A multi-objective real-time scheduling method based on prediction mechanism and energy storage mechanism

Abstract

The application discloses a multi-target real-time scheduling method based on a prediction mechanism and an energy storage mechanism. When multi-target real-time scheduling is performed, the application firstly analyzes historical data of multiple scheduling targets, and predicts and processes data in a future period of time to obtain predicted data. The energy storage mechanism is introduced to solve the problem of insufficient resources in the multi-target scheduling process. Finally, the fast and elitist non-dominated sorting genetic algorithm is used to achieve the purpose of real-time scheduling according to the predicted data and the energy storage mechanism.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a multi-objective real-time scheduling method based on prediction and energy storage mechanisms. Background Technology

[0002] Currently, rivers and waterways generally have comprehensive benefit objectives, including power generation, navigation, flood control, and ecological protection. These objectives are conflicting and mutually restrictive, with the contradiction being particularly prominent for hydropower stations that balance navigation and power generation benefits. Currently, inland waterway transportation mainly relies on manual scheduling, which is slow, of low quality, and lacks scientific rigor. Furthermore, real-time scheduling is not possible. This invention addresses the following key problems:

[0003] (1) Hydropower stations that combine power generation and navigation suffer from problems due to untimely acquisition of power load data. (2) The current weekly and daily navigation application plans for hydropower stations that combine power generation and navigation do not conform to the actual development of water transport. There is a large error between the advance declaration of ships and the actual arrival time of the hydropower station, resulting in low scheduling efficiency of the hydropower station or long waiting time for ships. (3) Due to the imbalance and insufficiency between the regulation of navigation and power generation benefits, the actual navigation time of hydropower stations that combine power generation and navigation is too short to meet the actual navigation demand.

[0004] The solution objectives and algorithms for multi-objective optimal scheduling of hydropower stations differ significantly from those for single-objective scheduling. The solution algorithms mainly fall into two categories: The first type uses traditional algorithms to solve the optimal scheduling model, transforming the multi-objective scheduling problem into a single-objective scheduling problem through methods such as weighting and penalty functions. The solution set obtained by this type of method is often not the optimal solution set, but rather generates a non-dominated solution set that satisfies the constraints. The second type uses modern intelligent algorithms or hybrid algorithms to solve the model. These methods are often computationally faster and can directly generate the optimal solution that satisfies the constraints. Currently, the following are some existing optimization models for hydropower stations.

[0005] A hydropower station optimal scheduling model with a daily calculation period is proposed. The model employs the principle of large-scale system decomposition and coordination and the Proof-of-Action (POA) algorithm to solve the model, optimizing the scheduling of hydropower stations on a daily basis. The scheduling process considers both scenarios—those without considering shipping—to investigate the degree of mutual influence between power generation and shipping.

[0006] An optimal scheduling model for hydropower stations with the goal of maximizing daily power generation is proposed. This model utilizes an immune differential evolution algorithm to analyze the data, establishing a refined model of the daily power generation of the hydropower station. The scheduling results are derived by using the power generation of the hydropower station as the optimal objective for scheduling analysis.

[0007] A hydropower station optimal scheduling model based on outflow. This model, grounded in the unsteady flow characteristics of the daily regulation process of a hydropower station, proposes a mathematical model based on the outflow process curve, considering navigation constraints. It utilizes the optimal square approximation method to establish a mathematical model for regulating the outflow process of the power station, and employs a pattern search method to solve the established model, obtaining a power station operation process that satisfies both navigation and optimal power generation.

[0008] However, while the daily-based hydropower station optimization scheduling model explores the relationship between power generation and navigation in actual operation and considers scheduling modes with and without navigation, its accuracy is insufficient. It cannot resolve the coordination between ship navigation and power generation in the short term. The hydropower station optimization scheduling model aiming to maximize daily power generation has a singular scheduling objective and inaccurate scheduling timeframe. The hydropower station optimization scheduling model based on downstream flow does not consider real-time ship arrivals and real-time grid load demand. Furthermore, none of the above three models explores real-time load demand and estimated ship arrival times, failing to meet the real-time scheduling requirements of hydropower stations.

