CN-121543836-B - Climate-driven hybrid energy storage comprehensive energy optimization method, system, equipment and storage medium

CN121543836BCN 121543836 BCN121543836 BCN 121543836BCN-121543836-B

Abstract

The invention discloses a climate-driven hybrid energy storage comprehensive energy optimization method, a system, equipment and a storage medium, which are applied to the technical field of comprehensive energy system planning and optimization and comprise the steps of generating an countermeasure network based on Wasserstein of fusion gradient punishment and small batch discrimination items, and generating an annual scene of wind and light time by time covering seasonal features; the method comprises the steps of taking a full-year wind-solar time-by-time scene as random input, constructing a capacity-scheduling combined optimization model of the hybrid energy storage comprehensive energy system, solving by adopting a Monte Carlo decomposition strategy to obtain optimal capacity configuration and a time-by-time scheduling strategy of each scene under the optimal capacity configuration, and carrying out energy storage duration integral evaluation on the time-by-time scheduling strategy to obtain utilization distribution of battery energy storage and hydrogen energy storage in the hybrid energy storage comprehensive energy system under different time scales. The invention obviously improves the robustness and the economy of the comprehensive energy system under the extreme climate condition.

Inventors

ZHAO YINGRU
TAN JIAWEI
XIE SHAN
JING RUI
LIN JIAN

Assignees

厦门大学

Dates

Publication Date: 20260512
Application Date: 20260116

Claims (7)

