Micro-grid economic dispatching method and device

Abstract

The application discloses a micro-grid economic dispatching method and device, relates to the technical field of micro-grid operation control, and solves the technical problem that the stability of the whole micro-grid operation is reduced due to the fact that the existing micro-grid economic dispatching method usually adopts fixed time intervals to make scheduling plans; the method comprises the following steps: generating a corresponding predicted power curve according to predicted environment data; generating a scheduling time period based on the predicted power curve; constructing a target function in an economic dispatching model corresponding to the micro-grid based on the scheduling time period; constructing a plurality of constraint conditions in the economic dispatching model corresponding to the micro-grid based on operation constraint data; the constraint conditions comprise power constraint conditions and basic constraint conditions; solving the target function in the economic dispatching model to obtain an optimal scheduling scheme; and scheduling power resources in the micro-grid based on the scheduling scheme; so that the economic dispatching is more targeted, and the stability of the whole micro-grid system is improved.

Description

Technical Field

[0001] This application belongs to the field of microgrid operation and control technology, specifically a microgrid economic dispatch method and device. Background Technology

[0002] With the rapid development of new energy technologies, microgrids, as autonomous systems capable of self-control, protection, and management, integrate distributed power sources, energy storage devices, and loads, and have become an important component of smart grids. Economic dispatch is a core element in the operation and management of microgrids, aiming to minimize system operating costs by rationally allocating the output of distributed power sources while meeting load demands and operational constraints.

[0003] Existing microgrid economic dispatch methods typically employ fixed time intervals for dispatch planning. However, renewable energy generation, such as photovoltaic and wind power, is significantly affected by weather conditions, exhibiting strong volatility and intermittency. Fixed-time dispatch methods often fail to accurately match the actual fluctuation characteristics of renewable energy power generation: during periods of stable power, overly granular time divisions increase the computational burden; while during periods of severe power fluctuations, overly coarse time divisions lead to delayed dispatch strategies. This inability to respond promptly to power changes affects the accuracy and economy of dispatch decisions, resulting in decreased stability of the entire microgrid operation. Therefore, there is an urgent need for a microgrid economic dispatch scheme that can dynamically divide dispatch periods based on the characteristics of renewable energy generation and comprehensively consider equipment operating costs. Summary of the Invention

[0004] This application provides a microgrid economic dispatch method and apparatus, which solves the technical problem that the existing microgrid economic dispatch methods usually use fixed time intervals to formulate dispatch plans, which makes the dispatch results unable to respond to power changes in a timely manner, thereby affecting the accuracy and economy of dispatch decisions and leading to a decrease in the stability of the entire microgrid operation.

[0005] To achieve the above objectives, this application adopts the following technical solution: Firstly, a microgrid economic dispatch method is provided, including: The system acquires predicted environmental data corresponding to a set time for each power generation unit in the microgrid, the predicted environmental data including parameter curves corresponding to several environmental parameters; inputs the predicted environmental data into the power prediction model corresponding to the power generation unit to obtain the corresponding predicted power curve; generates several scheduling periods based on the several predicted power curves; the power prediction model is obtained through training an artificial intelligence model. The objective function in the economic dispatch model corresponding to the microgrid is constructed based on the dispatch period corresponding to each power generation unit; the power generation cost objective function includes the operating cost and switching cost corresponding to each power generation unit, as well as the energy storage cost, electricity purchase cost and electricity sales cost of the energy storage unit; Obtain the operational constraint data of the microgrid; construct several constraints in the economic dispatch model corresponding to the microgrid based on the operational constraint data; the economic dispatch model includes a power generation cost objective function and several constraints; the constraints include power constraints and basic constraints. The objective function in the economic dispatch model is solved to obtain the optimal dispatch scheme; the power resources in the microgrid are dispatched based on the dispatch scheme.

