High-proportion new energy power system scheduling method and related device

Abstract

The invention relates to the field of power system automation, and discloses a high-proportion new energy power system scheduling method and a related device. Determining a look-ahead time window containing a high-risk time point in a future preset time period of the power system through supply-demand balance risk analysis; constructing a plurality of power system operation scenes by taking the look-ahead time window as a time domain, and obtaining a plurality of high-risk operation scenes by simulating the operation state of the power system in each power system operation scene; and constructing an electric power system real-time scheduling optimization model considering supply and demand imbalance cost, reserve capacity cost and operation risk cost, and solving the electric power system real-time scheduling optimization model by taking the plurality of high-risk operation scenes as a scheduling basis to obtain an electric power system scheduling scheme. And high-efficiency, stable and economical operation of the power system under high-proportion new energy access is realized.

Description

Technical Field

[0001] The present invention belongs to the field of power system automation and relates to a high-proportion new energy power system scheduling method and related devices. Background Art

[0002] With the global energy structure transforming towards clean and low-carbon directions, the proportion of renewable energy represented by wind energy and photovoltaic power in the power system is gradually increasing, and the new power system is gradually replacing the traditional power system. However, the volatility, intermittency, and uncertainty of new energy output have brought great challenges to the safe, stable, and economic operation of the power system. In particular, how to adapt to new energy consumption and power supply-demand balance in real-time scheduling decisions has become a current research hotspot and difficulty.

[0003] Currently, the status quo and challenges of new power system scheduling decisions are mainly reflected in the following aspects. First, the fluctuation and uncertainty of new energy output. Wind energy and photovoltaic power generation are significantly affected by meteorological conditions, with randomness and intermittency, resulting in difficult prediction of new energy output. Traditional scheduling methods usually rely on deterministic models and cannot effectively cope with the random fluctuations of new energy output, easily leading to the risk of power supply-demand imbalance. Second, the timeliness requirement of real-time scheduling decisions. In the context of high-proportion new energy access, the operating state of the power system changes rapidly, and real-time scheduling needs to be updated frequently to adapt to supply-demand changes. The traditional scheduling decision cycle is long, the calculation efficiency is low, and it is difficult to meet the real-time operation requirements. Third, the multi-objective scheduling and multi-scenario adaptability problems. The new power system scheduling needs to comprehensively consider multiple objectives, including maximizing new energy consumption, power supply-demand balance, system economy, and operation safety. However, the existing scheduling methods lack the ability to deduce multiple scenarios and are difficult to comprehensively evaluate future possible operation risks and constraint conditions, resulting in poor adaptability of the scheduling scheme in actual operation.

[0004] Therefore, how to improve the adaptability and accuracy of power scheduling decisions has become an urgent problem to be solved currently. Summary of the Invention

[0005] The purpose of the present invention is to overcome the shortcomings of insufficient foresight and poor adaptability of the existing scheduling scheme in the above-mentioned prior art, and provide a high-proportion new energy power system scheduling method and related devices.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] In the first aspect of the present invention, a dispatching method for a high-proportion new energy power system is provided, including: based on new energy output prediction data, load prediction data, and grid state data, determining a forward-looking time window including high-risk time points in the power system within a preset future time period through supply-demand balance risk analysis; constructing several power system operation scenarios with the forward-looking time window as the time domain, and obtaining several high-risk operation scenarios by simulating the operation states of the power system under each power system operation scenario; constructing a real-time dispatching optimization model of the power system considering the cost of supply-demand imbalance, reserve capacity cost, and operation risk cost, and solving the real-time dispatching optimization model of the power system with several high-risk operation scenarios as the dispatching basis to obtain a power system dispatching plan.

[0008] Optionally, the determining a forward-looking time window including high-risk time points in the power system within a preset future time period through supply-demand balance risk analysis based on new energy output prediction data and load prediction data includes: obtaining the supply-demand deviation of each time point in the power system within a preset future time period according to the new energy output prediction data, load prediction data, and grid state data; obtaining the supply-demand imbalance risk value of each time point in the power system within a preset future time period through a supply-demand imbalance risk assessment function according to the supply-demand deviation of each time point in the power system within a preset future time period; wherein, the supply-demand imbalance risk assessment function is:

[0009]

[0010] wherein, R(t) is the supply-demand imbalance risk value of the power system at time point t, |ΔP(t)| is the absolute value of the supply-demand deviation of the power system at time point t, is the standard deviation of the new energy output prediction data, is the standard deviation of the load prediction data.

[0011] Taking the time points with the supply-demand imbalance risk value greater than the preset supply-demand imbalance risk threshold as the high-risk time points in the power system within a preset future time period; generating a forward-looking time window including high-risk time points in the power system within a preset future time period according to the high-risk time points in the power system within a preset future time period.

[0012] Optionally, before determining the forward-looking time window including high-risk time points in the power system within a preset future time period through supply-demand balance risk analysis, it further includes: removing the noise in the new energy output prediction data, load prediction data, and grid state data by using a low-pass filter or wavelet decomposition method.

[0013] Optionally, constructing several power system operation scenarios with the forward time window as the time domain includes: based on new energy output prediction data, load prediction data, and power grid status data, taking the forward time window as the time domain, and generating several groups of new energy output prediction data, load prediction data, and power grid status data at each time point within the forward time window through the Monte Carlo simulation method and the stochastic process theory; constructing power system operation scenarios based on several groups of new energy output prediction data, load prediction data, and power grid status data at each time point within the forward time window to obtain several power system operation scenarios.