1. A climate driven hybrid energy storage integrated energy optimization method, comprising: Step 1, generating an countermeasure network based on Wasserstein of fusion gradient penalty and small batch discrimination items, and generating an annual scene which covers seasonal features; Step 2, constructing a hybrid energy storage comprehensive energy system capacity-scheduling joint optimization model which aims at annual total cost and comprises electric power, hydrogen energy, heat energy coupling and equipment operation constraint by taking the annual wind-solar time-by-time scenes as random input, and solving the hybrid energy storage comprehensive energy system capacity-scheduling joint optimization model by adopting a Monte Carlo decomposition strategy to obtain optimal capacity configuration and a time-by-time scheduling strategy of each scene under the optimal capacity configuration; step 3, performing energy storage duration integral evaluation on the time-by-time scheduling strategy to obtain utilization distribution of battery energy storage and hydrogen energy storage in the hybrid energy storage comprehensive energy system under different time scales; in the step 1, wasserstein fusing gradient penalty and small batch discrimination items generates an countermeasure network, specifically: Using generators Criticizing device Is a countermeasure to the structure; introducing small batch discrimination on WGAN-GP architecture, wherein the generator Wind and light are respectively used as 2-channel input, and after 1D convolution, downsampling, GLU residual error and deconvolution upsampling, 2-channel annual sequence is output, the criticizer Splicing small-batch distance features for the three-layer full-connection network and the middle layer; The loss function is as follows: WGAN-basic goal of GP: Wherein, the 、 Basic loss values of the criticizer and the generator respectively; Representing a sample of real climate data, the distribution of which is noted as ; Representing false samples generated by a generator, the distribution of which is noted as ; Representing mathematical expectations; representing the scoring output of the criticizer to the input samples; The gradient penalty coefficient is used for adjusting the intensity of the gradient penalty term; for the random interpolation sampling point between the real sample and the generated sample, is defined as Wherein Is that Random numbers between the two; Representing the criticizer output versus the input Is a gradient of (2); Representation of A norm; Setting criticizer for small batch of discrimination items Is the middle layer of Is characterized in that Defining a similarity measure for a batch : Wherein, the And Respectively represent the first in the current batch And (b) Feature vectors of the samples in the middle layer of the criticizer; representing euclidean distances between features; adding generator and criticizer losses, forming: Wherein, the And Total loss of the generator and the criticizer after adding the small batch of discrimination items; Batch size when training for model; And The weight coefficients of the small batch items in the generator and the criticizer are used for adjusting the strength of the diversity constraint; is an exponential function, used to map distances into similarity weights.
2. The climate driven hybrid energy storage integrated energy optimization method according to claim 1, wherein in step 2, the capacity-scheduling joint optimization model of the hybrid energy storage integrated energy system is specifically: decision variables and parameters: Main capacity decision-making photovoltaic installed capacity Fan installed capacity Capacity of electrolyzer Capacity of fuel cell Capacity of battery Hydrogen energy storage equivalent energy ; Time-by-time operation variable in climate scene: 、、、、、 Wherein, the method comprises the steps of, Generating power for photovoltaic time by time; generating power for wind power time by time; The power consumption of the electrolytic cell is realized; Generating power for the fuel cell; charging power for the battery energy storage; Storing energy and discharging power for the battery; The time periods are numbered in order to be able to, ; The state of charge of the battery time by time; the equivalent state of charge is the hydrogen energy storage time-by-time; 、 The power purchase power and the power selling power of the power grid are respectively; Objective function: annual total cost minimization: Wherein, the Investment cost for system annual; The annual operation and maintenance cost of the equipment is saved; The annual net electricity purchasing cost; the device is a device set, and comprises a photovoltaic device, a fan, an electrolytic tank, a fuel cell, a battery and a hydrogen energy storage container; is a device Is a function of the installed capacity of the device; is a device Unit volume investment unit price of (2); is a device Capital recovery coefficients of (a); Is the discount rate; is a device Is used for the service life of the (a); is a device At the position of Operating power of the time period; is a device Unit power operation cost unit price; Is that Time-of-purchase electricity price in time period; Is that The online electricity price of the time period; electricity, hydrogen balance and equipment constraints: electric power balance: Wherein, the Is that System base electrical load demand for a period of time; hydrogen energy balance, in terms of equivalent energy: Wherein, the And Respectively is Charging and discharging power of hydrogen energy storage in a period of time; self-loss coefficient for hydrogen storage; And The efficiency of the hydrogen storage and release processes, respectively; And The maximum rated power upper limit of hydrogen storage and hydrogen release respectively; And Is that The binary variables respectively represent hydrogen storage and hydrogen release states and are used for ensuring mutual exclusion of the two; cell and fuel cell restraint: Wherein, the Hydrogen production efficiency for the electrolyzer; generating efficiency for the fuel cell; Is the lower heating value of hydrogen; Lithium battery energy storage system constraints: Wherein, the The self-discharge rate coefficient of the battery; And The charge and discharge efficiencies of the battery are respectively; And Maximum charge and discharge power limits of the battery respectively; And Is that The binary variable is used for guaranteeing mutual exclusion of charging and discharging; fan output constraint: Wherein, the Is that Real-time wind speed at time intervals; is the cut-in wind speed; to cut out wind speed; Is the rated wind speed; Rated power of the fan; The comprehensive power generation efficiency of the fan is achieved; Photovoltaic output constraint: Wherein, the The illumination intensity under standard test conditions; Is that Actual solar radiation intensity for a period of time; the comprehensive efficiency of the photovoltaic system is achieved; Mutual exclusion constraint of power grid interconnection: Wherein: And The power purchase power and the power selling power of the power grid are respectively; And The maximum allowable electricity selling and electricity purchasing power of the grid-connected point are respectively; And Is that And the binary variable ensures that electricity selling and electricity purchasing actions cannot occur simultaneously.
3. The climate-driven hybrid energy storage comprehensive energy optimization method according to claim 1, wherein in step 2, a monte carlo decomposition strategy is adopted to solve a capacity-scheduling joint optimization model of the hybrid energy storage comprehensive energy system, so as to obtain an optimal capacity configuration and a time-by-time scheduling strategy of each scene under the optimal capacity configuration, wherein the method specifically comprises the following steps: sampling the generated annual wind-solar time-by-time scene for a plurality of times to form a plurality of annual time sub-models, independently solving, polymerizing to obtain capacity configuration distribution, and selecting a distribution center as the optimal capacity configuration based on solution set distribution; and fixing the optimal capacity configuration, and solving a time-by-time scheduling strategy of each scene under the optimal capacity configuration.
4. The climate driven hybrid energy storage comprehensive energy optimization method according to claim 1, wherein in step 3, energy storage duration integral evaluation is performed on the time-by-time scheduling strategy to obtain utilization distribution of battery energy storage and hydrogen energy storage in a hybrid energy storage comprehensive energy system under different time scales, specifically: Generating an energy storage time-by-time energy curve based on the time-by-time scheduling strategy; And carrying out vertical axis discrete-progressive scanning on the energy storage time-by-time energy curve, recording the product of each section of energy storage energy multiplied by the duration as an energy storage time duration integral index, and accumulating the whole year and the whole scene to obtain an energy storage time duration distribution map.
5. A climate driven hybrid energy storage integrated energy optimization system utilizing a climate driven hybrid energy storage integrated energy optimization method of any of claims 1-4, comprising: The annual wind-light time-by-time scene generation module is used for generating an countermeasure network based on Wasserstein of fusion gradient penalty and small batch discrimination items and generating an annual wind-light time-by-time scene covering seasonal features; The time-by-time scheduling strategy solving module is used for constructing a mixed energy storage comprehensive energy system capacity-scheduling combined optimization model which takes annual total cost as a target and comprises electric power, hydrogen energy, heat energy coupling and equipment operation constraint by taking the annual wind-solar time-by-time scene as random input, and solving the mixed energy storage comprehensive energy system capacity-scheduling combined optimization model by adopting a Monte Carlo decomposition strategy to obtain optimal capacity configuration and a time-by-time scheduling strategy of each scene under the optimal capacity configuration; And the energy storage duration integral evaluation module is used for carrying out energy storage duration integral evaluation on the time-by-time scheduling strategy to obtain the utilization distribution of battery energy storage and hydrogen energy storage in the hybrid energy storage comprehensive energy system under different time scales.
6. An electronic device, comprising: A memory for storing a computer program; A processor for implementing the steps of a climate driven hybrid energy storage integrated energy optimization method according to any of claims 1-4 when executing said computer program.
7. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps of a climate driven hybrid energy storage comprehensive energy optimization method according to any of claims 1-4.