[0006] Based on the above technical solution, in the microgrid economic dispatch method provided in this application, the following steps are taken: First, predictive environmental data corresponding to a set time for each power generation unit in the microgrid is obtained. This predictive environmental data is then input into the power estimation model corresponding to the power generation unit to obtain the corresponding predicted power curve. Several dispatch periods are generated based on these predicted power curves. An objective function in the economic dispatch model corresponding to the microgrid is constructed based on the dispatch periods corresponding to each power generation unit. Second, operational constraint data of the microgrid is obtained. Third, several constraints in the economic dispatch model corresponding to the microgrid are constructed based on the operational constraint data. The economic dispatch model includes a power generation cost objective function and several constraints. These constraints include power constraints and basic constraints. The objective function in the economic dispatch model is solved to obtain the optimal dispatch scheme. Power resources in the microgrid are dispatched based on the dispatch scheme. By predicting the power of each power generation unit, several dispatch periods are dynamically divided into future dispatch periods. Corresponding power constraints are set in each dispatch period. Simultaneously, switching costs are introduced into the objective function, making the economic dispatch more targeted and thus improving the stability of the entire microgrid system.

[0007] Current dispatching models often focus on explicit costs such as fuel costs, neglecting implicit switching costs such as equipment wear and tear and lifespan reduction incurred during the start-up, shutdown, or output adjustment of power generation equipment. This can lead to increased maintenance costs and shortened equipment lifespan due to frequent switching of unit states during actual dispatching. This embodiment introduces equipment wear, start-up and shutdown energy consumption, and lifespan reduction caused by the switching of power generation unit states, avoiding frequent start-up and shutdown of units or significant adjustments to output in pursuit of small electricity price gains. This effectively extends equipment lifespan and improves system operational stability.

[0008] In conjunction with the first aspect above, in one possible implementation, one training method for the power prediction model includes: Acquire several historical environmental data sets and power variation curves of the power generation unit under the aforementioned environmental data; the environmental data includes parameter curves corresponding to several environmental parameters; extract the parameter values ​​at each moment of the parameter curves and the power values ​​at each moment of the power variation curves; integrate several parameter values ​​and power values ​​at the same moment into a set of training data or test data; thereby construct several sets of training data and test data. The artificial intelligence model is trained using training data and tested using test data to obtain an artificial intelligence model with several parameter values ​​as input and the power value of the corresponding power generation unit under the environmental data as output. The power values ​​at different times are integrated into a power rate change curve, and the power change curve is used as the output of the predicted power curve. Finally, a power prediction model is obtained, in which the input is a number of parameter change curves and the output is the predicted power curve. The artificial intelligence model includes deep neural network models, etc.

[0009] In conjunction with the first aspect above, in one possible implementation, generating several scheduling periods based on several predicted power curves includes: To obtain several predicted power curves, it is understood that before proceeding to this step, each predicted power curve needs to be truncated. Specifically, the overlapping portions of several power generation prediction regions on the time axis are truncated; the power output values ​​corresponding to each moment in each predicted power curve are extracted, and several power output values ​​at the same moment are integrated into a stability assessment set; based on several power output values ​​within the stability assessment set, a stability assessment value and output power corresponding to the stability assessment set are generated; and several scheduling periods are generated based on the stability assessment values ​​and output power corresponding to each stability assessment set.

[0010] In conjunction with the first aspect above, in one possible implementation, generating a stability assessment value and output power corresponding to the stability assessment set based on several power output values ​​within the stability assessment set includes: Each power output value is obtained and fitted into a power change curve according to the order of the termination time of the corresponding predicted environmental data and the predicted power curve. When the power change curve converges, the stability evaluation value of the curve is set to 1, and the convergence value of the power change curve is recorded as the output power at the time corresponding to the stability evaluation set. When the power change curve at a certain time converges, it means that as time approaches, the predicted power value at that time is closer to a fixed value, and the prediction result at that time is more stable. Then the stability score corresponding to that time is set to the highest, which is "1". When the power change curve does not converge, the power change curve is substituted into a set stability evaluation function to obtain the stability evaluation value corresponding to the power change curve, and the output power at the time corresponding to the stability evaluation set; one expression of the stability evaluation function includes: ; ; in, for The stability score corresponding to each moment. The curve shows the power change. The set constant power value; This is the starting point corresponding to the power change curve. The endpoint corresponding to the power change curve; for The output power at any given moment.