[0009] Therefore, existing technologies still need to be improved and enhanced. Summary of the Invention

[0010] The technical problem to be solved by this invention is to provide a multi-objective real-time scheduling method based on prediction and energy storage mechanisms, addressing the aforementioned deficiencies of existing technologies. This method aims to solve the problem that existing power plant optimization scheduling models do not explore load demand and expected ship arrival times in real time, thus failing to meet the real-time scheduling requirements of hydropower plants.

[0011] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0012] In a first aspect, the present invention provides a multi-objective real-time scheduling method based on a prediction mechanism and an energy storage mechanism, characterized in that the method includes:

[0013] Obtain predicted power load demand data and estimated ship arrival times, and initialize the water level during the scheduling period based on the power load demand data and estimated ship arrival times to obtain an initial population, wherein the initial population is the water level during each scheduling period;

[0014] The initial population is subject to constraint correction to obtain the corrected population, and the shipping target and the power grid peak shaving target are obtained. Based on the shipping target and the power grid peak shaving target, the ship passage conditions are determined, and it is determined whether to use the energy storage device to generate electricity based on the ship passage conditions.

[0015] The modified population is sorted in a non-dominated hierarchical manner based on the optimal power generation efficiency and the optimal navigation efficiency. A genetic algorithm is used for selection, crossover and mutation operations to generate a new population. The generated new population is then sorted in a non-dominated hierarchical manner and classified into levels based on the optimal power generation efficiency and the optimal navigation efficiency.

[0016] The corrected population is selected according to the classification until the number of selected populations equals the number of initial populations. The termination condition is determined, and the result is output when the termination condition is met. The result is the water level of the hydropower station for each time period after scheduling.

[0017] In one implementation, obtaining the predicted power load demand data and the estimated arrival time of the ship includes:

[0018] Historical load data is acquired and analyzed, wherein the historical load data is the historical power load data of the hydropower station;

[0019] The power load demand data is obtained by using a prediction algorithm to predict the load over a future period.

[0020] In one implementation, obtaining the predicted power load demand data and the estimated arrival time of the ship includes:

[0021] Historical navigation data of ships in the waterway is acquired, and a multi-layer neural network model is built using deep learning algorithms. The historical navigation data includes the historical time of ships arriving at the hydropower station and passing through the lock in the past period.

[0022] Based on the multi-layer neural network model and the historical navigation data, the estimated arrival time of the ship at the hydropower station is predicted, thus obtaining the estimated arrival time of the ship.

[0023] In one implementation, initializing the water level during the scheduling period to obtain an initial population includes:

[0024] The water level of the hydropower station during each scheduling period is encoded according to time to obtain the initial population.

[0025] In one implementation, the step of performing a constraint correction on the initial population to obtain a corrected population includes:

[0026] Obtain preset constraints, wherein the constraints are: hydropower station unit constraints, downstream flow constraints, initial and final water level constraints of the hydropower station, and ship navigation conditions, wherein the hydropower station unit constraints include hydropower station unit start-up and shutdown time constraints and unit output constraints.

[0027] Based on the aforementioned constraints, the initial population is screened to obtain a revised population.

[0028] In one implementation, determining ship passage conditions based on the shipping target and the power grid peak-shaving target, and determining whether to utilize energy storage devices to generate electricity based on the ship passage conditions, includes:

[0029] Determine the ship passage conditions based on the shipping objectives and the power grid peak-shaving objectives;

[0030] Based on the aforementioned vessel passage conditions, determine whether the vessel is permitted to pass;

[0031] If ships can pass through, determine whether the power generation is sufficient;

[0032] If the power generation is insufficient, energy storage devices are used to generate electricity.

[0033] In one implementation, the ship passage conditions are that the ship needs to pass through a stable discharge flow and water level, cannot shut down the generator units, and needs to generate a stable amount of electricity.