[0014] Optionally, obtaining high-risk operation scenarios by simulating the operation status of the power system under each power system operation scenario includes: simulating the operation status of the power system under each power system operation scenario through a pre-trained long short-term memory network model or a pre-trained generative adversarial network model to obtain the power supply-demand imbalance value, node voltage over-limit value, and power flow congestion value of the power system under each power system operation scenario; taking the power system operation scenarios with the power supply-demand imbalance value greater than the power supply-demand imbalance risk threshold, the power system operation scenarios with the node voltage over-limit value greater than the node voltage over-limit risk threshold, and the power system operation scenarios with the power flow congestion value greater than the power flow congestion risk threshold as high-risk operation scenarios.

[0015] Optionally, the objective function of the power system real-time scheduling optimization model is:

[0016]

[0017] where C is the power system operation cost, T is the scheduling period, C 失调 (t) is the power supply-demand imbalance cost, C 备用 (t) is the reserve capacity cost, C 风险 (t) is the operation risk cost; the constraint conditions of the power system real-time scheduling optimization model include: power supply-demand balance constraint, equipment operation constraint, and safe operation constraint.

[0018] Optionally, solving the power system real-time scheduling optimization model with several high-risk operation scenarios as the scheduling basis to obtain the power system scheduling plan includes: setting the weights of each high-risk operation scenario according to the risk values of each high-risk operation scenario; generating several initial power system scheduling plans, obtaining the objective function values of each initial power system scheduling plan under each high-risk operation scenario and weighted superposition according to the weights of each high-risk operation scenario to obtain the weighted objective function values of each initial power system scheduling plan; taking the minimization of the weighted objective function value as the optimization objective and optimizing the initial power system scheduling plan through the genetic algorithm to obtain the final power system scheduling plan.

[0019] Optionally, it further includes: obtaining real-time new energy output prediction data, load prediction data, power grid status data, new energy output prediction error, load prediction error, and power system dispatching plan execution deviation at preset time intervals to obtain rolling optimization data, and rolling optimizing the power system dispatching plan according to the rolling optimization data.

[0020] The second invention of the present invention provides a high-proportion new energy power system dispatching system, including: a forward-looking time window determination module, configured to determine a forward-looking time window including high-risk time points in a future preset time period of the power system through supply-demand balance risk analysis based on new energy output prediction data, load prediction data, and power grid status data; a scenario processing module, configured to construct several power system operation scenarios with the forward-looking time window as the time domain, and obtain several high-risk operation scenarios by simulating the operation status of the power system under each power system operation scenario; a dispatching plan generation module, configured to construct a real-time dispatching optimization model of the power system considering the cost of supply-demand imbalance, reserve capacity cost, and operation risk cost, and solve the real-time dispatching optimization model of the power system with several high-risk operation scenarios as the dispatching basis to obtain the power system dispatching plan.

[0021] The third invention of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned high-proportion new energy power system dispatching method are implemented.

[0022] The fourth invention of the present invention provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned high-proportion new energy power system dispatching method are implemented.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] The dispatching method for a high-proportion new energy power system of the present invention first determines, based on new energy output prediction data, load prediction data, and power grid status data, a forward-looking time window containing high-risk time points within a preset future time period for the power system through supply-demand balance risk analysis, dynamically predicts the power supply-demand changes and system operation risks within an upcoming period of time, plans the dispatching time window in advance, and enhances the forward-looking and adaptability of dispatching decisions. Next, through intelligent deduction of multiple power system operation scenarios within the forward-looking time window, simulates the operation states of the power system under each power system operation scenario to obtain several high-risk operation scenarios, provides decision support for real-time dispatching, and improves the adaptability of the dispatching plan under complex power system operation scenarios. Finally, by constructing a real-time dispatching optimization model for the power system considering the cost of supply-demand imbalance, reserve capacity cost, and operation risk cost, realizes multi-objective path optimization, can effectively balance new energy output fluctuations, power supply-demand, and system economy, achieves maximized new energy consumption, and reduces the problems of wind curtailment and light curtailment. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a flowchart of the dispatching method for a high-proportion new energy power system according to an embodiment of the present invention.

[0026] Figure 2 It is a structural block diagram of the dispatching system for a high-proportion new energy power system according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] In order to enable those skilled in the art of the present technology to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0028] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or are inherent to these processes, methods, products, or devices.

[0029] The present invention will be further described in detail below with reference to the accompanying drawings:

[0030] Refer to Figure 1 , in an embodiment of the present invention, a high - proportion new - energy power system scheduling method is provided, specifically a deterministic power generation and consumption joint real - time scheduling decision - making method for a new - type power system, aiming to improve the real - time scheduling ability, adaptability and supply - demand balance level of the power system under the condition of high - proportion new - energy access.

[0031] Specifically, the high - proportion new - energy power system scheduling method of the present invention includes the following steps:

[0032] S1: Based on new - energy output prediction data, load prediction data and power - grid state data, determine a forward - looking time window containing high - risk time points in the power system within a future preset time period through supply - demand balance risk analysis.