[0011] In conjunction with the first aspect above, in one possible implementation, generating several scheduling periods based on the stability assessment values ​​and output power corresponding to each stability assessment set includes: S1: Obtain several stability evaluation values. When the stability evaluation value is greater than the set stability threshold, record the time corresponding to the stability evaluation set as a stable point. S2: The time interval between adjacent stable points is recorded as a stable time interval; and the stable time intervals are numbered according to their chronological order. S3: Select the stable time period with the smallest number as the candidate scheduling time period; S4: Obtain the duration of the candidate scheduling period and the output power corresponding to the two stable points; S5: Determine if the difference between the output power of the two stable points is less than the set power difference threshold. If yes, proceed to S6; otherwise, proceed to S9. S6: Obtain the candidate scheduling time periods corresponding to the two stable points, and determine whether the duration of the candidate scheduling time period is greater than the set fluctuation limit time period threshold; if yes, proceed to S7; if no, proceed to S8. S7: Set the candidate scheduling period as the scheduling period, and record the average output power of the two stable points as the output power of the scheduling period; proceed to S11; S8: Obtain the output power corresponding to several moments within the candidate scheduling period, calculate the variance corresponding to the output power, and determine whether the variance is less than the set variance threshold. If yes, proceed to S7; otherwise, proceed to S9. S9: Select several output powers within the selected scheduling period, fit the output powers into a power change curve for the period, calculate the point with the largest rate of change in the power change curve for the period as the cutoff point, and proceed to S10. S10: Use stage points to split the candidate scheduling period into two stable periods, and renumber the stable periods; proceed to S2; S11: Obtain the existing scheduling period and determine whether the total duration of the scheduling period is equal to the total scheduling duration; if yes, proceed to S12; otherwise, renumber the stable period and proceed to S2. S12: Output all scheduling periods and the corresponding output power for each scheduling period.

[0012] In conjunction with the first aspect above, in one possible implementation, the objective function in the economic dispatch model corresponding to the microgrid, constructed based on the dispatch time periods corresponding to each power generation unit, includes: Obtain the operating cost function and switching cost function corresponding to each power generation unit; obtain the energy storage cost function, electricity purchase cost function, and electricity sales cost function corresponding to each energy storage unit; construct the objective function of the economic dispatch model based on the operating cost function, switching cost function, energy storage cost function, electricity purchase cost function, and electricity sales cost function, wherein the objective function is: ; in, Set the total operating cost of the microgrid over a future time period. Let $\frac{m}{n}$ be the operating cost of the $n$-th power generation unit during the $m$-th scheduling period. This represents the operation switching cost caused by the switching of the operating status of the nth power generation unit from the (m-1)th scheduling period to the mth scheduling period. If the operating status does not change from the (m-1)th scheduling period to the mth scheduling period, the corresponding operation switching cost is 0. Let $ be the energy storage cost of the k-th energy storage device. For electricity purchase costs; This is the cost of selling electricity.

[0013] In conjunction with the first aspect above, in one possible implementation, one method for obtaining the switching cost includes: Obtain the first operating status code of the nth power generation unit during the (m-1)th scheduling period, and the second operating status code of the nth power generation unit during the mth scheduling period; construct a state switching code based on the first and second operating status surfaces, and obtain the corresponding switching cost by querying the switching cost lookup table corresponding to the power generation unit based on the state switching code; the switching cost lookup table includes query items and switching cost items; the query items include several state switching codes; the switching cost items include several switching costs corresponding to the state switching codes.

[0014] In conjunction with the first aspect mentioned above, one possible implementation method based on power constraints includes: Based on the output power of each power generation unit, the energy storage power and output power of each energy storage unit in the operational constraint data, as well as the purchased power, sold power, and load demand power, power balance constraints are constructed; the power balance constraints are as follows: ; in, The output power of the nth power generation unit must satisfy its corresponding output power constraint condition. This represents the output power corresponding to the k-th energy storage unit. Let k be the energy storage power corresponding to the k-th energy storage unit. For the power purchased by the microgrid, This refers to the electricity sales capacity of the microgrid. The load demand power of the microgrid; The power constraints include power balance constraints and output power constraints corresponding to each power generation unit.

