Port microgrid optimization scheduling method, device, equipment, storage medium and product

Abstract

The invention belongs to the technical field of energy management and power system optimization, and particularly discloses a port microgrid optimization scheduling method, device and equipment, a storage medium and a product. According to the application, the microgrid operation optimization model is constructed according to the predicted load demand, the predicted wind power generation power and the predicted solar power generation power; solving a target scheduling strategy according to the power grid operation optimization model and the multi-dimensional microgrid operation constraint conditions; and performing optimal scheduling on the port microgrid according to the target scheduling strategy. Through the above mode, comprehensive optimization of microgrid operation is realized by using the microgrid operation optimization model on the basis of considering the operation cost and the environment cost, the operation stability is ensured according to the multi-dimensional microgrid operation constraint condition, and then optimization scheduling is performed on the port microgrid according to the target scheduling strategy. Therefore, the accuracy of dispatching the port micro-grid can be effectively improved, the operation economy and the energy utilization efficiency of the micro-grid are improved, and carbon emission is reduced.

Description

Technical Field

[0001] This application belongs to the field of energy management and power system optimization technology, and more specifically, relates to port microgrid optimization scheduling methods, devices, equipment, storage media and products. Background Technology

[0002] Traditional port energy systems rely heavily on fossil fuels, leading to increased energy costs and exacerbating environmental pollution. Developing low-carbon, efficient energy systems is urgently needed to achieve sustainable port development. Port microgrids, as a novel energy system, can meet these needs by integrating various distributed energy sources such as wind power, solar power, and energy storage. However, the randomness and uncertainty of renewable energy sources like wind and solar power, along with the volatility of port load demand, make the optimal scheduling of port microgrids a significant technical challenge.

[0003] Currently, the common dispatching method for port microgrids is manual dispatching, where personnel continuously experiment and optimize based on historical experience and precedents from other fields. However, the operating conditions of port microgrids are complex and variable, making manual dispatching unsuitable. Consequently, the accuracy of dispatching port microgrids using this method is relatively low. Summary of the Invention

[0004] In view of the shortcomings of the prior art, the purpose of this application is to provide a method, device, equipment, storage medium and product for optimizing the scheduling of port microgrids, which aims to solve the problem of low accuracy in scheduling port microgrids in the prior art.

[0005] To achieve the above objectives, in a first aspect, this application provides a method for optimized scheduling of port microgrids, comprising: Predicted wind power generation capacity is determined based on meteorological information and historical wind power generation information associated with wind power generation, and predicted solar power generation capacity is determined based on meteorological information and historical solar power generation information associated with solar power generation. The predicted load demand is determined based on the port's historical load data, and a microgrid operation optimization model is constructed based on the predicted load demand, the predicted wind power generation, and the predicted solar power generation. Multi-dimensional microgrid operation constraints are constructed, and the target scheduling strategy is solved based on the target optimization algorithm, according to the power grid operation optimization model and the multi-dimensional microgrid operation constraints. The port microgrid is optimized and scheduled according to the target scheduling strategy.

[0006] In one embodiment, the step of determining the predicted wind power output based on meteorological information and historical wind power generation information associated with wind power generation includes: The target mode decomposition strategy is used to decompose historical wind power generation information to obtain multiple wind eigenmode function components; Feature extraction of meteorological information associated with wind power generation; Each of the wind eigenmode function components is combined with the extracted meteorological features to obtain the current combined data; The current combined data is formatted and then normalized. Using a long short-term memory model with an attention mechanism, the wind power generation corresponding to each of the wind intrinsic mode function components is predicted based on the current combined data after normalization. The long short-term memory model with an attention mechanism is obtained by optimizing the learning rate and number of neurons of the original long short-term memory model using the whale optimization algorithm and an attention mechanism. The predicted wind power output is obtained by superimposing the individual wind power outputs.

[0007] In one embodiment, the step of determining the predicted load demand based on the port's historical load data includes: The historical load data of the port is decomposed using the target mode decomposition strategy to obtain multiple load intrinsic mode function components; The format of each load intrinsic mode function component is converted, and the converted load intrinsic mode function components are standardized to obtain multiple standardized load intrinsic mode function components. The Long Short-Term Memory model, which learns long-term dependencies in time-series data, predicts individual load demands corresponding to the various standardized load intrinsic mode function components. The predicted load demand is obtained by superimposing the individual load demands.

[0008] In one embodiment, the step of constructing a microgrid operation optimization model based on the predicted load demand, the predicted wind power generation, and the predicted solar power generation includes: The main grid's electricity purchase cost is determined based on the main grid's electricity purchase price and power consumption, and the main grid's electricity sales cost is determined based on the main grid's electricity sales price and power consumption. The current operation and maintenance cost is determined based on the proportion coefficient of the operation and maintenance cost of each part and the output power of each part, and the current depreciation cost is determined based on the purchase and installation cost of each part and the service life of each part. The operating cost function for the microgrid during the minimum time period is generated based on the main grid electricity purchase cost, the main grid electricity sales cost, the current operation and maintenance cost, and the current depreciation cost. The cost of treating pollutants by purchasing electricity from the main grid for the microgrid is determined based on the cost coefficient for treating the target type of pollutants, the emissions of the target type of pollutants generated by the operation of the main grid, and the power purchased by the microgrid from the main grid at the current moment. The microgrid environmental cost function is determined based on the pollutant treatment costs. A microgrid operation optimization model is constructed based on the predicted load demand, the predicted wind power generation, the predicted solar power generation, the microgrid operating cost function within the minimum time period, and the microgrid environmental cost function.