[0033] S2: Construct several power - system operation scenarios with the forward - looking time window as the time domain, and obtain several high - risk operation scenarios by simulating the operation states of the power system under each power - system operation scenario.

[0034] S3: Construct a real - time scheduling optimization model for the power system considering the cost of supply - demand imbalance, reserve - capacity cost and operation - risk cost, and solve the real - time scheduling optimization model for the power system with several high - risk operation scenarios as the scheduling basis to obtain a power - system scheduling plan.

[0035] For the high - proportion new - energy power system scheduling method of the present invention, first, based on new - energy output prediction data, load prediction data and power - grid state data, determine a forward - looking time window containing high - risk time points in the power system within a future preset time period through supply - demand balance risk analysis, dynamically predict the power supply - demand changes and system operation risks in the next period of time, plan the scheduling time window in advance, and enhance the forward - looking and adaptability of scheduling decisions. Then, through intelligent deduction of multiple power - system operation scenarios within the forward - looking time window, simulate the operation states of the power system under each power - system operation scenario to obtain several high - risk operation scenarios, provide decision - making support for real - time scheduling, and improve the adaptability of the scheduling plan under complex power - system operation scenarios. Finally, by constructing a real - time scheduling optimization model for the power system considering the cost of supply - demand imbalance, reserve - capacity cost and operation - risk cost, realize multi - objective path optimization, effectively balance the new - energy output fluctuation, power supply - demand and system economy, maximize new - energy consumption, and reduce the problems of wind and light curtailment.

[0036] Explanatory. In order to improve the adaptability and accuracy of power dispatching decisions, technologies such as forward-looking time window generation, scenario rehearsal, and path optimization have gradually become the research focus. The forward-looking time window refers to predicting and planning the operating state of the power system within a certain period in the future (such as 4 - 6 hours) before real-time dispatching decisions, providing time range support. Scenario rehearsal is to simulate and deduce various possible future operating states based on historical data, real-time monitoring data, and prediction models. The path optimization technology is to optimize the real-time dispatching path based on the forward-looking time window and scenario rehearsal results to meet multiple objectives such as new energy consumption and power grid safe operation.

[0037] However, existing dispatching methods mainly rely on static prediction and deterministic models, unable to effectively generate a forward-looking time window and unable to plan the dispatching path in advance to cope with future operating risks. The current scenario rehearsal methods cannot effectively conduct intelligent rehearsal of multiple future scenarios, resulting in insufficient adaptability of the dispatching plan to multiple factors such as new energy output and load fluctuations. The current path optimization technology cannot fully match new energy output and user load, resulting in serious phenomena of wind curtailment and light curtailment, affecting the utilization rate of new energy.

[0038] Based on the above problems, the present invention determines a forward-looking time window including high-risk time points within a preset future time period for the power system through supply-demand balance risk analysis, improving the forward-looking nature of dispatching. By simulating the operating state of the power system under each power system operating scenario, several high-risk operating scenarios are obtained, improving the adaptability of the dispatching plan. Furthermore, a real-time dispatching optimization model of the power system considering the cost of supply-demand imbalance, reserve capacity cost, and operating risk cost is constructed, and the real-time dispatching optimization model of the power system is solved based on several high-risk operating scenarios to achieve multi-objective collaborative path optimization, balancing new energy consumption, supply-demand balance, and operating safety. Finally, the high-proportion new energy power system dispatching method of the present invention can effectively improve the dispatching decision-making ability and operating stability of the power system under the condition of high-proportion new energy access. Among them, a high-proportion new energy power system generally refers to a power system with a new energy electricity proportion of 20%, but not limited thereto.

[0039] In a possible implementation manner, the forward-looking time window including high-risk time points of the power system within a preset future time period determined through supply-demand balance risk analysis based on new energy output prediction data and load prediction data includes: obtaining the supply-demand deviation of each time point of the power system within the preset future time period according to the new energy output prediction data, the load prediction data, and the power grid state data; obtaining the supply-demand imbalance risk value of each time point of the power system within the preset future time period through a supply-demand imbalance risk assessment function according to the supply-demand deviation of each time point of the power system within the preset future time period, and taking the time points with the supply-demand imbalance risk value greater than the preset supply-demand imbalance risk threshold as the high-risk time points of the power system within the preset future time period; and generating a forward-looking time window including high-risk time points of the power system within the preset future time period according to the high-risk time points of the power system within the preset future time period.

[0040] Among them, the supply-demand imbalance risk assessment function is:

[0041]

[0042] Among them, R(t) is the supply-demand imbalance risk value of the power system at time point t, and |ΔP(t)| is the absolute value of the supply-demand deviation of the power system at time point t. is the standard deviation of the new energy output prediction data. is the standard deviation of the load prediction data.

[0043] Explanatorily, the new energy output prediction data includes wind power output, photovoltaic output, and energy storage charge and discharge status, etc. Among them, wind power output: Based on real-time meteorological data (such as wind speed, wind direction, air pressure, etc.) and historical output data, predict the wind power output within a future time period. Photovoltaic output: According to real-time weather conditions (such as irradiance, temperature, etc.) and historical data, predict the power generation capacity of photovoltaic modules. Energy storage charge and discharge status: The charge or discharge status of energy storage devices, which is predicted based on battery power, load demand, and power grid state.