[0034] Secondly, embodiments of the present invention also provide a multi-objective real-time scheduling device based on a prediction mechanism and an energy storage mechanism, wherein the device includes:

[0035] The prediction module is used to acquire predicted power load demand data and estimated ship arrival times, and initialize the water level during the scheduling period based on the power load demand data and estimated ship arrival times to obtain an initial population, wherein the initial population is the water level during each scheduling period.

[0036] The population correction module is used to perform a constraint correction on the initial population to obtain the corrected population, and to obtain the shipping target and the power grid peak shaving target. Based on the shipping target and the power grid peak shaving target, the module determines the ship passage conditions and determines whether to use the energy storage device to generate electricity based on the ship passage conditions.

[0037] The multi-objective scheduling module is used to perform non-dominated hierarchical sorting of the corrected population based on the optimal power generation benefit and the optimal navigation benefit, use genetic algorithm selection, crossover and mutation operations to generate a new population, and perform non-dominated hierarchical sorting and level division of the generated new population based on the optimal power generation benefit and the optimal navigation benefit.

[0038] The result output module is used to select the corrected population according to the level classification until the number of selected populations is equal to the number of initial populations, determine the termination condition, and output the result when the termination condition is met. The result is the water level of the hydropower station for each time period after scheduling.

[0039] Thirdly, embodiments of the present invention also provide a terminal device, wherein the terminal device includes a memory, a processor, and a multi-objective real-time scheduling program based on a prediction mechanism and an energy storage mechanism stored in the memory and executable on the processor. When the processor executes the multi-objective real-time scheduling program based on the prediction mechanism and the energy storage mechanism, it implements the steps of the multi-objective real-time scheduling method based on the prediction mechanism and the energy storage mechanism described in any of the above schemes.

[0040] Fourthly, embodiments of the present invention also provide a computer-readable storage medium, wherein the computer-readable storage medium stores a multi-objective real-time scheduling program based on a prediction mechanism and an energy storage mechanism, and when the multi-objective real-time scheduling program based on the prediction mechanism and the energy storage mechanism is executed by a processor, it implements the steps of the multi-objective real-time scheduling method based on the prediction mechanism and the energy storage mechanism as described in any of the above schemes.

[0041] Beneficial Effects: Compared with existing technologies, this invention provides a multi-objective real-time scheduling method based on prediction and energy storage mechanisms. First, it acquires predicted power load demand data and estimated ship arrival times. Based on these data, it initializes the water level for each scheduling period, obtaining an initial population, where the initial population represents the water level for each scheduling period. Then, it performs constrained corrections on the initial population to obtain a corrected population. It also acquires navigation and grid peak-shaving targets, determines ship passage conditions based on these targets, and decides whether to utilize energy storage devices for power generation. Next, it performs non-dominated hierarchical sorting and genetic algorithm selection, crossover, and mutation operations on the corrected population based on optimal power generation and navigation efficiency, generating a new population. Finally, it performs non-dominated hierarchical sorting and grading on the new population based on optimal power generation and navigation efficiency. Finally, the corrected population is selected according to the hierarchical classification until the number of selected populations equals the number of initial populations. A termination condition is then determined, and when the termination condition is met, the result is output, which represents the water level of the hydropower station for each time period after scheduling. This invention can obtain the scheduling target demand of the hydropower station in real time through power load forecasting and ship arrival time forecasting; through an energy storage mechanism, it can solve the problem of insufficient power generation by the hydropower station during ship navigation, thereby enabling the hydropower station to simultaneously meet the needs of power generation and navigation, extending the actual navigation time. Attached Figure Description

[0042] Figure 1 This is a flowchart illustrating a specific implementation of the multi-objective real-time scheduling method based on prediction and energy storage mechanisms provided in this invention.

[0043] Figure 2A flowchart illustrating the practical application of the multi-objective real-time scheduling method based on prediction and energy storage mechanisms provided in this embodiment of the invention.

[0044] Figure 3 The diagram illustrates the logic principle of a multi-objective real-time scheduling method based on prediction and energy storage mechanisms, as provided in this embodiment of the invention.

