of multiple objectives, rather than an optimal solution in the single dimension of profit or risk, because the two types of objectives have fundamental differences in the optimization direction (maximizing benefits and minimizing risks). This compromise solution takes into account the system's economy and operational safety to a certain extent, but inevitably sacrifices the ultimate performance of a single objective.

[0058] Therefore, this embodiment moderately weakens the risk constraint during global modeling, introducing global system risk as a soft constraint in the form of a threshold. By gradually increasing and relaxing the risk threshold, the global profit maximization problem is iteratively solved under different risk tolerance levels, thereby obtaining a set of profit-risk trade-off solutions. Finally, based on the weight coefficients of the two objective functions, the obtained solution set is comprehensively evaluated, and a scheduling scheme that satisfies the Pareto optimality is selected.

[0059] Building upon this, if the ideal return-risk objective function is solved only from the global system level, the model dimensions and the scale of decision variables will significantly expand with the increase in the number of participating entities and the refinement of the time scale. The search space will grow exponentially, easily leading to problems such as decreased convergence speed, susceptibility to local optima, and computational efficiency failing to meet the timeliness requirements of intraday scheduling decisions. If only the local time level is considered, and the system's operational risk or return is independently optimized for a certain period or several adjacent periods during intraday operation, the cumulative effect of imbalance risk over time and its cross-period transmission characteristics may be ignored.

[0060] The optimal scheduling model considering global-local balance of responsibility and risk is specifically as follows at the local level: Divide the entire day into local time blocks of varying lengths (e.g., 96 time slots, each 15 minutes long). ,for example The values ​​are 4, 8, 12, 16, 24, 32, and 48, which correspond to 1, 2, 3, 4, 6, 8, and 12 hours, respectively.

[0061] Scaling local risk thresholds proportionally: ; ; In the formula, Define the system risk threshold for the j-th local time period; , This is the lower limit of the system risk threshold; The system risk threshold interval is adjusted; Let x be the length of the xth local time interval; The length of a day.

[0062] The rolling optimization scheduling is specifically as follows: In the future The goal is to maximize the system's operational benefits within a given time period. This is combined with local risk margins to optimize the actual output behavior of the main entities and the grid dispatch load. The constraints of the rolling optimization dispatch are consistent with those of the risk collaborative control model and the economic dispatch model. The optimal economic and safe power system dispatch result is obtained by iteratively solving the problem using the particle swarm optimization algorithm.

[0063] In summary, this embodiment considers constructing a micro-scheduling balance zone optimization scheduling model that takes into account global-local balance responsibility risks. The framework is as follows: Figure 1 First, an objective function is constructed at the global system level to maximize the main entity's revenue and minimize its risk. Based on this, multiple types of constraints and progressively increasing system risk thresholds are considered, and the particle swarm optimization algorithm is used to solve the problem. Second, the actual output behavior of entities responsible for balancing and the changes in dispatched load are analyzed, providing operational boundaries and initial values ​​for subsequent local-level solutions. Finally, due to the smaller local time scale, stricter risk control measures are adopted, and rolling scheduling optimization is further considered, taking into account time-segmented scheduling, output ramp-up constraints, and local risk thresholds. Ultimately, the ideal main entity output behavior and dispatched load are solved to achieve economical and safe power system dispatch and risk control.

[0064] Furthermore, the local rolling scheduling model is based on the future To maximize system operating benefits within a given time period, this approach combines local risk margins with optimization of actual power output behavior and grid dispatch load. The Particle Swarm Optimization (PSO) algorithm is then employed to obtain the optimal economic and secure power system dispatch result. The constraints of the risk-coordinated control model involving constraints on regulation and global balance responsibility, as well as the economic dispatch model considering global balance responsibility risk, are consistent.