[0015] In conjunction with the first aspect above, in one possible implementation, the output power constraint condition can be constructed in the following ways: Obtain the output power of each power generation unit for each scheduling period, and construct output power constraints for the corresponding power generation unit based on the output power; one expression of the output power constraints for the power generation unit includes: ; in, This represents the output power corresponding to the nth power generation unit. This represents the maximum output power corresponding to the nth power generation unit; one way to express the maximum output power includes: ; in, This represents the output power of the nth power generation unit during the mth scheduling period. By setting specific power constraints for each power generation unit during different constant power periods, the power constraints become more targeted, enhancing the accuracy of long-term economic scheduling.

[0016] In conjunction with the first aspect above, in one possible implementation, the other constraints include power generation unit ramp-up constraints and energy storage unit constraints, etc. The ramp-up constraint condition for the power generation unit is: ; in, This represents the output power of the nth power generation unit during the mth scheduling period. This represents the output power of the nth power generation unit during the (m-1)th scheduling period. For the maximum gradeability, For switching time; The constraints of the energy storage unit are: ; ; ; ; The first equation represents the dynamic constraint condition for the state of charge of the energy storage unit; where: Let be the state of charge of the k-th energy storage unit at time t; Let be the charging efficiency of the k-th energy storage unit; Let be the discharge efficiency of the k-th energy storage unit; Let be the charging power of the k-th energy storage unit at time t; Let be the discharge power of the k-th energy storage unit at time t; For scheduling time; The second equation represents the uplink and downlink constraints for the energy storage unit; where: This represents the minimum permissible state of charge for the energy storage unit. This represents the maximum permissible state of charge of the energy storage unit. The third equation represents the charging and discharging power constraint condition for the energy storage unit; where: This represents the upper limit of the energy storage power of the k-th energy storage unit; This represents the upper limit of the output power of the k-th energy storage unit; The fourth equation is the mutual exclusion constraint condition for charging and discharging of the energy storage unit.

[0017] Secondly, a microgrid economic dispatch device is provided, comprising: a data acquisition module, a power prediction module, a data processing module, and a dispatch module; The data acquisition module is used to acquire the predicted environmental data corresponding to the set time of each power generation unit in the microgrid, as well as the operating constraint data of the microgrid. The power prediction module is used to input the predicted environmental data into the power prediction model corresponding to the power generation unit to obtain the corresponding predicted power curve. It can be understood that each power generation unit has its own corresponding power prediction model. Multiple lightweight models are used instead of a powerful copy model, which saves data processing time. The data processing module includes an economic scheduling model construction unit and an economic scheduling model solving unit; The economic dispatch model construction unit is used to construct the objective function in the economic dispatch model corresponding to the microgrid based on the dispatch period corresponding to each power generation unit; and to construct several constraint conditions in the economic dispatch model corresponding to the microgrid based on the operational constraint data; the constraint conditions include power constraint conditions and basic constraint conditions. The economic scheduling model solving unit is used to solve the objective function in the economic scheduling model to obtain the optimal scheduling scheme; The scheduling module is used to schedule the power resources in the microgrid based on the scheduling scheme.

[0018] This application provides a microgrid economic dispatch method and apparatus, which can acquire predicted environmental data corresponding to a set time for each power generation unit in the microgrid; input the predicted environmental data into the power prediction model corresponding to the power generation unit to obtain the corresponding predicted power curve; generate several dispatch periods based on several predicted power curves; construct the objective function in the economic dispatch model corresponding to the microgrid based on the dispatch periods corresponding to each power generation unit; acquire the microgrid's operating constraint data; construct several constraints in the economic dispatch model corresponding to the microgrid based on the operating constraint data; the economic dispatch model includes a power generation cost objective function and several constraints; the constraints include power constraints and basic constraints; solve the objective function in the economic dispatch model to obtain the optimal dispatch scheme; dispatch the power resources in the microgrid based on the dispatch scheme; by predicting the power of each power generation unit, the future dispatch period is dynamically divided into several dispatch periods, and corresponding power constraints are set in each dispatch period. At the same time, switching costs are introduced into the objective function, making the economic dispatch more targeted, thereby improving the stability of the entire microgrid system.