[0045] Figure 4 The schematic diagram of a multi-objective real-time regulation device based on prediction and energy storage mechanisms provided in this embodiment of the invention.

[0046] Figure 5 A schematic diagram of a terminal device provided in an embodiment of the present invention. Detailed Implementation

[0047] To make the objectives, technical solutions, and effects of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0048] This embodiment provides a multi-objective real-time scheduling method based on prediction and energy storage mechanisms. Specifically, this embodiment first acquires predicted power load demand data and estimated ship arrival times. Based on these data, water levels within the scheduling period are initialized to obtain an initial population, where the initial population represents the water level for each scheduling period. Then, the initial population undergoes constraint correction to obtain a corrected population. Navigation objectives and grid peak-shaving objectives are acquired. Based on these objectives, ship passage conditions are determined, and the use of energy storage devices for power generation is assessed. Next, the corrected population is subjected to non-dominated hierarchical sorting based on optimal power generation and navigation efficiency. Genetic algorithms are used for selection, crossover, and mutation operations to generate a new population. This new population is then subjected to non-dominated hierarchical sorting and grading based on optimal power generation and navigation efficiency. Finally, the corrected population is selected according to the hierarchical classification until the number of selected populations equals the number of initial populations. A termination condition is then determined, and when the termination condition is met, the result is output, which represents the water level of the hydropower station for each time period after scheduling. This invention can obtain the scheduling target demand of the hydropower station in real time through power load forecasting and ship arrival time forecasting; through an energy storage mechanism, it can solve the problem of insufficient power generation by the hydropower station during ship navigation, thereby enabling the hydropower station to simultaneously meet the needs of power generation and navigation, extending the actual navigation time.

[0049] Exemplary methods

[0050] The multi-objective real-time scheduling method based on prediction and energy storage mechanisms of the present invention is applied to terminal devices, including intelligent product terminals such as computers and mobile phones. Specifically, as shown in the figure... Figure 1 As shown in the figure, the multi-objective real-time scheduling method based on prediction mechanism and energy storage mechanism in this embodiment includes the following steps:

[0051] Step S100: Obtain the predicted power load demand data and the expected arrival time of the ships, and initialize the water level during the scheduling period based on the power load demand data and the expected arrival time of the ships to obtain an initial population, wherein the initial population is the water level during each scheduling period.

[0052] This embodiment is mainly divided into three parts. The first part uses machine learning and deep learning algorithms to predict the power load demand data of the hydropower station, comprehensively considering the impact of season, weather, and time of day on the power load demand of the hydropower station, and selecting prediction algorithms with high prediction accuracy. The second part uses the shipping trajectory of ships, taking into account factors such as the ship's latitude, longitude, speed, and load, to predict the expected arrival time of ships. The third part uses energy storage mechanisms to perform real-time scheduling of the hydropower station based on the expected arrival time of ships and the power load demand data, ensuring the navigation time and power generation efficiency of the hydropower station.

[0053] Specifically, such as Figure 2 As shown in the diagram, this embodiment first acquires historical load data and analyzes it. The historical load data refers to the historical power load data of the hydropower station. Then, a prediction algorithm is used to predict the load for a future period, obtaining the power load demand data. In this embodiment, as... Figure 3 As shown, when predicting electricity demand data, the prediction algorithm can be Support Vector Machine (SVR), Long Short Memory Network (LSTM), Gated Recurrent Neural Network (GRU), Differential Autoregressive Moving Average (ARIMA), etc. In specific applications, this embodiment can use historical load data to train the above models to obtain trained models, and then predict electricity load demand data based on the trained model.

[0054] This embodiment also acquires historical navigation data of ships in the waterway and uses a deep learning algorithm to build a multi-layer neural network model. The historical navigation data includes the historical times of ships arriving at and passing through the locks of the hydropower station within a past period. Then, based on the multi-layer neural network model and the historical navigation data, the estimated arrival time of the ship at the hydropower station locks is predicted, thus obtaining the estimated arrival time of the ship. In specific applications, this embodiment can also use historical navigation data to train the neural network model to obtain a multi-layer neural network model, and then automatically predict the time based on the multi-layer neural network model to obtain the estimated arrival time of the ship. In this embodiment, when predicting these two parts, the data with the highest prediction accuracy is compared and transmitted to the real-time scheduling module.