[0065] It should be noted that since most countries and regions only publish the overall load information of provinces / regions, and the load behavior of each entity is not publicly available data, this embodiment selects a certain market platform and a certain region in Denmark that includes 9 typical power generation entities of thermal power, 9 wind power, and 6 photovoltaic power as the research object. The basic parameters of the installed capacity of each type of power generation entity are shown in Table 1 below.

[0066] Table 1. Basic Information of Market Entities .

[0067] This embodiment selects a typical day as the simulation scenario. Figure 2 This involves standardizing the efforts of each entity under the original conditions. Balancing incentive prices. To balance market electricity prices, penalties for curtailing wind and solar power will be imposed. , And performance evaluation and penalty electricity prices All values ​​are taken as 20 (EUR / MWh); the secondary, primary, and constant cost coefficients of coal consumption for the nine thermal power plants are referenced from IEEE 118 node data. The balance ratio coefficient φ supporting the nth type of power plant in time period t is also used. n,t Allocation based on installed capacity ratio means fully utilizing regulatory resources to equally support all types of entities; the contribution of the nth type of entity to system risk. Both are consistent and are 1. , The system risk threshold is set at 40,000 MW and 60,000 MW, respectively; e represents the adjusted system risk threshold interval of 10 MW; system risk and benefit are considered proportionally, i.e., θ is 0.5; the upper and lower limits of the main dispatch output amplitude ε1 and ε2 are set at 0.9 and 1.1, respectively; the penalty coefficient α for assuming balancing responsibility is 0.3; and the upper limit of the renewable energy curtailment rate η is 0.9. Simultaneously, the local time period is divided into T... x =4, 8, 12, 16, 24, 32, 48, to represent 1, 2, 3, 4, 6, 8, 12 hours respectively.

[0068] Meanwhile, to verify the effectiveness of the proposed method and model, this paper also designed three schemes for comparative analysis.

[0069] Option 1: Only implement overall risk control.

[0070] Option 2: Overall economic scheduling and risk control.

[0071] Option 3: Global-local economic scheduling and risk control.

[0072] Figure 3This section illustrates the spatiotemporal distribution of risks for each entity under Scheme 1, based on the power output adjustment behavior of the balancing responsibility mechanism. From the perspective of entity type, photovoltaic (PV) has the lowest risk because its renewable energy regulation capacity relies primarily on wind and solar curtailment, leading to reduced spot market revenue and penalties for curtailment. Therefore, it has less incentive to assume balancing responsibility. Thermal power has the highest risk because its units have strong regulation capabilities and adjust their output behavior to obtain spot market access and balancing incentives. Furthermore, the increase in installed capacity and actual output enhances the flexibility of output adjustment, making thermal power, the seventh entity with the largest installed capacity and output load, the one with the greatest balancing risk. From a temporal distribution perspective, since renewable energy lacks upward regulation capabilities, its risk is almost zero during dispatching, actual output, and the multiple early morning and evening periods when the balancing responsibility is zero.

[0073] Figure 4 The system risk and benefit of 2000 risk threshold adjustments under Scheme 2 are shown in Table 2, which includes 5 Pareto frontiers.

[0074] Table 2. Risk and return profile of differentiated solutions .

[0075] The scheduling schemes corresponding to the Pareto front points mentioned above can no longer further improve system benefits without worsening system risks, or further reduce system risks without reducing system benefits, within the current feasible region. This reflects the optimal trade-off boundary between economy and security.

[0076] Furthermore, in adjusting the risk margin, Scheme Two maintains a risk level almost entirely between 32,000 and 37,000 MW. Based on the normalized upper and lower limits of risk and return, the optimal point is Pareto1, resulting in a market risk of 32,317.30 MW for the selected dispatch scheme, corresponding to a return of 85,943.30 EUR. Compared to Scheme One, the risk decreases to 25,044.32 MW, but the return is negative, decreasing by more than 1.5 times. This indicates that when only risk control is considered, it is difficult to simultaneously achieve economic benefits.