[0044] The load prediction data includes total load and sub-region load prediction, etc. Among them, total load: Predict the load demand for a future time period (usually 4 to 6 hours), considering the influence of factors such as season, time period, and climate. Sub-region load prediction: Predict the load demand of different regions according to the electricity consumption conditions of different regions. Especially in a multi-region power grid, the load prediction of key nodes is crucial.

[0045] Grid status data includes the output status of generating units, grid node voltages, frequencies, power flow status, and energy storage device status, etc. Among them, the output status of generating units: the output prediction of each generating unit in the future time period, taking into account the operating status of the equipment, maintenance plans, and fault warnings. Grid node voltages, frequencies, power flow status: real-time monitoring of grid operation parameters to predict the fluctuation trends of voltages and frequencies. Energy storage device status: information such as the charge-discharge strategy, current power level, and maximum capacity of the energy storage system.

[0046] The future preset time period can be 4 to 6 hours in the future. For the acquisition of various types of data, based on the new energy output prediction model, predict the new energy power output in the next 4 to 6 hours to obtain new energy output prediction data; use the load curve prediction algorithm (such as the LSTM model) to predict the change in load demand in the next 4 to 6 hours to obtain load prediction data; grid status data is obtained through real-time monitoring.

[0047] The calculation of the supply-demand deviation can be obtained by the following formula:

[0048] ΔP(t) = P RES (t) + P gen (t) - L(t)

[0049] Among them, ΔP(t) is the supply-demand deviation at time point t, P RES (t) is the new energy output prediction data at time point t, P gen (t) is the output of conventional power sources at time point t, and L(t) is the load prediction data at time point t.

[0050] The supply-demand imbalance risk assessment function represents the risk level of supply-demand imbalance at time point t, which is defined by combining the absolute value of the supply-demand deviation and system uncertainty. |ΔP(t)| is used to reflect the degree of supply-demand imbalance. The larger the value of R(t), the higher the risk of supply-demand imbalance under the condition of higher uncertainty. The supply-demand imbalance risk threshold is set according to the actual operation experience of the system. By comparison, the time points with the supply-demand imbalance risk value greater than the preset supply-demand imbalance risk threshold are used as the high-risk time points of the power system in the future preset time period.

[0051] When generating a forward-looking time window of the power system in the future preset time period that includes the high-risk time points based on the high-risk time points of the power system in the future preset time period, adaptively set a time period including all high-risk time points as the forward-looking time window, which can start from the scheduling start time point to the latest high-risk time point, or a certain time can be reserved after the latest high-risk time point.

[0052] In a possible implementation, before the look-ahead time window in which the power system includes high-risk time points within a preset future time period is determined through supply-demand balance risk analysis, the method further includes: removing noise in the new energy output prediction data, load prediction data, and grid state data by using a low-pass filter or wavelet decomposition method.

[0053] Explanatory, the random noise components in the data are removed by using a low-pass filter or wavelet decomposition method, and the true supply-demand deviation signal is retained. Among them, the low-pass filter: is suitable for removing high-frequency noise and retaining the low-frequency trend of the supply-demand deviation, and is commonly used in signal processing. The wavelet decomposition method: performs multi-scale analysis on the supply-demand deviation, decomposes it into low-frequency and high-frequency parts to remove high-frequency noise. The new energy output prediction data, load prediction data, and grid state data after removing the rapidly fluctuating noise part are smoother and can more accurately reflect the true supply-demand imbalance signal, providing reliable inputs for subsequent risk assessment and screening of high-risk time points.

[0054] In a possible implementation, constructing several power system operation scenarios with the look-ahead time window as the time domain includes: based on the new energy output prediction data, load prediction data, and grid state data, taking the look-ahead time window as the time domain, generating several sets of new energy output prediction data, load prediction data, and grid state data at each time point within the look-ahead time window through the Monte Carlo simulation method and stochastic process theory; constructing power system operation scenarios based on several sets of new energy output prediction data, load prediction data, and grid state data at each time point within the look-ahead time window to obtain several power system operation scenarios.

[0055] Explanatory, the goal of scenario simulation is to simulate various possible supply-demand scenarios through intelligent deduction based on the future operating state within the look-ahead time window, extract key constraints and high-risk scenarios, and provide accurate and dynamic decision-making basis for real-time scheduling path optimization. Scenario simulation is based on data-driven methods such as machine learning and deep learning, and performs intelligent deduction on the new energy output prediction data, load prediction data, and grid state data within the look-ahead time window to identify possible system operating states and key constraint conditions, and predict high-risk operating scenarios.

[0056] Generate several sets of new energy output prediction data, load prediction data, and grid state data at each time point within the look-ahead time window through the Monte Carlo simulation method and stochastic process theory, and construct power system operation scenarios based on this to obtain several power system operation scenarios.

[0057] In a possible implementation, obtaining the high-risk operation scenarios by simulating the operation states of the power system under various power system operation scenarios includes: simulating the operation states of the power system under various power system operation scenarios through a pre-trained long short-term memory network model or a pre-trained generative adversarial network model to obtain the power supply-demand imbalance value, the node voltage over-limit value, and the power flow congestion value of the power system under each power system operation scenario; and taking the power system operation scenarios with the power supply-demand imbalance value greater than the power supply-demand imbalance risk threshold, the power system operation scenarios with the node voltage over-limit value greater than the node voltage over-limit risk threshold, and the power system operation scenarios with the power flow congestion value greater than the power flow congestion risk threshold as the high-risk operation scenarios.

