CN-122022636-A - Intelligent scheduling method and system for wind-light-hydrogen storage integrated logistics park

CN122022636ACN 122022636 ACN122022636 ACN 122022636ACN-122022636-A

Abstract

The invention discloses an intelligent scheduling method and system for a wind, light and hydrogen storage integrated logistics park, which relate to the technical field of computers, and are used for constructing a data collaborative system, executing multidimensional data alignment, calculating a hydrogen production quantity prediction function by utilizing an inference model, calculating a dynamic hydrogenation admission threshold value based on pressure feedback and logistics demand change, predicting a hydrogen production trend according to the prediction function, predicting whether the power generation of a power system is in a peak period or a trough period, limiting hydrogenation by combining the threshold value in the trough period, guiding vehicle energy supplementing in the peak period, aiming at coping with hydrogen production demand mismatch caused by wind and light resource fluctuation, changing post management into prospective regulation to reduce potential safety hazards, guaranteeing energy supply and optimizing an energy storage strategy.

Inventors

DU SONGLIN
ZHANG LIYANG
LEI TAO
Sadamu Shadik
WANG BINGQUAN
WANG XIAOFENG
YU JIONG
DU XUSHENG

Assignees

杭州骋风而来数字科技有限公司
新疆丝路融创网络科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260410

Claims (10)