[0009] In one embodiment, the step of constructing multi-dimensional microgrid operating constraints includes: Power balance constraints are constructed based on the output power of photovoltaic systems, wind turbines, fuel cells, water electrolysis hydrogen production equipment, energy storage output power, power consumption of energy storage equipment when storing electricity, main grid power parameters, and current load power. Construct output constraints for distributed power sources based on their minimum and maximum output values; Construct energy storage constraints based on the current discharge power and current charging power of the energy storage system, the minimum discharge power and maximum discharge power of the energy storage device, the minimum charging power and maximum charging power of the energy storage device, the state variables of the energy storage system, and the minimum and maximum energy storage capacity of the energy storage system. The constraints of the hydrogen storage tank are constructed based on the current hydrogen release power and current hydrogen charging power, minimum hydrogen charging power, maximum hydrogen charging power, state variables of the hydrogen storage system, minimum hydrogen storage capacity, and maximum hydrogen storage capacity. Based on the power purchased, the power sold, and the rules for power purchase and sales, the constraints for power purchase and sales are constructed. Based on the power balance constraint, the distributed power output constraint, the energy storage constraint, the hydrogen storage tank constraint, and the power purchase and sale constraint, multi-dimensional microgrid operation constraints are obtained.

[0010] In one embodiment, the step of solving the target scheduling strategy based on the target optimization algorithm, according to the power grid operation optimization model and the multi-dimensional microgrid operation constraints, includes: Based on the objective optimization algorithm, a microgrid scheduling strategy and strategy adjustment parameters are randomly assigned to each particle according to the power grid operation optimization model and the multi-dimensional microgrid operation constraints. Calculate the fitness value of each particle after allocation, and solve for the individual optimal position and the global optimal position based on the fitness value; The microgrid scheduling strategy and strategy adjustment parameters for each particle are updated based on the individual optimal position and the global optimal position. When the preset termination conditions are met, the updated optimal scheduling policy will be used as the target scheduling policy.

[0011] Secondly, this application provides a port microgrid optimized dispatching device, comprising: The prediction module is used to determine the predicted wind power output based on meteorological information and historical wind power output associated with wind power generation, and to determine the predicted solar power output based on meteorological information and historical solar power output associated with solar power generation. The module is used to determine the predicted load demand based on the port's historical load data, and to build a microgrid operation optimization model based on the predicted load demand, the predicted wind power generation, and the predicted solar power generation. The construction module is also used to construct multi-dimensional microgrid operation constraints and, based on the target optimization algorithm, solve the target scheduling strategy according to the power grid operation optimization model and the multi-dimensional microgrid operation constraints. The scheduling module is used to optimize the scheduling of the port microgrid according to the target scheduling strategy.

[0012] Thirdly, this application provides an electronic device, comprising: at least one memory for storing a program; and at least one processor for executing the program stored in the memory, wherein when the program stored in the memory is executed, the processor is configured to execute the method described in the first aspect or any possible implementation thereof.

[0013] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when run on a processor, causes the processor to perform the method described in the first aspect or any possible implementation thereof.

[0014] Fifthly, this application provides a computer program product that, when run on a processor, causes the processor to perform the method described in the first aspect or any possible implementation thereof.

[0015] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here.

[0016] Overall, the technical solutions conceived in this application have the following beneficial effects compared with the prior art: (1) By optimizing the learning rate and number of neurons of the original long short-term memory (LSTM) model using the whale optimization algorithm and attention mechanism, an LSTM model with attention mechanism is obtained. This LSTM model with attention mechanism is then used to predict wind power generation and solar power generation, and a LSTM model that learns long-term dependencies in time series data is used to predict load demand, effectively improving the accuracy of the predicted parameters. By constructing a microgrid operation optimization model using these parameters, and considering operating costs and environmental costs, comprehensive optimization of microgrid operation is achieved, improving energy efficiency and reducing carbon emissions.

[0017] (2) This application constructs multi-dimensional microgrid operation constraints, including power balance constraints, distributed power output constraints, energy storage constraints, and hydrogen storage tank constraints, to ensure the stability and reliability of the port microgrid under various operating conditions. Using an objective optimization algorithm to solve for the optimal target scheduling strategy can effectively improve the economic efficiency of microgrid operation, reduce operating costs, and increase economic benefits. Through the grid operation optimization model and the target scheduling strategy, negative environmental impacts can be reduced, promoting the green and low-carbon operation of the port microgrid.

[0018] The predicted wind power generation is determined based on meteorological and historical wind power generation information associated with wind power generation, and the predicted solar power generation is determined based on meteorological and historical solar power generation information associated with solar power generation. The predicted load demand is determined based on historical load data of the port, and a microgrid operation optimization model is constructed based on the predicted load demand, predicted wind power generation, and predicted solar power generation. Multi-dimensional microgrid operation constraints are constructed, and a target scheduling strategy is solved based on the target optimization algorithm, the microgrid operation optimization model, and the multi-dimensional microgrid operation constraints. The port microgrid is then optimized and scheduled according to the target scheduling strategy. Through the above methods, after predicting the predicted load demand, predicted wind power generation, and predicted solar power generation, a microgrid operation optimization model is constructed using these parameters. Considering operating costs and environmental costs, comprehensive optimization of microgrid operation is achieved. The multi-dimensional microgrid operation constraints ensure operational stability, and the port microgrid is optimized and scheduled according to the target scheduling strategy. This effectively improves the accuracy of port microgrid scheduling, thereby improving the economic efficiency and energy utilization efficiency of microgrid operation, and reducing carbon emissions. Attached Figure Description

[0019] Figure 1 This is one of the flowcharts illustrating the port microgrid optimization scheduling method provided in the embodiments of this application; Figure 2 This is a system architecture diagram of the port microgrid provided in the embodiments of this application; Figure 3 This is a schematic diagram of the process for training a long short-term memory model with an attention mechanism, as provided in the embodiments of this application. Figure 4 This is a schematic diagram of the optimized scheduling results provided in the embodiments of this application; Figure 5 This is the second flowchart illustrating the port microgrid optimization scheduling method provided in the embodiments of this application; Figure 6 This is a schematic diagram of the module structure of the port microgrid optimized scheduling device provided in the embodiments of this application; Figure 7 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0021] In this article, the term "and / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The symbol " / " in this article indicates that the related objects are in an "or" relationship; for example, A / B means A or B.