[0058] Explanatory, through a pre-trained long short-term memory network model or a pre-trained generative adversarial network model, the intelligent deduction of the operation states of the power system under various power system operation scenarios is realized, and the deduction fidelity is improved.

[0059] In a possible implementation, the objective function of the power system real-time scheduling optimization model is:

[0060]

[0061] where C is the power system operation cost, T is the scheduling period, and C 失调 (t) is the power supply-demand imbalance cost, C 备用 (t) is the reserve capacity cost, and C 风险 (t) is the operation risk cost.

[0062] The constraint conditions of the power system real-time scheduling optimization model include: power supply-demand balance constraint, equipment operation constraint, and safe operation constraint.

[0063] Explanatory, the power system real-time scheduling optimization model fully considers power supply-demand balance and new energy consumption, and starts from multi-objective optimization, including power supply-demand imbalance cost, reserve capacity cost, and operation risk cost.

[0064] Specifically, the power supply-demand imbalance cost is mainly determined by the magnitude of the power supply-demand deviation and the duration of the deviation:

[0065] C 失调 (t) = α 失调 ·|ΔP(t)|

[0066] where ΔP(t): power supply-demand deviation, representing the difference between power supply and demand at time t, and the calculation formula is ΔP(t) = P RES (t) + P gen (t) - L(t). α 失调: Weight factor for the cost of supply-demand imbalance, used to adjust the economic loss or impact degree of the imbalance. Calculation basis: Supply-demand deviation ΔP(t): directly reflects the severity of the supply-demand imbalance. Weight factor α 失调 : This factor can be determined according to the economic loss assessment model of the power grid or actual operation experience. A larger α 失调 indicates a greater impact of the imbalance on the economy or system stability.

[0067] The cost of reserve capacity is determined as follows: The volatility of new energy output affects the stable operation of the power grid. Conventional power sources serve as reserve capacity to ensure the power grid provides backup power during low output periods of new energy. Its cost formula:

[0068] C 备用 (t) = β 备用 ·P 备用 (t)

[0069] Where, P 备用 (t): Output of reserve capacity, that is, the available backup power capacity at a certain moment, usually obtained through dispatching calculations to ensure rapid power supply replenishment during supply-demand imbalance. β 备用 : Unit cost coefficient of reserve capacity, used to reflect the economic cost of reserve capacity, such as start-up cost, operation cost, etc. P 备用 (t): Considering potential supply-demand imbalance, calculated based on the actual load fluctuations of the power grid and system reliability requirements. β 备用 : Determined according to the power grid dispatching strategy, type of reserve units, energy cost, etc. Since starting reserve units requires additional costs, the cost of reserve capacity is higher than the operation cost of conventional generating units.

[0070] The cost of operation risk is calculated and determined by the operation risk value and weight factor:

[0071] C 风险 (t) = γ 风险 ·R 风险 (t)

[0072] Where, R 风险 (t): Represents the risk degree of the power grid at time t, usually obtained through simulation calculations or risk assessment models, involving factors such as voltage over-limit, power flow congestion, and frequency fluctuation. Usually evaluated based on the safety indicators of the power grid, such as over-limit conditions of voltage, frequency, power flow, etc. Can be calculated through simulation models or statistical methods based on historical data. γ 风险 : Reflects the economic cost of risk. A higher γ 风险 indicates a higher cost of operation risk, usually set according to the reliability requirements of the power grid and the operating status of equipment.

[0073] Regarding the constraint conditions of the real-time scheduling optimization model for the power system, they can be formed by extracting the key constraint conditions of high-risk operation scenarios, including the supply-demand balance constraint: used to ensure the supply-demand balance; the equipment operation constraint: used to ensure that the equipment operates within the safe range; and the safe operation constraint: used to ensure that indicators such as the grid voltage, frequency, and power flow do not exceed the safe range.

[0074] In a possible implementation manner, solving the real-time scheduling optimization model for the power system based on several high-risk operation scenarios to obtain the power system scheduling plan includes: setting the weights of each high-risk operation scenario according to the risk values of each high-risk operation scenario; generating several initial power system scheduling plans, obtaining the objective function values of each initial power system scheduling plan under each high-risk operation scenario, and weighted superposition according to the weights of each high-risk operation scenario to obtain the weighted objective function values of each initial power system scheduling plan; taking the minimization of the weighted objective function value as the optimization objective, and optimizing the initial power system scheduling plan through the genetic algorithm to obtain the final power system scheduling plan.

[0075] Explanatorily, the risk values that each high-risk operation scenario may bring (such as the cost of supply-demand imbalance and the risk of equipment overload) may be different. Therefore, it is necessary to perform weighted processing on each high-risk operation scenario according to the risk values. The weighted high-risk operation scenarios can reflect the contribution of different scenarios to the optimization of the scheduling plan. More important high-risk operation scenarios will be given higher weights during the optimization process. The scheduling plan is iteratively optimized through heuristic methods such as the genetic algorithm, simulating the possible scheduling plans under multiple high-risk operation scenarios, and then selecting the scheduling plan with lower risk and lower cost. The final scheduling plan includes the generator output: the output scheduling of conventional power sources at each time point; the new energy output: the scheduling strategies of new energy devices such as wind power and photovoltaic power; the load scheduling: the strategies for adjusting according to the real-time load demand; and the reserve capacity scheduling: the usage strategies of reserve capacity to cope with sudden supply-demand imbalances or equipment failures.