[0055] After obtaining the power load data and the estimated arrival time of ships, this embodiment further encodes the water level of the hydropower station within each scheduling period according to time to obtain the initial population. For example, taking a day as an example, if the scheduling period is 1 hour, then the encoding length is 24, so 24 initial populations can be obtained, and each initial population corresponds to the water level within a scheduling period.

[0056] Step S200: Perform constraint correction on the initial population to obtain the corrected population, and obtain the shipping target and the power grid peak shaving target. Based on the shipping target and the power grid peak shaving target, determine the ship passage conditions, and determine whether to use the energy storage device to generate electricity based on the ship passage conditions.

[0057] Specifically, this embodiment first obtains preset constraints, which include: hydropower station unit constraints, downstream flow constraints, initial and final water level constraints of the hydropower station, and ship navigation conditions. The hydropower station unit constraints include generator start-up and shutdown time constraints and generator output constraints. Then, based on these constraints, the initial population is screened, removing some populations that do not meet the constraints, resulting in a corrected population. This correction of the initial population in this embodiment facilitates more precise scheduling of the hydropower station's water level in subsequent steps.

[0058] During real-time dispatching, when ships are passing through, the hydropower station units cannot be shut down to maintain the necessary conditions for passage; therefore, stable power generation is required. Power generation dispatching needs to remain consistent with the dispatching plan of the previous period. To this end, this embodiment determines the ship passage conditions based on the navigation target and the power grid peak-shaving target, i.e., determining the hydropower station's navigation time and power generation benefits. Then, based on the ship passage conditions, it is determined whether the ship can pass. If the ship can pass, it is determined whether the power generation is sufficient. If the power generation is insufficient, energy storage devices are used to generate electricity. If the ship cannot pass, power grid peak-shaving is implemented to ensure smooth passage. It can be seen that the ship passage conditions in this embodiment require stable discharge flow and water level during ship passage, preventing unit shutdown and requiring stable power generation. Therefore, if power generation is insufficient during navigation, energy storage devices are used for discharge to meet the power grid's peak-shaving load requirements. Based on this mechanism, this embodiment utilizes energy storage to supply power to the grid in cases of insufficient power generation during navigation, ensuring that neither navigation benefits nor power generation benefits are affected.

[0059] Step S300: Perform non-dominated hierarchical sorting on the corrected population based on optimal power generation efficiency and optimal navigation efficiency; use genetic algorithm for selection, crossover and mutation operations to generate a new population; and perform non-dominated hierarchical sorting and grading on the generated new population based on optimal power generation efficiency and optimal navigation efficiency.

[0060] Specifically, this embodiment uses a non-dominated sorting genetic algorithm to analyze the corrected population. Since the corrected population corresponds to the water level during each scheduling period, the most suitable population can be selected based on the objectives of optimal power generation and optimal navigation benefits. Specifically, this embodiment performs non-dominated hierarchical sorting on the corrected population based on optimal power generation and optimal navigation benefits, uses a genetic algorithm for selection, crossover, and mutation operations to generate a new population. The generated new population is then subjected to non-dominated hierarchical sorting and grading based on optimal power generation and optimal navigation benefits. In this embodiment, grading can be achieved by calculating congestion distance, thereby realizing the water level scheduling of the hydropower station.

[0061] Step S400: Select the corrected population according to the grade classification until the number of selected populations is equal to the number of initial populations, determine the termination condition, and output the result when the termination condition is met. The result is the water level of the hydropower station for each time period after scheduling.