Description

Climate-driven hybrid energy storage comprehensive energy optimization method, system, equipment and storage medium Technical Field The invention relates to the technical field of comprehensive energy system planning and optimization, in particular to a climate-driven hybrid energy storage comprehensive energy optimization method, system, equipment and storage medium. Background In recent years, renewable energy has rapidly increased in specific gravity in energy systems. Clean energy sources such as solar energy and wind energy have remarkable intermittence and fluctuation, and challenges are brought to safe and economic operation of the comprehensive energy system. The comprehensive energy system is used as a novel energy utilization mode, and can realize interconnection, intercommunication and coordination optimization of various energy forms such as electricity, heat, hydrogen and the like. By integrating heterogeneous energy units such as wind energy, photovoltaic power generation, cogeneration, energy storage and the like, the comprehensive energy system can improve the energy utilization efficiency and reduce the running cost. Meanwhile, the coupling operation of the multi-energy units increases the complexity of the model, and higher requirements are put on the optimal scheduling of the system. In integrated energy systems, not only are the loads significantly fluctuating and random, but the renewable energy power generation side is also subject to climatic conditions with strong uncertainty. This double uncertainty exposes the system to supply and demand imbalance risks at different time scales. For example, load fluctuations and changes in light intensity during the day may lead to short-term power gaps, while seasonal wind energy output changes may lead to long-term energy imbalances. To cope with such uncertainty, typical day methods, deterministic optimization, or scene generation methods are commonly employed in academia and engineering practice. However, the typical day method is difficult to cover the weather characteristics diversified throughout the year, and the deterministic method may ignore the extreme situation caused by fluctuation, so that the system planning result deviates from the actual requirement. The scene generation method improves the adaptability of the model to uncertainty to a certain extent, but still has the problems of unstable generated samples and overlarge scene quantity, which cause excessive calculation load. Energy storage technology is considered as an important means for relieving renewable energy source fluctuation and improving system flexibility. The battery energy storage system has the characteristics of quick response and high efficiency, is suitable for energy balance, frequency modulation and peak shaving at the level of hours, has large-scale and long-period energy storage potential, and can realize energy transfer at the scale of week, month and even season. However, how to reasonably configure and schedule short-term energy storage and long-term energy storage to make the advantages of the short-term energy storage and the long-term energy storage complementary is a key problem faced in current comprehensive energy system planning. In addition, as the scale of the comprehensive energy system expands, the climate data and the operation data volume increase sharply, the dimension and complexity of the optimization model are continuously improved. When the traditional mixed integer programming method is used for processing high-dimensional annual time sequence and large-scale scenes, the calculation cost is obviously increased, and the solving speed and stability are difficult to meet the requirements of practical engineering application. Therefore, how to provide a climate-driven hybrid energy storage comprehensive energy optimization method, system, device and storage medium, so as to improve model calculation efficiency and rationality of energy storage system configuration while guaranteeing authenticity and diversity of climate scenes is a problem to be solved by those skilled in the art. Disclosure of Invention In view of this, the present invention provides a climate driven hybrid energy storage integrated energy optimization method, system, device and storage medium. The method is characterized in that a climate-driven scene generation technology is introduced, wind and light resource time sequence data with authenticity and diversity can be generated, the problem of capacity allocation and scheduling time by time all year is solved by adopting a Monte Carlo decomposition optimization method on the basis, the difficulty that the traditional optimization model is excessively heavy in calculation load in a high-dimensional scene is solved, and meanwhile, an energy storage duration integral index is provided for quantifying the synergistic effect of short-time battery energy storage and long-time hydrogen energy storage, so that the reliability an