[0022] The terms "first" and "second," etc., used in the specification and claims herein are used to distinguish different objects, not to describe a specific order of objects. For example, "first response message" and "second response message," etc., are used to distinguish different response messages, not to describe a specific order of response messages.

[0023] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0024] Based on this, the embodiments of this application provide a method for optimized scheduling of port microgrids, referring to... Figure 1 , Figure 1 This is one of the flowcharts illustrating the port microgrid optimization scheduling method provided in this application embodiment. In this embodiment, the port microgrid optimization scheduling method includes steps S10 to S40: Step S10: Determine the predicted wind power generation capacity based on meteorological information and historical wind power generation information associated with wind power generation, and determine the predicted solar power generation capacity based on meteorological information and historical solar power generation information associated with solar power generation.

[0025] It should be noted that meteorological information associated with wind power generation includes, but is not limited to, wind speed, temperature, and air pressure. To effectively improve the accuracy of wind power generation forecasts, historical wind power generation information, which can be historical wind power generation figures, needs to be incorporated into the forecast. Similarly, meteorological information associated with solar power generation includes, but is not limited to, irradiance and temperature. Likewise, to effectively improve the accuracy of solar power generation forecasts, historical solar power generation information, which can be historical solar power generation figures, needs to be incorporated into the forecast.

[0026] It should be understood that predicting wind power generation and solar power generation can be done in the same way. Specifically, the learning rate and number of neurons of the original Long Short-Term Memory (LSTM) model are optimized using the whale optimization algorithm and an attention mechanism to obtain an LTM model with an attention mechanism. This attention-based LTM model is then used to predict wind power generation and solar power generation, respectively. (Reference) Figure 2 , Figure 2 The system architecture diagram of the port microgrid is as follows: the entire system architecture is divided into two parts with the AC bus as the boundary. The first part includes the main grid, wind power system, photovoltaic system and load system, which represents the conversion between AC and DC. The second part includes hydrogen energy storage system and electrochemical energy storage system, which also represent the conversion between AC and DC.

[0027] Further, the step of determining the predicted wind power generation based on meteorological information and historical wind power generation information associated with wind power generation includes: decomposing historical wind power generation information using a target mode decomposition strategy to obtain multiple wind intrinsic mode function components; extracting features from the meteorological information associated with wind power generation; combining each of the wind intrinsic mode function components with the extracted meteorological features to obtain current combined data; converting the format of the current combined data and normalizing the converted data; predicting the wind power generation corresponding to each of the wind intrinsic mode function components based on the normalized current combined data using a long short-term memory model with an attention mechanism; wherein the long short-term memory model with an attention mechanism is obtained by optimizing the learning rate and number of neurons of the original long short-term memory model using a whale optimization algorithm and an attention mechanism; and superimposing the wind power generation values ​​to obtain the predicted wind power generation.

[0028] It is understandable that the target mode decomposition strategy refers to the strategy of decomposing information into multiple intrinsic mode function components, thereby decomposing a non-stationary time series into several intrinsic mode function components. Each intrinsic mode function component represents a part of different frequency components in the original information, which helps to reveal patterns hidden in complex information. This target mode decomposition strategy can be a variational mode decomposition (VMD) strategy. Taking historical wind power generation information as the decomposition object, after decomposing it into multiple wind intrinsic mode function components using the target mode decomposition strategy, each wind intrinsic mode function component is combined with the extracted meteorological features. Then, the format of the current combined data is converted into a supervised learning format, and the prediction problem is constructed as a supervised learning problem, divided into training set and test set. Through normalization processing, the data is scaled to the [0,1] interval to eliminate the influence of different units and magnitudes, ensure data consistency and stability, and improve the training efficiency and prediction accuracy of subsequent prediction models.

[0029] It should be understood that when the object of decomposition is historical solar power generation information, the same method can be used. When predicting the solar power generation corresponding to each solar intrinsic mode function component, a long short-term memory model with attention mechanism can also be used, and then the solar power generation is superimposed to obtain the predicted solar power generation.

[0030] It should be noted that the reference Figure 3 , Figure 3 The flowchart illustrates the training process for a Long Short-Term Memory (LSTM) model with an attention mechanism. Specifically, it introduces the Whale Optimization Algorithm (WOA) to optimize the hyperparameters of the original LSTM model. The learning rate is used as a parameter of the Adam optimizer, specifically as the initial learning rate. The Adam optimizer adjusts the model parameters during training to minimize the loss function. This optimized initial learning rate guides the update of model weights during training, aiming for faster convergence and better model performance. The WOA algorithm searches for optimal parameters, using the original LSTM's learning rate and the number of hidden layer neurons as optimization targets. After obtaining the optimal parameters for the neural network, training is performed. The specific solution steps are as follows: (1) Taking the historical solar power generation information and five solar intrinsic mode function components as the decomposition objects, the VMD decomposition of the historical solar power generation information time series signal is performed and decomposed into five solar intrinsic mode function components. Each solar intrinsic mode function component is iterated. In each iteration, meteorological features are loaded, and the solar intrinsic mode function components and the extracted meteorological features are merged. The prediction problem is constructed into a supervised learning problem. The data is converted into a supervised learning format. The dataset can be divided into training set and test set according to 8:2. The data is normalized and then decomposed into input and output. The data is then imported into the LSTM-Attention model.