[0076] In a possible implementation manner, the high-proportion new energy power system scheduling method further includes: obtaining real-time new energy output prediction data, load prediction data, grid status data, new energy output prediction errors, load prediction errors, and power system scheduling plan execution deviations at preset time intervals to obtain rolling optimization data, and rolling optimizing the power system scheduling plan according to the rolling optimization data.

[0077] Explanatory, real-time feedback and rolling optimization aim to achieve the dynamic adaptability and real-time update of the scheduling plan, ensure the real-time matching of the system operation state and the scheduling path, and thus effectively solve the uncertainty problem brought by the rapid change of the power grid operation state under the condition of high proportion of new energy access. The rolling optimization mechanism dynamically adjusts and rolls upates the scheduling plan based on real-time monitoring data, combined with the forward-looking time window and the results of multi-scenario deduction, forming a closed-loop scheduling control to ensure that the scheduling plan always maintains the optimal state during the execution process.

[0078] Mainly obtain real-time new energy output prediction data, load prediction data, power grid state data, new energy output prediction error, load prediction error, and the execution deviation of the power system scheduling plan. Among them, the new energy output prediction error and the load prediction error are used to evaluate the error of the scheduling plan, which is obtained by calculating the error between the actual execution and the operation scenario. The execution deviation of the power system scheduling plan refers to the deviation between the actually executed scheduling plan and the operation scenario, reflecting the difference between the actually executed scheduling plan and the issued scheduling plan. The error and the execution deviation are the basis for dynamically adjusting the scheduling plan to ensure the accuracy of the real-time scheduling plan.

[0079] Based on the real-time new energy output prediction data, load prediction data, power grid state data, new energy output prediction error, load prediction error, and the execution deviation of the power system scheduling plan, combined with the forward-looking time window and the newly obtained high-risk operation scenarios, an optimization algorithm (such as model interaction-driven optimization and dynamic constraint adjustment, etc.) is used to regenerate the power system scheduling plan, and the rolling optimization of the power system scheduling plan is realized by performing it once every preset time interval (for example, 15 minutes).

[0080] Finally, the optimized power system scheduling plan is issued to control the conventional power sources, new energy output, and load regulation, and the execution effect is monitored in real time to form a feedback closed loop to ensure that the system state is always in the optimal state.

[0081] In a possible implementation manner, for a certain provincial power grid with a new energy output ratio of 20% and including 2000 computing nodes, the high-proportion new energy power system scheduling method of the present invention is adopted, and rolling optimization is performed every 15 minutes. The final application effects are as follows: scheduling calculation time: 45 seconds; new energy consumption rate: 96.5%; reduction of the supply-demand balance deviation: 15%. It can be seen that the high-proportion new energy power system scheduling method of the present invention can effectively realize the efficient, stable, and economic operation of the power system under the condition of high-proportion new energy access.

[0082] The following is an apparatus embodiment of the present invention, which can be used to execute the method embodiment of the present invention. For the details not disclosed in the apparatus embodiment, please refer to the method embodiment of the present invention.

[0083] SeeFigure 2 , in another embodiment of the present invention, a high-proportion new energy power system scheduling system is provided, which can be used to implement the above-mentioned high-proportion new energy power system scheduling method. Specifically, the high-proportion new energy power system scheduling system includes a forward-looking time window determination module, a scenario processing module, and a scheduling plan generation module.

[0084] Among them, the forward-looking time window determination module is used to determine the forward-looking time window including high-risk time points in the future preset time period of the power system through supply-demand balance risk analysis based on new energy output prediction data, load prediction data, and power grid status data; the scenario processing module is used to construct several power system operation scenarios with the forward-looking time window as the time domain, and obtain several high-risk operation scenarios by simulating the operation status of the power system under each power system operation scenario; the scheduling plan generation module is used to construct a real-time scheduling optimization model of the power system considering the cost of supply-demand imbalance, reserve capacity cost, and operation risk cost, and solve the real-time scheduling optimization model of the power system with several high-risk operation scenarios as the scheduling basis to obtain the power system scheduling plan.

[0085] In a possible implementation manner, the determining the forward-looking time window including high-risk time points in the future preset time period of the power system through supply-demand balance risk analysis based on new energy output prediction data and load prediction data includes: obtaining the supply-demand deviation of each time point in the future preset time period of the power system according to the new energy output prediction data, load prediction data, and power grid status data; obtaining the supply-demand imbalance risk value of each time point in the future preset time period of the power system through the supply-demand imbalance risk assessment function according to the supply-demand deviation of each time point in the future preset time period of the power system.

[0086] Among them, the supply-demand imbalance risk assessment function is:

[0087]

[0088] Among them, R(t) is the supply-demand imbalance risk value of the power system at time point t, |ΔP(t)| is the absolute value of the supply-demand deviation of the power system at time point t, is the standard deviation of the new energy output prediction data, is the standard deviation of the load prediction data.

[0089] The time points with supply-demand imbalance risk values greater than the preset supply-demand imbalance risk threshold are used as the high-risk time points of the power system in the future preset time period; according to the high-risk time points of the power system in the future preset time period, a forward-looking time window including high-risk time points in the future preset time period of the power system is generated.