[0019] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram illustrating the steps of the microgrid economic dispatch method in this application; Figure 2 This is a flowchart illustrating the scheduling time slot division process in this application; Figure 3 This is a schematic diagram of the module connections of the microgrid economic dispatch system in this application. Detailed Implementation

[0022] The technical solutions of this application will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0023] Please see Figure 1 The first aspect of this application provides a microgrid economic dispatch method, including: The process involves acquiring predicted environmental data for each power generation unit in a microgrid at a set time. This predicted environmental data includes parameter curves corresponding to several environmental parameters. Specifically, the power generation units in a microgrid typically include renewable energy equipment such as photovoltaic (PV) power generation systems and wind power generation systems. For PV power generation units, the predicted environmental data mainly refers to irradiance curves and ambient temperature curves; for wind power generation units, it mainly refers to wind speed curves and wind direction data. This data is usually collected in real time by meteorological monitoring sensors deployed within the microgrid, or obtained from an external meteorological service platform via a communication interface for a future set time period, such as predicted data for the next 24 hours. It should be understood that the form of the predicted environmental data is not limited to continuous curves; it can also be discrete time-series data points, as long as it reflects the trend of environmental parameters changing over time. This data forms the basis for subsequent power prediction, and its accuracy directly affects the reliability of scheduling decisions.

[0024] The predicted environmental data is input into the power prediction model corresponding to the power generation unit to obtain the corresponding predicted power curve. The power prediction model is trained using an artificial intelligence model; specifically, each power generation unit is equipped with a corresponding power prediction model. This model is a mathematical model trained using historical environmental data and historical power data through machine learning algorithms such as deep neural networks and support vector machines. The model learns the nonlinear mapping relationship between environmental parameters and power generation. When predicted environmental data is input, the model can output the predicted power values ​​of the power generation unit at various future times, thus forming the predicted power curve. Compared to traditional physical modeling methods, the artificial intelligence-based model can better capture the power change characteristics under complex environments and improve prediction accuracy. This step realizes the transformation from environmental factors to power generation capacity, providing data support for subsequent time period division.

[0025] Several scheduling periods are generated based on several predicted power curves. Specifically, traditional scheduling methods typically divide a day into 24 or 48 fixed periods, ignoring the fluctuations in renewable energy generation power across different time periods. This embodiment dynamically divides future time into several scheduling periods based on the fluctuation characteristics of the predicted power curves. For example, when the predicted power curve is very stable over a certain period, it may be divided into a longer scheduling period; while when the predicted power curve fluctuates sharply over a certain period, it may be divided into multiple shorter scheduling periods. This dynamic division mechanism ensures that the scheduling periods match the actual power generation characteristics, laying the foundation for subsequently building an accurate economic scheduling model.

[0026] The objective function in the economic dispatch model corresponding to the microgrid is constructed based on the dispatch period corresponding to each power generation unit; the power generation cost objective function includes the operating cost and switching cost corresponding to each power generation unit, as well as the energy storage cost, electricity purchase cost and electricity sales cost of the energy storage unit; Obtain the operational constraint data of the microgrid; construct several constraints in the economic dispatch model corresponding to the microgrid based on the operational constraint data; the economic dispatch model includes a power generation cost objective function and several constraints; the constraints include power constraints and basic constraints. Specifically, the construction of an economic dispatch model comprises two core components: an objective function and constraints. The objective function typically aims to minimize operating costs, and its construction depends on the dispatch period, as parameters such as generation costs and electricity purchase prices may differ across time periods. Operational constraints include power limits between the microgrid and the main grid, the state-of-charge boundary of the energy storage system, and line transmission capacity limitations. Constraints ensure the safety and feasibility of the dispatch scheme during actual execution; for example, power balance constraints ensure matching generation with consumption, and ramp constraints ensure that the rate of change in unit output remains within permissible limits. By incorporating dynamically generated dispatch periods into the model construction process, the optimization problem becomes more closely aligned with actual operational scenarios.