[0031] (2) Initialize the parameters of the WOA algorithm, including but not limited to the population size, maximum number of iterations, and search dimension. Define the search range for the initial learning rate and the number of hidden layer neurons, initialize the population, apply boundary conditions, and replace individuals outside the boundary with random values ​​within the boundary. Start the iteration, set the current best individual as the leader whale, update the position of each individual, and recalculate the fitness value. Specifically, if the updated value of an individual is less than the fitness value of the current best individual, then set that individual as the new leader whale and record the current best fitness. During the update process, the three main operations of the WOA algorithm need to be used to simulate the behavior of whales, including spiral hunting, encirclement hunting, and food searching. Whales will choose to perform spiral attacks or encirclement hunting according to the following formula:

[0032] in, Indicates the distance between the prey and the whale. Indicates the spiral shape coefficient. The size of the whale determines whether it will launch a spiral attack or surround and prey on its prey. The value takes a range from [0,1], representing a random probability. Represents the distance vector. Represents the convergence factor vector. The value is taken in the range [-1, 1] to control the direction and step size of the helix. Generally, the probability of helical attack and encirclement predation is 50% each. During encirclement predation, if... Then, the whale's position is updated according to the following formula:

[0033] in, This indicates the whale's current spatial location. This indicates the current spatial location of the optimal whale. This indicates the number of rounds to be calculated. If... Then, the whale's position is updated according to the following formula:

[0034] in, This indicates the reference whale's location, with a randomly selected value.

[0035] It is also important to emphasize that in calculating the fitness value, the learning rate and the number of neurons in the original LSTM hidden layer are first obtained from the population parameters. During the creation of the LSTM-Attention model, these two parameters are imported into the model for compilation and training. After training, the model is used for prediction. In this embodiment, Mean Squared Error (MSE) can be used as the fitness function for the whale algorithm to calculate the whale's fitness value. The fitness function can be expressed as:

[0036] in, This represents the fitness value of the whale. Indicates the total quantity. Indicates the predicted value. This represents the actual value.

[0037] It should be noted that after calculating the fitness values ​​of the whales, the fitness values ​​of the population are sorted, the global optimum is updated, and after the number of iterations reaches MaxIter, the optimal position is obtained. The optimal solution of the LSTM model obtained at this point using the WOA algorithm is the optimal parameter, including the optimal learning rate and the number of hidden layer neurons. These two parameters are input into the VMD-WOA-LSTM-Att-ention model for formal prediction. Each solar intrinsic mode function component can be loaded with the optimal learning rate and the number of hidden layer neurons during prediction. Initialization is performed according to the different characteristics of each solar intrinsic mode function component, which can accurately predict the power of each solar power generation. Then, the power of each solar power generation is superimposed to obtain the predicted solar power generation.

[0038] Step S20: Determine the predicted load demand based on the port's historical load data, and construct a microgrid operation optimization model based on the predicted load demand, the predicted wind power generation, and the predicted solar power generation.

[0039] It should be understood that the predicted load demand refers to the actual load required by the port in the future. After predicting the predicted load demand using the long short-term memory model, a microgrid operation optimization model is constructed by combining the predicted wind power generation and the predicted solar power generation. Among them, the multi-objective optimization model refers to the model constructed to achieve comprehensive optimization of microgrid operation based on considering operating costs and environmental costs.

[0040] Furthermore, the step of determining the predicted load demand based on the port's historical load data includes: decomposing the port's historical load data using a target mode decomposition strategy to obtain multiple load intrinsic mode function components; converting the format of each load intrinsic mode function component and standardizing the converted load intrinsic mode function components to obtain multiple standardized load intrinsic mode function components; predicting the individual load demand corresponding to each standardized load intrinsic mode function component by using a long short-term memory model that learns long-term dependencies in time series data; and superimposing the individual load demands to obtain the predicted load demand.

[0041] Understandably, for predicting load demand, the same approach can be used to decompose the port's historical load data, convert the format of each load intrinsic mode function component, and standardize it. In this case, the long short-term memory model needs to learn the long-term dependencies in the time series data and predict the individual load demand corresponding to each standardized load intrinsic mode function component. Then, the individual load demands are superimposed to form the predicted load demand.

[0042] Further, the step of constructing a microgrid operation optimization model based on the predicted load demand, the predicted wind power generation, and the predicted solar power generation includes: determining the main grid's electricity purchase cost based on the main grid's electricity purchase price and power generation, and determining the main grid's electricity sales cost based on the main grid's electricity sales price and power generation; determining the current operation and maintenance cost based on the proportion coefficient of each part's operation and maintenance cost and the output power of each part, and determining the current depreciation cost based on the purchase and installation cost of each part and the service life of each part; generating an operating cost function for the microgrid within the minimum time period based on the main grid's electricity purchase cost, the main grid's electricity sales cost, the current operation and maintenance cost, and the current depreciation cost; determining the pollutant treatment cost of the microgrid purchasing electricity from the main grid based on the cost coefficient for treating target pollutants, the emission of target pollutants generated by the main grid operation, and the current power generation of the microgrid purchasing electricity from the main grid; determining the microgrid's environmental cost function based on the pollutant treatment cost; and constructing a microgrid operation optimization model based on the predicted load demand, the predicted wind power generation, the predicted solar power generation, the operating cost function for the microgrid within the minimum time period, and the microgrid's environmental cost function.

[0043] It should be understood that before constructing the power grid operation optimization model, it is necessary to separately construct the operating cost function for the microgrid during the minimization period, which characterizes operating costs, and the environmental cost function for the microgrid, which characterizes environmental costs. On the one hand, the specific method for constructing the operating cost function for the microgrid during the minimization period can be as follows: (1) The cost of purchasing electricity from the main power grid is determined based on the purchase price and power capacity of the main power grid, specifically as follows:

[0044] in, This indicates the cost of purchasing electricity from the main power grid. This indicates the purchase price of electricity from the main power grid. This indicates the power purchased by the main power grid.

[0045] (2) The cost of electricity sales on the main power grid is determined based on the electricity sales price and sales capacity of the main power grid, specifically as follows:

[0046] in, This represents the cost of electricity sold through the main power grid. This indicates the electricity price sold through the main power grid. This indicates the power output of the main power grid.