[0090] In a possible implementation, before determining the forward-looking time window including high-risk time points of the power system within a preset future time period through supply-demand balance risk analysis, it further includes: using a low-pass filter or wavelet decomposition method to remove the noise in the new energy output prediction data, load prediction data, and power grid state data.

[0091] In a possible implementation, constructing several power system operation scenarios with the forward-looking time window as the time domain includes: based on the new energy output prediction data, load prediction data, and power grid state data, with the forward-looking time window as the time domain, through the Monte Carlo simulation method and stochastic process theory, generating several groups of new energy output prediction data, load prediction data, and power grid state data at each time point within the forward-looking time window; constructing power system operation scenarios based on several groups of new energy output prediction data, load prediction data, and power grid state data at each time point within the forward-looking time window to obtain several power system operation scenarios.

[0092] In a possible implementation, obtaining high-risk operation scenarios by simulating the operation state of the power system under each power system operation scenario includes: through a pre-trained long short-term memory network model or a pre-trained generative adversarial network model, simulating the operation state of the power system under each power system operation scenario to obtain the supply-demand imbalance value, node voltage overlimit value, and power flow congestion value of the power system under each power system operation scenario; taking the power system operation scenarios with the supply-demand imbalance value greater than the supply-demand imbalance risk threshold, the power system operation scenarios with the node voltage overlimit value greater than the node voltage overlimit risk threshold, and the power system operation scenarios with the power flow congestion value greater than the power flow congestion risk threshold as high-risk operation scenarios.

[0093] In a possible implementation, the objective function of the power system real-time scheduling optimization model is:

[0094]

[0095] where C is the operation cost of the power system, T is the scheduling period, C 失调 (t) is the supply-demand imbalance cost, C 备用 (t) is the reserve capacity cost, C 风险 (t) is the operation risk cost.

[0096] The constraint conditions of the power system real-time scheduling optimization model include: supply-demand balance constraint, equipment operation constraint, and safe operation constraint.

[0097] In a possible implementation manner, solving the real-time scheduling optimization model of the power system based on several high-risk operation scenarios to obtain the power system scheduling scheme includes: setting the weights of each high-risk operation scenario according to the risk values of each high-risk operation scenario; generating several initial power system scheduling schemes, obtaining the objective function values of each initial power system scheduling scheme under each high-risk operation scenario, and weighted superposing according to the weights of each high-risk operation scenario to obtain the weighted objective function values of each initial power system scheduling scheme; taking minimizing the weighted objective function value as the optimization objective, and optimizing the initial power system scheduling scheme through a genetic algorithm to obtain the final power system scheduling scheme.

[0098] In a possible implementation manner, it further includes a rolling optimization module, which is used to obtain real-time new energy output prediction data, load prediction data, grid state data, new energy output prediction error, load prediction error, and power system scheduling scheme execution deviation at each preset time interval to obtain rolling optimization data, and roll and optimize the power system scheduling scheme according to the rolling optimization data.

[0099] All relevant contents of each step involved in the embodiments of the foregoing high-proportion new energy power system scheduling method can be cited in the function descriptions of the corresponding functional modules of the high-proportion new energy power system scheduling system in the embodiments of the present invention, and will not be elaborated here.

[0100] The division of modules in the embodiments of the present invention is illustrative, and is only a logical function division. In actual implementation, there may be other division methods. In addition, in each embodiment of the present invention, each functional module can be integrated in one processor, or can exist independently physically, or two or more modules can be integrated in one module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules.

[0101] In another embodiment of the present invention, a computer device is provided. The computer device includes a processor and a memory. The memory is used to store a computer program, and the computer program includes program instructions. The processor is used to execute the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, and is suitable for implementing one or more instructions. Specifically, it is suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method flow or corresponding function. The processor described in the embodiment of the present invention can be used for the operation of the high-proportion new energy power system scheduling method.

[0102] In another embodiment of the present invention, a storage medium is also provided, specifically a computer-readable storage medium (Memory). The computer-readable storage medium is a memory device in the computer device and is used to store programs and data. It can be understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and, of course, the extended storage medium supported by the computer device. The computer-readable storage medium provides a storage space, and the operating system of the terminal is stored in this storage space. And, one or more instructions suitable for being loaded and executed by the processor are also stored in this storage space. These instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The one or more instructions stored in the computer-readable storage medium can be loaded and executed by the processor to implement the corresponding steps of the high-proportion new energy power system scheduling method in the above embodiment.

[0103] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0104] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0105] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0106] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0107] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: still can modify the specific implementation manners of the present invention or make equivalent replacements, and any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for dispatching a high-proportion new energy power system, characterized in that: include: Based on the new energy output forecast data, load forecast data and grid status data, the forward-looking time window containing high-risk time points in the future preset time period of the power system is determined through supply and demand balance risk analysis; Several power system operation scenarios are constructed using the forward-looking time window as the time domain, and several high-risk operation scenarios are obtained by simulating the operation status of the power system under each power system operation scenario; A real-time dispatch optimization model for the power system is constructed which takes into account the cost of supply-demand imbalance, reserve capacity cost and operation risk cost. The real-time dispatch optimization model for the power system is solved based on several high-risk operation scenarios to obtain the power system dispatch plan.