[0062] After analyzing the corrected population using a non-dominated sorting genetic algorithm, this embodiment prioritizes the selected population based on hierarchical division until the number of selected populations equals the initial population size. At this point, the analysis using the non-dominated sorting genetic algorithm is stopped. A termination condition is then determined; in this embodiment, the termination condition is that the similarity between all populations is at the same level. If the termination condition is met, the result is output, which is the water level of the hydropower station for each time period after scheduling. If the termination condition is not met, the process is repeated. Figure 3 Genetic operations are performed in order to re-evaluate the termination condition until it is met.

[0063] In summary, this embodiment utilizes predicted power load data for scheduling, rather than historical power load data. This ensures the reliability of the scheduling data and increases the credibility of the scheduling. This embodiment predicts the arrival time of vessels applying for passage to the hydropower station, ensuring that the hydropower station's revenue is not affected by errors in the scheduling plan due to various uncontrollable factors causing vessels to fail to arrive within the declared time, and preventing excessively long vessel waiting times. This embodiment incorporates an energy storage mechanism to ensure that the hydropower station can meet the grid's power load demand during vessel passage, solving the problem of vessels being unable to pass due to grid peak shaving. This embodiment allows for real-time scheduling; when vessels are passing through, scheduling is performed using energy storage devices based on the current grid load demand, thereby achieving the dual goals of optimizing navigation and power generation.

[0064] Exemplary device

[0065] Based on the above embodiments, the present invention also provides a multi-objective real-time scheduling device based on a prediction mechanism and an energy storage mechanism, such as... Figure 4As shown, the device in this embodiment includes: a prediction module 10, a population correction module 20, a multi-objective scheduling module 30, and a result output module 40. Specifically, the prediction module 10 is used to acquire predicted power load demand data and the expected arrival time of ships, and initialize the water level within the scheduling period based on the power load demand data and the expected arrival time of ships to obtain an initial population, wherein the initial population is the water level within each scheduling period. The population correction module 20 is used to perform constraint correction on the initial population to obtain a corrected population, and acquire the shipping target and the power grid peak-shaving target, determine the ship passage conditions based on the shipping target and the power grid peak-shaving target, and determine whether to use the energy storage device to generate electricity based on the ship passage conditions. The multi-objective scheduling module 30 is used to perform non-dominated hierarchical sorting on the corrected population based on optimal power generation efficiency and optimal navigation efficiency, use genetic algorithm selection, crossover and mutation operations to generate a new population, and perform non-dominated hierarchical sorting and level division on the generated new population based on optimal power generation efficiency and optimal navigation efficiency. The result output module 40 is used to select the corrected population according to the level classification until the number of selected populations is equal to the number of initial populations, determine the termination condition, and output the result when the termination condition is met. The result is the water level of the hydropower station for each time period after scheduling.

[0066] In one implementation, the prediction module 10 of this embodiment includes:

[0067] The historical load data analysis unit is used to acquire historical load data and analyze the historical load data, wherein the historical load data is the historical power load data of the hydropower station.

[0068] The power demand data prediction unit is used to predict the load for a future period of time using a prediction algorithm to obtain the power load demand data.

[0069] In one implementation, the prediction module 10 of this embodiment further includes:

[0070] The historical navigation data analysis unit is used to acquire historical navigation data of ships in the waterway and to build a multi-layer neural network model using deep learning algorithms. The historical navigation data includes the historical time of ships arriving at the hydropower station and passing through the lock in the past period.

[0071] The ship arrival time prediction unit is used to predict the estimated arrival time of a ship at the hydropower station lock based on the multi-layer neural network model and the historical navigation data, thereby obtaining the estimated arrival time of the ship.

[0072] In one implementation, the prediction module 10 of this embodiment further includes:

[0073] The water level coding unit is used to encode the water level of the hydropower station in each scheduling period according to time to obtain the initial population.

[0074] In one implementation, the population correction module 20 includes:

[0075] The constraint condition acquisition unit is used to acquire preset constraint conditions, wherein the constraint conditions are: hydropower station unit constraints, downstream flow constraints, hydropower station initial and final water level constraints, and ship navigation conditions, wherein the hydropower station unit constraints include hydropower station unit start-up and shutdown time constraints and unit output constraints.