[0047] (3) Determine the current operation and maintenance cost based on the proportion coefficient of each part's operation and maintenance cost and the output power of each part, specifically as follows:

[0048] in, This indicates the current operation and maintenance cost. This represents the proportion of operation and maintenance costs for each component. This indicates the output power of each component.

[0049] (4) Determine the current depreciation cost based on the purchase and installation costs of each part and the service life of each part, as follows:

[0050] in, This represents the current depreciation cost. Indicates the first Part of the purchase and installation costs, Indicates the first Partial service life.

[0051] (5) Generate the operating cost function for the microgrid during the minimum time period based on the main grid's electricity purchase cost, main grid's electricity sales cost, current operation and maintenance cost, and current depreciation cost, as follows:

[0052] in, This represents the operating cost function of the microgrid during the period in which it is minimized. Indicates the period.

[0053] On the other hand, the specific way to construct the microgrid environmental cost function can be: (1) The cost of treating pollutants from the main grid is determined based on the cost coefficient for treating the target type of pollutants, the emissions of the target type of pollutants generated by the operation of the main grid, and the power purchased by the microgrid from the main grid at the current moment. Specifically:

[0054] in, This represents the cost of pollutant treatment for the electricity purchased by the microgrid from the main grid. Indicates the type of pollutant. This indicates the amount of target pollutants emitted during the operation of a large power grid; This represents the amount of electricity the microgrid is purchasing from the main grid at the current moment. Indicates the current moment.

[0055] (2) The environmental cost function of the microgrid is determined based on the cost of pollutant treatment, as follows:

[0056] in, This represents the environmental cost function of a microgrid.

[0057] It should be noted that after constructing the operating cost function for the microgrid during the minimum period representing operating costs and the environmental cost function for the microgrid representing environmental costs, a grid operation optimization model is constructed by combining predicted load demand, predicted wind power generation, and predicted solar power generation. This enables comprehensive optimization of microgrid operation, as well as improved energy efficiency and reduced carbon emissions.

[0058] Step S30: Construct multi-dimensional microgrid operation constraints, and solve the target scheduling strategy based on the target optimization algorithm, according to the power grid operation optimization model and the multi-dimensional microgrid operation constraints.

[0059] It should be understood that the multi-dimensional microgrid operation constraints include, but are not limited to, power balance constraints, distributed generation output constraints, energy storage constraints, hydrogen storage tank constraints, and power purchase and sale constraints. Among these, the power balance constraint is used to ensure that the total power generation of the port microgrid matches the total load demand at any given time to maintain the stability of the power system; the distributed generation output constraint is used to limit the output of distributed power sources such as wind and solar power to prevent equipment overload or damage; the energy storage constraint is used to control the charging and discharging process of energy storage equipment, including charging and discharging rates and maximum energy storage capacity, to extend the service life of the equipment; the hydrogen storage tank constraint is used to control the hydrogen storage capacity of the hydrogen storage tank; and the power purchase and sale constraint is used to control the purchase and sale of electricity to be carried out simultaneously, and to ensure that both the purchased and sold power are within their maximum and minimum ranges.

[0060] Furthermore, the step of constructing multi-dimensional microgrid operation constraints includes: constructing power balance constraints based on the output power of the photovoltaic system, wind turbine output power, fuel cell output power, water electrolysis hydrogen production equipment output power, energy storage output power, power consumption of the energy storage device when storing electricity, main grid power parameters, and current load power; constructing distributed power generation output constraints based on the minimum and maximum output values ​​of distributed power sources; and constructing distributed power generation output constraints based on the current discharge and charging power of the energy storage system, the minimum and maximum discharge power of the energy storage device, and the minimum and maximum charging power of the energy storage device. Electric energy storage constraints are constructed based on electrical power, energy storage system state variables, minimum and maximum energy storage capacity of the energy storage system; hydrogen storage tank constraints are constructed based on the current hydrogen release power, current hydrogen charging power, minimum and maximum hydrogen charging power, hydrogen storage system state variables, minimum and maximum hydrogen storage capacity; electricity purchase and sale constraints are constructed based on the purchased power, sold power, and rules governing electricity purchase and sale; and multi-dimensional microgrid operation constraints are obtained based on the power balance constraints, distributed power output constraints, electric energy storage constraints, hydrogen storage tank constraints, and electricity purchase and sale constraints.

[0061] It is understandable that power balance constraints, distributed power output constraints, energy storage constraints, hydrogen storage tank constraints, and electricity purchase and sale constraints are constructed separately, and then multi-dimensional microgrid operation constraints are obtained from the constraints constructed above.

[0062] (1) The specific method for constructing power balance constraints is as follows:

[0063] in, Indicates the output power of the photovoltaic system. Indicates the output power of the fan. Indicates the output power of the fuel cell. This indicates the output power of the water electrolysis hydrogen production equipment. Indicates the energy storage output power. This indicates the power consumed by the energy storage device when storing electricity. This represents the main grid power parameter, which can be either the input or output power of the main grid. It is positive when purchasing electricity and negative when selling electricity. This represents the load power at the current moment. Indicates the current moment.

[0064] (2) The specific method for constructing the output constraints of distributed power sources is as follows:

[0065] in, Indicates the first The minimum output value of the distributed power source. Indicates the first The maximum output of the distributed power source.

[0066] (3) The specific method for constructing the constraints of electric energy storage is as follows:

[0067]

[0068]

[0069]

[0070]

[0071] in, and These represent the current discharge power and current charging power of the energy storage system, respectively. and These represent the minimum and maximum discharge power of the energy storage device, respectively. and These represent the minimum and maximum charging power of the energy storage device, respectively. and Both represent state variables of the energy storage system, namely energy storage discharge and energy storage charging, respectively. and These represent the minimum and maximum energy storage capacity of the energy storage system, respectively.