[0077] Based on the globally optimal economic risk scheduling results, seven different local time-sharing partitioning schemes are further considered. To reasonably reduce computational complexity while ensuring solution accuracy, this paper increases the search interval of the system risk threshold by 10 times to reduce the number of iterations. On this basis, a micro-scheduling balancing scheme that takes into account both global and local risk responsibilities is constructed, and its risk and benefit performance is compared and analyzed. The relevant results are shown in Table 3.

[0078] Table 3. Risk and Return Analysis of Differentiated Time-Segmented Blocks .

[0079] The results show that as the length of the time-segment blocks decreases (i.e., the number of time segments increases), the system risk and return levels improve to varying degrees in most cases, such as... Figure 5 Furthermore, the optimization effect gradually increases with the refinement of the local time period division, verifying that time decomposition plays a positive role in tapping the potential of local scheduling.

[0080] However, the specific scheme is not entirely monotonically increasing. For example, when the number of time periods... When the time interval is 12 hours, local optimization shows a significant increase in returns within the first 12 hours, rising from 33044.29 to 42423.94, an increase of nearly 30%. However, due to the weakening of the overall coordination effect of cross-time period coupling constraints during local time period optimization, the resource coordination capability between the two 12-hour periods is limited, leading to a decrease in the utilization efficiency of adjustable resources in the latter 12-hour period, with the return decrease exceeding the return increase in the first 12 hours. Ultimately, the sum of the returns from the two local time periods is lower than the total system return obtained from global optimization throughout the entire time period.

[0081] And with Further reduction in time complexity reveals significant advantages in risk and reward. On one hand, the substantial reduction in solution dimensionality effectively avoids the PSO algorithm's tendency to get trapped in local extrema in high-dimensional nonlinear spaces. The progressively reduced search accuracy and convergence quality of the particle swarm optimization within time-segmented blocks are greatly improved, thereby uncovering deep solution spaces "missed" by the global model due to computational complexity. On the other hand, the time-decoupled micro-scheduling strategy is more sensitive to capturing short-term risks and electricity price signals. This high-frequency and aggressive response mode, while sacrificing some long-term resource smoothness, maximizes the arbitrage potential of flexible resources in the instantaneous dimension. Although this process correspondingly increases the overall computation time, it ultimately achieves a globally-locally optimal scheduling scheme considering both risk and reward.

[0082] In summary, this invention constructs a risk quantification method that incorporates system regulation capacity, system imbalance, and deviations in the balance responsibility borne by the main entities. Based on this, an economic dispatch model considering global-local balance responsibility and risk-coordinated control is built, and an incremental risk threshold and the PSO algorithm are combined with Pareto optimality to find the optimal power grid dispatch scheme. Results show that the proposed method can fully consider the economic efficiency and risk level in optimized dispatch.

[0083] This invention also relates to an optimized scheduling system that considers global-local balance responsibility risk, comprising: The analysis module is used to analyze the imbalance mechanism in the micro-scheduling equilibrium zone and determine the composition and distribution characteristics of the imbalance. The quantification module is used to quantify the operational risk of imbalance between power grid dispatch and actual output based on the composition and distribution characteristics of the imbalance. The modeling module is used to construct a risk collaborative control model that considers global balance responsibility, with the goal of minimizing the risk of unbalanced operation of the system on the same day and in combination with the main body's scheduling output constraints; at the same time, with the goal of maximizing the main body's additional revenue on the same day after optimized scheduling, an economic scheduling model that considers global balance responsibility is constructed in combination with the main body's actual output constraints. The solution module integrates the risk collaborative control model and the economic dispatch model into a unified solution framework, forming an optimized dispatch model that considers global-local balance responsibility risks. At the global level, the model uses the risk collaborative control model to calculate multiple system risk thresholds, solves multiple benefit-risk trade-off dispatch schemes through these thresholds, and selects the dispatch scheme based on Pareto optimality. At the local level, the entire day is divided into multiple local time blocks, and rolling optimization dispatch is performed to output the optimal grid dispatch load and main body actual output scheme.