[0076] The population correction unit is used to filter the initial population based on the constraints to obtain a corrected population.

[0077] In one implementation, the population correction module 20 further includes:

[0078] The passage condition determination unit is used to determine the ship passage conditions based on the shipping target and the power grid peak-shaving target;

[0079] The vessel passage determination unit is used to determine whether a vessel can pass based on the vessel passage conditions.

[0080] The power generation determination unit is used to determine whether the power generation is sufficient if the ship can pass through.

[0081] A power generation control unit is used to generate electricity using energy storage devices if the power generation is insufficient.

[0082] The conditions for ship passage are that the discharge flow and water level must be stable, the generating units cannot be shut down, and the power generation must be stable.

[0083] The working principle of each module in the multi-objective real-time scheduling device based on prediction and energy storage mechanisms in this embodiment is the same as that of each step in the above method embodiment, and will not be repeated here.

[0084] Based on the above embodiments, the present invention also provides a terminal device, the principle block diagram of which is shown in Figure 5. The terminal device may include one or more processors 100 (… Figure 5(Only one is shown in the image), a memory 101, and a computer program 102 stored in the memory 101 and executable on one or more processors 100, such as a program for controlling an app's theme scene. When one or more processors 100 execute the computer program 102, they can implement the various steps in the method embodiment for controlling an app's theme scene. Alternatively, when one or more processors 100 execute the computer program 102, they can implement the functions of each module / unit in the apparatus embodiment for multi-objective real-time scheduling based on prediction and energy storage mechanisms, which is not limited here.

[0085] In one embodiment, the processor 100 may be a central processing unit (CPU), or 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. The general-purpose processor may be a microprocessor or any conventional processor.

[0086] In one embodiment, memory 101 may be an internal storage unit of an electronic device, such as a hard drive or RAM. Memory 101 may also be an external storage device of the electronic device, such as a plug-in hard drive, smart memory card (SMC), secure digital card (SD) card, flash memory card, etc. Furthermore, memory 101 may include both internal and external storage units. Memory 101 is used to store computer programs and other programs and data required by the terminal device. Memory 101 can also be used to temporarily store data that has been output or will be output.

[0087] Those skilled in the art will understand that Figure 5 The block diagram shown is merely a partial structural diagram related to the present invention and does not constitute a limitation on the terminal device to which the present invention is applied. The specific terminal device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0088] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the methods described above. Any references to memory, storage, operational databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual operating data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAM bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0089] 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 foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-objective real-time scheduling method based on prediction and energy storage mechanisms, characterized in that, The method includes: Obtain predicted power load demand data and estimated ship arrival times, and initialize the water level during the scheduling period based on the power load demand data and estimated ship arrival times to obtain an initial population, wherein the initial population is the water level during each scheduling period; The initial population is subject to constraint correction to obtain the corrected population, and the shipping target and the power grid peak shaving target are obtained. Based on the shipping target and the power grid peak shaving target, the ship passage conditions are determined, and it is determined whether to use the energy storage device to generate electricity based on the ship passage conditions. The modified population is sorted in a non-dominated hierarchical manner based on the optimal power generation efficiency and the optimal navigation efficiency. A genetic algorithm is used for selection, crossover and mutation operations to generate a new population. The generated new population is then sorted in a non-dominated hierarchical manner and classified into levels based on the optimal power generation efficiency and the optimal navigation efficiency. The corrected population is selected according to the classification until the number of selected populations is equal to the number of initial populations. The termination condition is determined and the result is output when the termination condition is met. The result is the water level of the hydropower station for each time period after scheduling. The step of determining ship passage conditions based on the shipping target and the power grid peak-shaving target, and determining whether to utilize energy storage devices to generate electricity based on the ship passage conditions, includes: Determine the ship passage conditions based on the shipping objectives and the power grid peak-shaving objectives; Based on the aforementioned vessel passage conditions, determine whether the vessel is permitted to pass; If ships can pass through, determine whether the power generation is sufficient; If the power generation is insufficient, energy storage devices are used to generate electricity.