[0072] (4) The specific method for constructing the constraints of the hydrogen storage tank is as follows:

[0073]

[0074]

[0075]

[0076]

[0077]

[0078]

[0079] in, and These represent the current hydrogen release power and the current hydrogen charging power of the hydrogen storage system, respectively. and These represent the minimum and maximum hydrogen charging power of the hydrogen storage system, respectively. and These are all state variables of the hydrogen storage system, representing hydrogen release and hydrogen filling of the hydrogen storage system, respectively. and These represent the minimum and maximum hydrogen storage capacities, respectively.

[0080] Step S40: Optimize the scheduling of the port microgrid according to the target scheduling strategy.

[0081] It is understandable that the target scheduling strategy refers to the optimal strategy for optimizing the scheduling of the port microgrid. After solving the target scheduling strategy based on the target optimization algorithm, the port microgrid is optimized and scheduled according to the target scheduling strategy to guide the operation of the port microgrid.

[0082] It should be understood that, reference Figure 4 , Figure 4 To optimize the dispatching results, the specific steps are as follows: From 00:00 to 14:00, since wind power exceeds the load, electricity is sold to the main grid from 00:00 to 03:00 and from 06:00 to 14:00 to generate revenue. Battery charging occurs at 04:00 and 05:00, and hydrogen production via water electrolysis takes place at 04:00, 05:00, 07:00, 10:00, and 11:00. Secondly, from 17:00 to 21:00, wind and solar power output is low, so electricity is purchased from the main grid. Finally, during off-peak electricity prices at 04:00, 05:00, 22:00, and 23:00, electricity is purchased from the main grid and stored in energy storage and hydrogen storage systems. During peak electricity prices from 15:00 to 20:00, the electricity is released, achieving "low-storage, high-generation" arbitrage. This embodiment uses a target scheduling strategy to optimize the scheduling of the port microgrid, which can effectively reduce operating costs and environmental costs.

[0083] This embodiment determines the predicted wind power output based on meteorological information and historical wind power output associated with wind power generation, and determines the predicted solar power output based on meteorological information and historical solar power output associated with solar power generation. It also determines the predicted load demand based on historical port load data, and constructs a microgrid operation optimization model based on the predicted load demand, predicted wind power output, and predicted solar power output. Multi-dimensional microgrid operation constraints are constructed, and a target scheduling strategy is solved based on the target optimization algorithm, the microgrid operation optimization model, and the multi-dimensional microgrid operation constraints. The port microgrid is then optimized and scheduled according to the target scheduling strategy. By predicting the predicted load demand, predicted wind power output, and predicted solar power output, and then constructing a microgrid operation optimization model using these parameters, the microgrid operation is comprehensively optimized while considering operating and environmental costs. The multi-dimensional microgrid operation constraints ensure operational stability, and the target scheduling strategy optimizes the scheduling of the port microgrid. This effectively improves the accuracy of port microgrid scheduling, thereby enhancing the economic efficiency and energy utilization of the microgrid operation, and reducing carbon emissions.

[0084] In one specific implementation, this application provides steps for solving the target scheduling strategy. Please refer to... Figure 5 , Figure 5 This is the second flowchart illustrating the port microgrid optimized scheduling method provided in this application. It includes steps T401 to T404: Step T401: Based on the objective optimization algorithm, a microgrid scheduling strategy and strategy adjustment parameters are randomly assigned to each particle according to the power grid operation optimization model and the multi-dimensional microgrid operation constraints.

[0085] It should be noted that the objective optimization algorithm can be a multi-objective particle swarm optimization (MOPSO) algorithm, which uses the objective optimization algorithm to find the optimal solution set that satisfies all constraints, i.e., the Pareto front. In this case, a position and velocity can be randomly assigned to each particle based on the objective optimization algorithm. Here, the position represents the microgrid scheduling strategy, and the velocity represents the strategy adjustment parameters, which include, but are not limited to, adjustment direction and magnitude.

[0086] Step T402: Calculate the fitness value of each assigned particle, and solve for the individual optimal position and the global optimal position based on the fitness value.

[0087] Understandably, after randomly assigning a microgrid scheduling strategy and strategy adjustment parameters to each particle, the fitness value of each particle after assignment, i.e., operating cost and environmental cost, can be calculated according to the objective function of the power grid operation optimization model. For each particle, if its current position is better than the previously recorded individual optimal position, its individual optimal position is updated. Simultaneously, the individual optimal positions of all particles are compared, and the best position is selected as the global optimal position. Furthermore, the Pareto dominance relation is applied; that is, when updating particle positions, the Pareto dominance relation is used to ensure solution diversity and prevent the algorithm from prematurely converging to a local optimum.

[0088] Step T403: Update the microgrid scheduling strategy and strategy adjustment parameters for each particle based on the individual optimal position and the global optimal position.

[0089] It should be understood that after determining the optimal position for each individual particle and the global optimal position, updates are made based on the particle's current velocity and position; that is, the microgrid scheduling strategy and strategy adjustment parameters are updated. One specific method for updating the velocity is as follows: .

[0090] in, Indicates update speed. Indicates the inertia factor. Represents the local velocity factor. Represents the global velocity factor. This represents the optimal position of an individual. Indicates the globally optimal position. Indicates the current speed.

[0091] On the other hand, the specific method for updating the location is as follows:

[0092] in, Indicates updating the position. Indicates the initial position.

[0093] Step T404: When it is determined that the preset termination condition is met, the updated optimal scheduling policy is taken as the target scheduling policy.

[0094] It should be noted that during the update process, the above steps will be repeated until the preset termination condition is met. The preset termination condition can be reaching the maximum number of iterations or the optimization degree of the solution reaching a preset threshold. At this time, the final particle swarm position can be output, and the updated optimal scheduling strategy corresponding to the final particle swarm position can be used as the target scheduling strategy. This target scheduling strategy can then be used to guide the operation of the port microgrid.