2. The high-proportion new energy power system dispatching method according to claim 1 is characterized in that: The forward-looking time window containing high-risk time points in the future preset time period of the power system is determined through supply-demand balance risk analysis based on the new energy output forecast data and load forecast data, including: According to the new energy output forecast data, load forecast data and grid status data, the supply and demand deviation of the power system at each time point in the future preset time period is obtained; According to the supply and demand deviation of the power system at each time point in the future preset time period, the supply and demand imbalance risk value of the power system at each time point in the future preset time period is obtained through the supply and demand imbalance risk assessment function; Among them, the supply-demand imbalance risk assessment function is: Among them, R(t) is the supply and demand imbalance risk value of the power system at time point t, |ΔP(t)| is the absolute value of the supply and demand deviation of the power system at time point t, is the standard deviation of the new energy output forecast data, is the standard deviation of the load forecast data; The time point when the supply-demand imbalance risk value is greater than the preset supply-demand imbalance risk threshold is regarded as a high-risk time point of the power system in the future preset time period; According to the high-risk time points of the power system in the future preset time period, a forward-looking time window containing the high-risk time points in the future preset time period is generated.

3. The high-proportion new energy power system dispatching method according to claim 1 is characterized in that: The method of determining the forward-looking time window containing high-risk time points in the future preset time period of the power system through supply and demand balance risk analysis also includes: using a low-pass filter or wavelet decomposition method to eliminate noise in new energy output forecast data, load forecast data and power grid status data.

4. The high-proportion new energy power system dispatching method according to claim 1 is characterized in that: The construction of several power system operation scenarios as the time domain with the forward-looking time window includes: Based on the new energy output forecast data, load forecast data and power grid status data, with the forward-looking time window as the time domain, several groups of new energy output forecast data, load forecast data and power grid status data at each time point in the forward-looking time window are generated through Monte Carlo simulation method and random process theory; Based on several groups of new energy output forecast data, load forecast data and grid status data at each time point in the forward-looking time window, the power system operation scenario is constructed to obtain several power system operation scenarios.

5. The high-proportion new energy power system dispatching method according to claim 1 is characterized in that: The high-risk operation scenarios obtained by simulating the operation status of the power system under various power system operation scenarios include: By pre-training the long short-term memory network model or the pre-training generative adversarial network model, the operating state of the power system under various power system operating scenarios is simulated to obtain the supply and demand imbalance value, node voltage over-limit value and power flow congestion value of the power system under various power system operating scenarios; The power system operation scenarios in which the supply-demand imbalance value is greater than the supply-demand imbalance risk threshold, the power system operation scenarios in which the node voltage overlimit value is greater than the node voltage overlimit risk threshold, and the power system operation scenarios in which the power flow congestion value is greater than the power flow congestion risk threshold are regarded as high-risk operation scenarios.

6. The high-proportion new energy power system dispatching method according to claim 1 is characterized in that: The objective function of the power system real-time dispatch optimization model is: Among them, C is the power system operation cost, T is the dispatch period, C 失调 (t) is the cost of supply-demand imbalance, C 备用 (t) is the spare capacity cost, C 风险 (t) Cost of operational risk; The constraints of the power system real-time dispatch optimization model include: supply and demand balance constraints, equipment operation constraints and safe operation constraints.

7. The high-proportion new energy power system dispatching method according to claim 6 is characterized in that: The power system real-time dispatch optimization model is solved based on several high-risk operation scenarios as the dispatch basis, and the power system dispatch scheme is obtained, including: Set the weight of each high-risk operation scenario according to its risk value; Generate several initial power system dispatching plans, obtain the objective function values ​​of each initial power system dispatching plan under each high-risk operation scenario, and perform weighted superposition according to the weights of each high-risk operation scenario to obtain the weighted objective function value of each initial power system dispatching plan; Taking minimizing the weighted objective function value as the optimization goal, the initial power system dispatching plan is optimized by genetic algorithm to obtain the final power system dispatching plan.

8. The high-proportion new energy power system dispatching method according to claim 1 is characterized in that: Also includes: Real-time new energy output forecast data, load forecast data, grid status data, new energy output forecast error, load forecast error and power system dispatch plan execution deviation are obtained at every preset time interval to obtain rolling optimization data, and the power system dispatch plan is optimized based on the rolling optimization data.

9. A high proportion of new energy power system dispatching system, characterized in that: include: A forward-looking time window determination module is used to determine the forward-looking time window containing high-risk time points in the power system within a preset time period in the future through supply and demand balance risk analysis based on new energy output forecast data, load forecast data and power grid status data; The scenario processing module is used to construct several power system operation scenarios with the forward-looking time window as the time domain, and obtain several high-risk operation scenarios by simulating the operation status of the power system under each power system operation scenario; The dispatching scheme generation module is used to construct a real-time dispatching optimization model for the power system that takes into account the cost of supply-demand imbalance, reserve capacity cost and operation risk cost, and solves the real-time dispatching optimization model for the power system based on several high-risk operation scenarios to obtain a dispatching scheme for the power system.

10. 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, the steps of the high-proportion new energy power system scheduling method as described in any one of claims 1 to 8 are implemented.

11. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the high-proportion new energy power system dispatching method as claimed in any one of claims 1 to 8 are implemented.

Images