1. The intelligent scheduling method for the wind-solar-hydrogen storage integrated logistics park is characterized by comprising the following steps of: acquiring operation data of a logistics park, wherein the operation data comprise meteorological data, hydrogen production electrolytic tank operation state data, hydrogen storage tank pressure data and residual hydrogen energy state of a hydrogen energy logistics vehicle, sampling each item of operation data, and then aligning according to sampling time to construct a basic data set comprising a time stamp; Based on the wind speed and irradiation intensity data in the basic data set, combining a power characteristic curve of a wind generating set and an energy conversion model of a photovoltaic module, calculating to obtain a predicted power generation sequence, and constructing a prediction function of hydrogen production quantity relative to time according to an electric hydrogen conversion efficiency coefficient of an electrolysis water system; The method comprises the steps of monitoring the current pressure value of a hydrogen storage tank in a logistics park in real time, and calculating to obtain dynamic hydrogenation admission thresholds in different time periods through the residual pressure of the hydrogen storage tank and logistics transportation requirements by combining the residual hydrogen energy distribution condition of on-line vehicles in the logistics park; Detecting the variation amplitude of a prediction function, judging whether the prediction function is in a trough region, when the hydrogen production amount obtained by the prediction function is in the trough region, calling the dynamic hydrogenation admission threshold value, comparing the residual hydrogen energy proportion of the current hydrogenation-requesting vehicle with the dynamic hydrogenation admission threshold value in real time, and if the residual hydrogen energy proportion of the hydrogenation-requesting vehicle is greater than the dynamic hydrogenation admission threshold value, giving a hydrogenation limiting instruction to the hydrogenation-requesting vehicle; and when the hydrogen yield obtained by the prediction function is in a peak area, sending a hydrogenation operation prompt instruction to a vehicle in which the residual hydrogen energy in the park is not in a full-load state on the basis of accounting the reserved safety margin of the hydrogen storage tank.
2. The intelligent scheduling method for the wind, light and hydrogen storage integrated logistics park according to claim 1, wherein the method for constructing the prediction function of the hydrogen production quantity relative to time comprises the following steps: learning the historical record data of wind speed and irradiation through a long-short-period memory neural network, setting a weather sequence with fixed duration and the environmental temperature at the current moment as input feature vectors, and outputting a weather prediction sequence in a preset time period in the future; Inputting the weather prediction sequence into a power conversion model extracted from a historical power generation record, and calculating algebraic sum of wind power generation power and photovoltaic power generation power; And for any predicted time, acquiring the power generated at the any predicted time, and calculating the power generated, the electrolytic hydrogen conversion rate and the electric energy consumption coefficient required by producing hydrogen in unit mass to obtain the hydrogen production rate at the any predicted time.
3. The intelligent scheduling method for the wind, light and hydrogen storage integrated logistics park according to claim 1, wherein the method for calculating the dynamic hydrogenation admission threshold comprises the following steps: Setting a rated maximum pressure value of a hydrogen storage tank, acquiring a current hydrogen storage tank pressure value, counting the residual hydrogen energy distribution conditions of all online vehicles in a park, and calculating to obtain an average residual hydrogen energy proportion; subtracting a quotient of dividing the current hydrogen storage tank pressure by the rated maximum pressure from a preset first weight constant, multiplying the obtained difference by the average residual hydrogen energy proportion, multiplying the difference by a logistics order strength correction factor, and finally adding the product result to a system protection factor; the logistics order intensity correction factor comprises the steps of calculating a difference value between the daily logistics traffic volume and the historical average logistics traffic volume, and adding a second weight constant to the weighted difference value to obtain the logistics order intensity correction factor.
4. The intelligent scheduling method for the wind, light and hydrogen storage integrated logistics park according to claim 1, wherein the scheduling strategy for the prediction function in the trough stage comprises the following steps: The method comprises the steps of calculating the ratio of the real-time value of the hydrogen production rate to the average hydrogen production rate in a preset sliding window in a time sequence of the duration of the window, setting a first preset proportion, and judging that a system enters a trough period if the ratio is lower than the first preset proportion and the first derivative of the hydrogen production rate along with the time function is negative; In this state, when the hydrogen energy logistics vehicle is identified to drive into the sensing area of the hydrogenation station, the residual hydrogen energy proportion of the hydrogen energy logistics vehicle is obtained, if the residual hydrogen energy proportion is larger than the current dynamic hydrogenation admission threshold value, the vehicle is judged not to belong to urgent hydrogenation demands, and a suggested hydrogenation time point is given according to the hydrogen yield prediction function.
5. The intelligent scheduling method for the wind, light and hydrogen storage integrated logistics park according to claim 1, wherein the scheduling strategy for the prediction function in the peak stage comprises the following steps: Acquiring an average value of hydrogen production rate in unit time as a hydrogen production rate threshold, and when the prediction function is larger than the hydrogen production rate threshold and the current hydrogen storage tank pressure reaches more than a third preset proportion of rated maximum pressure; And distributing an incoming route and a hydrogenation station for the invited vehicles according to the idle state of each hydrogenation station in the park and the geographic position of the hydrogen energy logistics vehicles by utilizing a path planning algorithm.
6. The intelligent scheduling system for the wind, light and hydrogen storage integrated logistics park is used for executing the intelligent scheduling method for the wind, light and hydrogen storage integrated logistics park according to any one of claims 1-5, and is characterized by comprising an operation data management module, a hydrogen production prediction module, an admission threshold calculation module, a trough scheduling module and a crest scheduling module; The operation data management module is used for acquiring multi-source heterogeneous operation data of the logistics park through the sensor network and the data acquisition interface, wherein the operation data comprise meteorological data, hydrogen production electrolytic tank operation state data, hydrogen storage tank pressure data and residual hydrogen energy state of a hydrogen energy logistics vehicle, and the operation data are sampled and time aligned to construct a basic data set containing a time dimension; the hydrogen production prediction module is used for calculating a predicted power generation sequence in a future preset period based on the basic data set, and establishing a prediction function of the hydrogen production rate relative to time by combining the electric hydrogen conversion efficiency coefficient of the electrolytic water system; the admission threshold calculation module is used for monitoring the current pressure value of the hydrogen storage tank in real time, counting the residual hydrogen energy distribution characteristics of the on-line vehicle, and calculating to obtain dynamic hydrogenation admission thresholds of different time periods by establishing a correlation model of the pressure of the hydrogen storage tank, the average residual hydrogen energy and the logistics transportation strength; The trough scheduling module is used for detecting the change rate of the prediction function in real time to identify the trough period of hydrogen production, comparing the residual hydrogen energy proportion of the vehicle requesting hydrogenation with the dynamic hydrogenation admission threshold value in real time in the trough period, and executing the limiting hydrogenation scheduling; And the peak scheduling module is used for searching the hydrogen energy logistics vehicles with residual hydrogen energy not reaching the full-load state when the prediction function indicates that the hydrogen production amount enters the peak period, and sending a hydrogenation operation prompt instruction and a station allocation instruction to the hydrogen energy logistics vehicles.
7. The intelligent scheduling system for the wind, light and hydrogen storage integrated logistics park of claim 6, wherein the operation data management module comprises a data acquisition unit, a data cleaning unit and a time alignment unit; The data acquisition unit is connected with the wind speed sensor, the irradiation sensor, the electrolytic cell sensor, the hydrogen storage tank pressure sensor and the vehicle-mounted terminal, the data cleaning unit is used for removing noise data and abnormal values, and the time alignment unit is used for unifying the data acquired by the data acquisition unit to a standard time axis.
8. The intelligent scheduling system for the wind, light and hydrogen storage integrated logistics park of claim 6, wherein the hydrogen production prediction module comprises a neural network training unit and a dynamics simulation unit; The neural network training unit is configured with a long-term memory neural network model and is used for outputting a meteorological prediction sequence, and the dynamics simulation unit is internally provided with a wind driven generator power characteristic model, a photovoltaic energy conversion model and an electric hydrogen conversion model and is used for generating the prediction function.
9. The intelligent scheduling system for the wind, light and hydrogen storage integrated logistics park according to claim 6, wherein the admission threshold calculation module comprises a state monitoring unit and a threshold calculation unit; the state monitoring unit collects the pressure of the hydrogen storage tank and the average energy level of the whole-garden vehicle in real time, and the threshold calculating unit calculates a dynamic hydrogenation admission threshold according to a preset weight constant and a logistics correction factor.
10. The intelligent scheduling system for the wind, light and hydrogen storage integrated logistics park of claim 6, wherein the wave trough scheduling module comprises a wave trough detection unit and a constraint control unit, and the wave crest scheduling module comprises a wave crest judging unit and an active guiding unit; The constraint control unit is used for executing vehicle admission judgment and issuing a limiting hydrogenation instruction; The active guiding unit integrates a path planning algorithm and is used for sending hydrogenation invitations to idle vehicles and distributing stations.