[0095] This embodiment, based on a target optimization algorithm, randomly assigns a microgrid scheduling strategy and strategy adjustment parameters to each particle according to the power grid operation optimization model and the multi-dimensional microgrid operation constraints. It calculates the fitness value of each particle after assignment and solves for the individual optimal position and the global optimal position based on the fitness value. The microgrid scheduling strategy and strategy adjustment parameters for each particle are updated based on the individual optimal position and the global optimal position. When a preset termination condition is met, the updated optimal scheduling strategy is used as the target scheduling strategy. Through this method, by randomly assigning a microgrid scheduling strategy and strategy adjustment parameters to each particle based on a target optimization algorithm, solving for the individual optimal position and the global optimal position based on the fitness value, and then updating the strategy based on the particle's current velocity and current position, the above steps are repeatedly executed until the preset termination condition is met. At this point, the updated optimal scheduling strategy corresponding to the final particle swarm position is used as the target scheduling strategy, thereby effectively improving the accuracy of solving for the target scheduling strategy.

[0096] The port microgrid optimization scheduling device provided in this application is described below. The port microgrid optimization scheduling device described below corresponds to the port microgrid optimization scheduling method described above. Please refer to... Figure 6 , Figure 6 This is a schematic diagram of the module structure of the port microgrid optimized scheduling device provided in this application embodiment, including: The prediction module T10 is used to determine the predicted wind power generation based on meteorological information and historical wind power generation information associated with wind power generation, and to determine the predicted solar power generation based on meteorological information and historical solar power generation information associated with solar power generation.

[0097] The construction module T20 is used to determine the predicted load demand based on the port's historical load data, and to construct a microgrid operation optimization model based on the predicted load demand, the predicted wind power generation, and the predicted solar power generation.

[0098] The construction module T20 is also used to construct multi-dimensional microgrid operation constraints and, based on the target optimization algorithm, solve the target scheduling strategy according to the power grid operation optimization model and the multi-dimensional microgrid operation constraints.

[0099] The scheduling module T30 is used to optimize the scheduling of the port microgrid according to the target scheduling strategy.

[0100] This embodiment determines the predicted wind power output based on meteorological information and historical wind power output associated with wind power generation, and determines the predicted solar power output based on meteorological information and historical solar power output associated with solar power generation. It also determines the predicted load demand based on historical port load data, and constructs a microgrid operation optimization model based on the predicted load demand, predicted wind power output, and predicted solar power output. Multi-dimensional microgrid operation constraints are constructed, and a target scheduling strategy is solved based on the target optimization algorithm, the microgrid operation optimization model, and the multi-dimensional microgrid operation constraints. The port microgrid is then optimized and scheduled according to the target scheduling strategy. By predicting the predicted load demand, predicted wind power output, and predicted solar power output, and then constructing a microgrid operation optimization model using these parameters, the microgrid operation is comprehensively optimized while considering operating and environmental costs. The multi-dimensional microgrid operation constraints ensure operational stability, and the target scheduling strategy optimizes the scheduling of the port microgrid. This effectively improves the accuracy of port microgrid scheduling, thereby enhancing the economic efficiency and energy utilization of the microgrid operation, and reducing carbon emissions.

[0101] It is understood that the detailed functional implementation of each of the above modules can be found in the description of the aforementioned method embodiments, and will not be repeated here.

[0102] It should be understood that the above-described device is used to execute the methods in the above embodiments. The implementation principle and technical effect of the corresponding program modules in the device are similar to those described in the above methods. The working process of the device can be referred to the corresponding process in the above methods, and will not be repeated here.

[0103] Based on the methods in the above embodiments, this application provides an electronic device, please refer to... Figure 7 , Figure 7 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application.

[0104] It should be noted that the system may include: a processor 10, a communication interface (Co Port Microgrid Optimization and Scheduling Services Interface) 20, a memory 30, and a communication bus 40. The processor 10, communication interface 20, and memory 30 communicate with each other via the communication bus 40. The processor 10 can call logical instructions stored in the memory 30 to execute the methods described in the above embodiments.

[0105] Furthermore, the logical instructions in the aforementioned memory 30 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0106] Based on the methods in the above embodiments, this application provides a computer-readable storage medium storing a computer program that, when run on a processor, causes the processor to execute the methods in the above embodiments.

[0107] Based on the methods in the above embodiments, this application provides a computer program product that, when run on a processor, causes the processor to execute the methods in the above embodiments.

[0108] It is understood that the processor in the embodiments of this application can be a central processing unit, or other general-purpose processors, digital signal processors, application-specific integrated circuits, field-programmable gate arrays, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. A general-purpose processor can be a microprocessor or any conventional processor.

[0109] The method steps in this application embodiment can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in random access memory, flash memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, registers, hard disks, portable hard disks, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor.

[0110] It is understood that the various numerical designations used in the embodiments of this application are merely for descriptive convenience and are not intended to limit the scope of the embodiments of this application. Those skilled in the art will readily understand that the above descriptions are merely preferred embodiments of this application and are not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for optimal scheduling of a port microgrid, characterized in that, The method comprises the following steps: determining predicted wind power generation according to meteorological information and historical wind power generation information associated with wind power generation, and determining predicted solar power generation according to meteorological information and historical solar power generation information associated with solar power generation; determining predicted load demand according to historical load data of the port, and constructing a micro-grid operation optimization model according to the predicted load demand, the predicted wind power generation and the predicted solar power generation; constructing a multi-dimensional micro-grid operation constraint condition, and solving a target scheduling strategy according to the grid operation optimization model and the multi-dimensional micro-grid operation constraint condition based on a target optimization algorithm; optimizing scheduling of the port micro-grid according to the target scheduling strategy.