[0084] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of the application as defined herein, and are to be considered as covering any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Thus, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. An optimized scheduling method considering global-local balance responsibility risk, characterized by the following steps: include: Analyze the imbalance mechanism in the micro-scheduling equilibrium zone and determine the composition and distribution characteristics of the imbalance quantity; Based on the composition and distribution characteristics of imbalance, the risk of unbalanced operation between power grid dispatch and actual output is quantitatively considered. With the goal of minimizing the risk of unbalanced operation of the system on the same day, and in combination with the main body's scheduling output constraints, a risk collaborative control model considering global balance responsibility is constructed; at the same time, with the goal of maximizing the main body's additional revenue on the same day after optimized scheduling, and in combination with the main body's actual output constraints, an economic scheduling model considering global balance responsibility is constructed. By integrating the risk collaborative control model and the economic scheduling model into a unified solution framework, an optimized scheduling model that considers global-local balance responsibility risk is formed. At the global level, the model uses the risk collaborative control model to calculate multiple system risk thresholds, solves multiple benefit-risk trade-off scheduling schemes through multiple system risk thresholds, and selects scheduling schemes based on Pareto optimality. At the local level, the entire day is divided into multiple local time blocks, and rolling optimization scheduling is performed to output the optimal power grid dispatch load and main actual output scheme. Among them, the operational risks of imbalance between power grid dispatch and actual output include the operational risks of imbalance considering the system's regulation capacity, the operational risks of deviation in the balance responsibility borne by the main system entity, and the operational risks of imbalance in the system. The specific quantitative formula for considering the balance risk of system regulation capacity is as follows: ； ； In the formula, Let be the lower bound for the equilibrium of the nth type of subject in time period t; The number of entities in the nth category; and For the nth type and i-th subject, the actual output and dispatched output during the t-th time period; To support the maximum adjustment capacity in time period t, which is the time period t. To support the balance ratio coefficient of the i-th subject of the n-th category in the t-th time period; The directional balance coefficient is the coefficient for time period t. To consider the risk of the nth type of subject in time period t, taking into account the system's adjustment capacity; The specific quantitative formula for considering the operational risk of the system's main body's deviation in balancing responsibility is as follows: ； In the formula, To account for the risk in time period t of the nth type of entity that bears the imbalance of responsibility for the system's main body; and For the nth type and i-th subject, during the t-th time period, the actual output-dispatch deviation and the initial output-dispatch deviation are respectively. To assume balancing responsibility for the i-th subject of the n-th category during the t-th time period; The specific quantitative formula for considering the operational risk of system imbalance is as follows: ； In the formula, To consider the risk of the nth type of subject in the t-th time period for system imbalance.

2. The optimized scheduling method considering global-local balance responsibility risk according to claim 1, characterized in that, The risk collaborative control model is as follows: ； In the formula, N represents the total number of system imbalance categories; T represents the total number of categories over a given time period.

3. The optimized scheduling method considering global-local balance responsibility risk according to claim 2, characterized in that, The constraints of the risk collaborative control model include upper and lower limits of the main dispatch output, the magnitude of the main dispatch output, the constant system dispatch load output before and after dispatch, and the ramp-up constraint of the main dispatch output.

4. The optimized scheduling method considering global-local balance responsibility risk according to claim 1, characterized in that, The economic scheduling model is as follows: ； In the formula, N represents the total number of system imbalance categories; T represents the total number of categories over a given time period. , , , These represent the new spot market revenue, balanced responsibility incentive revenue, penalty expenses, and cost expenses for the nth category and the tth period, respectively.

5. The optimization scheduling method considering global-local balance responsibility risk according to claim 4, characterized in that, The specific revenue from the new spot market is as follows: ； The specific benefits of balancing responsibility incentives are as follows: ； ； The specific penalty fees are as follows: ； The specific costs and expenses are as follows: ； In the formula, , , For the t-th time period, the spot electricity price, the balancing responsibility incentive electricity price, and the assessment penalty electricity price; The initial actual output of the i-th subject in the n-th category during the t-th time period; , Let α be the optimized scheduling cost and initial cost of the i-th subject in the n-th category during the t-th time period; α be the penalty coefficient for assuming balancing responsibility; and Δt be the time step.