Description

Intelligent scheduling method and system for wind-light-hydrogen storage integrated logistics park Technical Field The invention relates to the technical field of computers, in particular to an intelligent scheduling method and system for a wind-solar-hydrogen storage integrated logistics park. Background By integrating wind energy, photovoltaic power generation, water electrolysis hydrogen production and energy storage technologies, the wind, light and hydrogen storage integrated logistics park can realize on-site absorption of renewable energy and green supply of hydrogen energy logistics vehicles. The wind, light and hydrogen storage integrated logistics park is characterized in that dynamic balance of energy production and consumption is achieved through an intelligent scheduling system. The system generally utilizes wind-solar complementary power generation to drive the electrolytic tank to generate hydrogen and stores the hydrogen in the hydrogen storage tank so as to meet the hydrogenation requirement of logistics vehicles. The prior art has strong randomness and volatility in processing wind and light resource sources, so that serious nonlinear mismatch exists between hydrogen production rate and hydrogen demand for vehicles. For example, the hydro-dispatch scheme is based on a static strategy of a vehicle arrival sequence, and potential safety hazards such as long-time queuing of vehicles and overload operation of compressors are easily caused in an overlapped period of low wind-light output and insufficient hydrogen storage pressure. Meanwhile, a prospective supply and demand prediction and dynamic threshold adjustment mechanism is lacking, so that a serious wind and light abandoning phenomenon is generated in the wind and light output peak period due to limited hydrogen storage space or misplacement of hydrogenation requirements, and the comprehensive energy utilization efficiency of the system is obviously reduced. Disclosure of Invention The invention aims to provide an intelligent scheduling method and system for a wind, light and hydrogen storage integrated logistics park, which are used for solving the problems in the prior art. The technical scheme of the invention is that the intelligent scheduling method for the wind-solar-hydrogen storage integrated logistics park is provided; The method comprises the following steps: Step 1, acquiring operation data of a logistics park through a sensor network and a data acquisition interface, wherein the operation data comprise meteorological data, hydrogen production cell operation state data, hydrogen storage tank pressure data and residual hydrogen energy state of a hydrogen energy logistics vehicle, performing high-frequency sampling on each operation data, and performing clock synchronization and alignment processing according to a uniform time stamp to construct a basic data set containing time dimension; Step 2, constructing a prediction function of hydrogen production quantity with respect to time, namely calculating a predicted power generation sequence in a future preset period by utilizing a deep learning model in combination with a physical constraint model based on wind speed and irradiation intensity data in the basic data set, and constructing a nonlinear prediction function of hydrogen production rate with respect to time by combining with an electric hydrogen conversion efficiency coefficient of an electrolytic water system; Step 3, calculating a dynamic hydrogenation admission threshold value, namely monitoring the current pressure value of a hydrogen storage tank in a logistics park in real time, counting the residual hydrogen energy distribution characteristics of all online vehicles in the logistics park, and calculating to obtain the dynamic hydrogenation admission threshold values in different time periods by establishing a correlation model of the pressure of the hydrogen storage tank, the average residual hydrogen energy and the logistics transportation strength; Step 4, executing a trough scheduling strategy, namely detecting the change rate of a prediction function in real time, identifying whether the prediction function enters a trough period of hydrogen production, calling the dynamic hydrogenation admission threshold when the prediction function is in the trough period, comparing the residual hydrogen energy proportion of the current vehicle requesting hydrogenation with the dynamic hydrogenation admission threshold in real time, and giving a limiting hydrogenation instruction or a delayed hydrogenation suggestion to the vehicle if the residual hydrogen energy proportion of the vehicle requesting hydrogenation is greater than the dynamic hydrogenation admission threshold; and 5, executing a peak scheduling strategy, namely actively searching hydrogen energy logistics vehicles in which residual hydrogen energy in the park does not reach a full-load state on the premise of ensuring the reserved safety margin of the hydrogen