2. The method of claim 1, wherein, The step of determining predicted wind power generation according to meteorological information and historical wind power generation information associated with wind power generation comprises the following steps: adopting a target modal decomposition strategy to decompose historical wind power generation information to obtain a plurality of wind intrinsic modal function components; extracting features from meteorological information associated with wind power generation; combining each wind intrinsic modal function component with the extracted meteorological features to obtain current combination data; performing format conversion on the current combination data, and performing normalization processing on the converted current combination data; predicting wind power generation corresponding to each wind intrinsic modal function component according to the normalized current combination data through a long short-term memory model with an attention mechanism, wherein the long short-term memory model with the attention mechanism is obtained by optimizing the learning rate and the number of neurons of an original long short-term memory model through a whale optimization algorithm and an attention mechanism; superimposing each wind power generation to obtain predicted wind power generation.

3. The method of claim 1, wherein, The step of determining predicted load demand according to historical load data of the port comprises the following steps: adopting a target modal decomposition strategy to decompose historical load data of the port to obtain a plurality of load intrinsic modal function components; performing format conversion on each load intrinsic modal function component, and performing standardization processing on the converted load intrinsic modal function component to obtain a plurality of standardized load intrinsic modal function components; predicting a single load demand corresponding to each standardized load intrinsic modal function component through a long short-term memory model that learns long-term dependencies in time series data; superimposing each single load demand to obtain predicted load demand.

4. The method of claim 1, wherein, The step of constructing a micro-grid operation optimization model according to the predicted load demand, the predicted wind power generation and the predicted solar power generation comprises the following steps: determining a main grid power purchase cost according to a main grid power purchase price and a main grid power purchase power, and determining a main grid power sale cost according to a main grid power sale price and a main grid power sale power; determining a current operation and maintenance cost according to a proportionality coefficient of the operation and maintenance cost of each part and the output power of each part, and determining a current depreciation cost according to a purchase and installation cost of each part and a service life of each part; generating a micro-grid operation cost function in a minimum time period according to the main grid power purchase cost, the main grid power sale cost, the current operation cost and the current depreciation cost; determining a pollution treatment cost of the micro-grid purchasing power from the main grid according to a cost coefficient of a target type of pollution, an emission amount of the target type of pollution generated by the main grid operation and a current time micro-grid power purchase from the main grid; determining a micro-grid environmental cost function according to the pollution treatment cost; constructing a micro-grid operation optimization model according to the predicted load demand, the predicted wind power generation, the predicted solar power generation, the micro-grid operation cost function in the minimum time period and the micro-grid environmental cost function.

5. The method of claim 1, wherein, The step of constructing the multi-dimensional micro-grid operation constraint condition comprises: constructing a power balance constraint condition according to the photovoltaic system output power, the fan output power, the fuel cell output power, the hydrogen production equipment output power, the energy storage output power, the energy storage device power consumption when storing electricity, the main grid power parameter and the current load power; constructing a distributed power output constraint condition according to the minimum output value and the maximum output value of the distributed power supply; constructing an electric energy storage constraint condition according to the current discharge power and the current charge power of the energy storage system, the minimum discharge power and the maximum discharge power of the electric energy storage device, the minimum charge power and the maximum charge power of the electric energy storage device, the energy storage system state variable, the minimum energy storage amount and the maximum energy storage amount of the energy storage system; constructing a hydrogen storage tank constraint condition according to the current hydrogen discharge power and the current hydrogen charge power, the minimum hydrogen charge power, the maximum hydrogen charge power, the hydrogen storage system state variable, the minimum hydrogen storage amount and the maximum hydrogen storage amount; constructing a purchase and sale power constraint condition according to the power purchase power, the power sale power and the purchase and sale power rules; obtaining the multi-dimensional micro-grid operation constraint condition according to the power balance constraint condition, the distributed power output constraint condition, the electric energy storage constraint condition, the hydrogen storage tank constraint condition and the purchase and sale power constraint condition.

6. The method of any one of claims 1 to 5, wherein, The step of solving the target scheduling strategy based on the target optimization algorithm according to the grid operation optimization model and the multi-dimensional micro-grid operation constraint condition comprises: allocating a micro-grid scheduling strategy and a strategy adjustment parameter to each particle randomly based on the target optimization algorithm according to the grid operation optimization model and the multi-dimensional micro-grid operation constraint condition; calculating the fitness value of each particle after allocation respectively, and solving the individual optimal position and the global optimal position according to the fitness value; updating the micro-grid scheduling strategy and the strategy adjustment parameter of each particle according to the individual optimal position and the global optimal position; determining the updated optimal scheduling strategy as the target scheduling strategy when the preset termination condition is met.

7. A port micro-grid optimal scheduling device, characterized in that, comprises: a prediction module configured to determine a predicted wind power generation according to weather information and historical wind power generation information associated with wind power generation, and to determine a predicted solar power generation according to weather information and historical solar power generation information associated with solar power generation; The construction module is configured to determine a predicted load demand according to historical load data of the port, and construct a micro-grid operation optimization model according to the predicted load demand, the predicted wind power and the predicted solar power; The construction module is further configured to construct a multi-dimensional micro-grid operation constraint condition, and solve a target scheduling strategy according to the grid operation optimization model and the multi-dimensional micro-grid operation constraint condition based on a target optimization algorithm; The scheduling module is configured to perform optimized scheduling on the port micro-grid according to the target scheduling strategy.

8. An electronic device, comprising: The computer program product comprises: at least one memory for storing a computer program; at least one processor for executing the program stored in the memory, and when the program stored in the memory is executed, the processor is configured to execute the method according to any one of claims 1-6.

9. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 8. When the computer program runs on the processor, the processor is caused to execute the method according to any one of claims 1-6.

10. A computer program product, characterised in that, When the computer program product runs on the processor, the processor is caused to execute the method according to any one of claims 1-6.

Images