6. The optimization scheduling method considering global-local balance responsibility risk according to claim 4, characterized in that, The constraints of the economic dispatch model include the upper and lower limits of the actual output of the main body, the upper and lower limits of the output of new energy sources, the main body's ramp-up constraint, and the balance risk constraint of the system's regulation capacity.

7. The optimization scheduling method considering global-local balance responsibility risk according to claim 1, characterized in that, The optimal scheduling model considering global-local balance of responsibility and risk is specifically as follows at the global level: The system imbalance risk calculated by the risk collaborative control model is denoted as R; the new revenue of the main entity after optimized scheduling calculated by the economic scheduling model is denoted as C. Calculate the system risk threshold: ； and ; In the formula, Let the system risk threshold be the j-th iteration. , This is the lower limit of the system risk threshold; e represents the adjusted system risk threshold interval; Under different system risk thresholds, the risk collaborative control model and economic scheduling model are solved to obtain a set of benefit-risk trade-off scheduling schemes; For each set of payoff-risk tradeoff scheduling schemes, a comprehensive evaluation function is used to evaluate them. The comprehensive evaluation function is as follows: F = (1-θ)×R + θ×C; In the formula, θ is the weighting coefficient of the economic scheduling objective function; All scheduling schemes are evaluated and then selected to obtain a scheduling scheme that satisfies the Pareto optimality.

8. The optimized scheduling method considering global-local balance responsibility risk according to claim 7, characterized in that, The optimal scheduling model considering global-local balance of responsibility and risk is specifically as follows at the local level: The entire day is divided into local time blocks of varying lengths, and the local risk thresholds are scaled proportionally: ； ； In the formula, Define the system risk threshold for the j-th local time period; , This is the lower limit of the system risk threshold; The system risk threshold interval is adjusted; Let x be the length of the xth local time interval; The length of a day; The rolling optimization scheduling is specifically as follows: In the future The goal is to maximize the system's operational benefits within a given time period. This is combined with local risk margins to optimize the actual output behavior of the main entities and the grid dispatch load. The constraints of the rolling optimization dispatch are consistent with those of the risk collaborative control model and the economic dispatch model. The optimal economic and safe power system dispatch result is obtained by iteratively solving the problem using the particle swarm optimization algorithm.

9. An optimized scheduling system considering global-local balance responsibility risk, characterized in that, An optimized scheduling method for implementing the global-local balance responsibility risk as described in any one of claims 1 to 8 includes: The analysis module is used to analyze the imbalance mechanism in the micro-scheduling equilibrium zone and determine the composition and distribution characteristics of the imbalance. The quantification module is used to quantify the operational risk of imbalance between power grid dispatch and actual output based on the composition and distribution characteristics of the imbalance. The modeling module is used to construct a risk collaborative control model that considers global balance responsibility, with the goal of minimizing the risk of unbalanced operation of the system on the same day and in combination with the main body's scheduling output constraints; at the same time, with the goal of maximizing the main body's additional revenue on the same day after optimized scheduling, an economic scheduling model that considers global balance responsibility is constructed in combination with the main body's actual output constraints. The solution module integrates the risk collaborative control model and the economic dispatch model into a unified solution framework, forming an optimized dispatch model that considers global-local balance responsibility risks. At the global level, the model uses the risk collaborative control model to calculate multiple system risk thresholds, solves multiple benefit-risk trade-off dispatch schemes through these thresholds, and selects the dispatch scheme based on Pareto optimality. At the local level, the entire day is divided into multiple local time blocks, and rolling optimization dispatch is performed to output the optimal grid dispatch load and main body actual output scheme.

Images