2. The multi-objective real-time scheduling method based on prediction and energy storage mechanisms according to claim 1, characterized in that, The acquisition of predicted electricity load demand data and estimated ship arrival times includes: Historical load data is acquired and analyzed, wherein the historical load data is the historical power load data of the hydropower station; The power load demand data is obtained by using a prediction algorithm to predict the load over a future period.

3. The multi-objective real-time scheduling method based on prediction and energy storage mechanisms according to claim 2, characterized in that, The acquisition of predicted electricity load demand data and estimated ship arrival times includes: Historical navigation data of ships in the waterway is acquired, and a multi-layer neural network model is built using deep learning algorithms. The historical navigation data includes the historical time of ships arriving at the hydropower station and passing through the lock in the past period. Based on the multi-layer neural network model and the historical navigation data, the estimated arrival time of the ship at the hydropower station is predicted, thus obtaining the estimated arrival time of the ship.

4. The multi-objective real-time scheduling method based on prediction and energy storage mechanisms according to claim 1, characterized in that, The initialization of water levels during the scheduling period to obtain an initial population includes: The water level of the hydropower station during each scheduling period is encoded according to time to obtain the initial population.

5. The multi-objective real-time scheduling method based on prediction and energy storage mechanisms according to claim 1, characterized in that, The process of performing restrictive modifications on the initial population to obtain the modified population includes: Obtain preset constraints, wherein the constraints are: hydropower station unit constraints, downstream flow constraints, hydropower station initial and final water level constraints, and ship navigation conditions, wherein the hydropower station unit constraints include hydropower station unit start-up and shutdown time constraints and unit output constraints. Based on the aforementioned constraints, the initial population is screened to obtain a revised population.

6. The multi-objective real-time scheduling method based on prediction and energy storage mechanisms according to claim 1, characterized in that, The conditions for ship passage are that the discharge flow and water level must be stable during ship passage, the generating units cannot be shut down, and the power generation must be stable.

7. A multi-objective real-time scheduling device based on a prediction mechanism and an energy storage mechanism, characterized in that, The apparatus is used to implement the steps of the multi-objective real-time scheduling method based on prediction mechanism and energy storage mechanism as described in any one of claims 1-6, and the apparatus includes: The prediction module is used to acquire predicted power load demand data and estimated ship arrival times, and initialize the water level during the scheduling period based on the power load demand data and estimated ship arrival times to obtain an initial population, wherein the initial population is the water level during each scheduling period. The population correction module is used to perform a constraint correction on the initial population to obtain the corrected population, and to obtain the shipping target and the power grid peak shaving target. Based on the shipping target and the power grid peak shaving target, the module determines the ship passage conditions and determines whether to use the energy storage device to generate electricity based on the ship passage conditions. The multi-objective scheduling module is used to perform non-dominated hierarchical sorting of the corrected population based on the optimal power generation benefit and the optimal navigation benefit, use genetic algorithm selection, crossover and mutation operations to generate a new population, and perform non-dominated hierarchical sorting and level division of the generated new population based on the optimal power generation benefit and the optimal navigation benefit. The result output module is used to select the corrected population according to the level classification until the number of selected populations is equal to the number of initial populations, determine the termination condition, and output the result when the termination condition is met. The result is the water level of the hydropower station for each time period after scheduling.

8. A terminal device, characterized in that, The terminal device includes a memory, a processor, and a multi-objective real-time scheduling program based on a prediction mechanism and an energy storage mechanism, which is stored in the memory and can run on the processor. When the processor executes the multi-objective real-time scheduling program based on the prediction mechanism and an energy storage mechanism, it implements the steps of the multi-objective real-time scheduling method based on the prediction mechanism and an energy storage mechanism as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a multi-objective real-time scheduling program based on a prediction mechanism and an energy storage mechanism. When the multi-objective real-time scheduling program based on the prediction mechanism and the energy storage mechanism is executed by a processor, it implements the steps of the multi-objective real-time scheduling method based on the prediction mechanism and the energy storage mechanism as described in any one of claims 1-6.

Images