[0027] The objective function in the economic dispatch model is solved to obtain the optimal dispatch scheme; based on the dispatch scheme, the power resources in the microgrid are dispatched. Specifically, after constructing the mathematical model containing the objective function and constraints, an optimization algorithm is used to solve it. The solution process involves finding the combination of decision variables that minimizes the objective function while satisfying all constraints, such as the output plans of each generating unit and the charging and discharging plans of the energy storage system. The final output dispatch scheme is sent as a control command to the microgrid control system, which then adjusts the inverters, energy storage converters, and other equipment of each generating unit to achieve optimal allocation of power resources.

[0028] By acquiring the predicted environmental data corresponding to the set time of each power generation unit in the microgrid; inputting the predicted environmental data into the power prediction model corresponding to the power generation unit to obtain the corresponding predicted power curve; generating several scheduling periods based on several predicted power curves; constructing the objective function in the economic dispatch model corresponding to the microgrid based on the scheduling periods corresponding to each power generation unit; acquiring the operating constraint data of the microgrid; constructing several constraint conditions in the economic dispatch model corresponding to the microgrid based on the operating constraint data; the economic dispatch model includes a power generation cost objective function and several constraint conditions; the constraint conditions include power constraint conditions and basic constraint conditions; solving the objective function in the economic dispatch model to obtain the optimal dispatch scheme; dispatching the power resources in the microgrid based on the dispatch scheme; by predicting the power of each power generation unit, dynamically dividing the future dispatch period into several scheduling periods, setting corresponding power constraint conditions in each scheduling period, and introducing switching costs into the objective function, the economic dispatch becomes more targeted, thereby improving the stability of the entire microgrid system. This embodiment realizes a complete closed loop from environmental data perception, power prediction, dynamic time period division to optimization solution, effectively solving the problem of scheduling strategy lag or frequent fluctuation caused by fixed time period division, and significantly improving the economy and stability of microgrid operation.

[0029] In one possible implementation, a training method for the power prediction model includes: acquiring several historical environmental data and power change curves of power generation units under the environmental data; the environmental data includes parameter curves corresponding to several environmental parameters; extracting parameter values ​​at each time point of the parameter curves and power values ​​at each time point of the power change curves; integrating several parameter values ​​and power values ​​at the same time point into a set of training data or test data; thereby constructing several training data and test data. Specifically, historical environmental data typically originates from the historical database stored in the microgrid monitoring system, covering records of changes in parameters such as sunlight intensity, ambient temperature, wind speed, and wind direction over the past few months or even years. The corresponding power variation curves represent the actual output power of the power generation units under these historical environmental conditions. It should be understood that, to improve the model's generalization ability, the selection of historical data should cover as many operating conditions as possible, including seasonal changes and different weather types such as sunny, cloudy, and rainy days, to avoid overfitting due to limited training data. The artificial intelligence model is trained using training data and tested using test data to obtain an artificial intelligence model with several parameter values ​​as input and the power value of the corresponding power generation unit under the environmental data as output. The power values ​​at different times are integrated into a power rate change curve, and the power change curve is used as the output of the predicted power curve. Finally, a power prediction model is obtained, in which the input is a number of parameter change curves and the output is the predicted power curve. The artificial intelligence model includes deep neural network models, etc. The training of this lightweight prediction model is a relatively existing technology, only the training data is different, so it will not be described in detail here.

[0030] In one possible implementation, several scheduling periods are generated based on several predicted power curves. This includes: acquiring several predicted power curves; it is understood that before this step, each predicted power curve needs to be truncated, specifically, the overlapping parts of several power generation prediction regions on the time axis are truncated; extracting the power output values ​​corresponding to each moment in each predicted power curve, and integrating several power output values ​​at the same moment into a stability assessment set; it should be understood that microgrids typically contain various types of power generation units such as photovoltaics and wind power, and the predicted power of different power generation units at the same moment may differ, and even the results of the same power generation unit may differ when environmental data changes; integrating these power output values ​​at the same moment into a set can comprehensively reflect the degree of certainty of the overall power generation capacity of the microgrid at that moment. If the numerical differences within the set are small, it indicates good prediction consistency and high system stability; if the differences are large, it indicates high uncertainty. Based on several power output values ​​within the stability assessment set, a stability assessment value and output power corresponding to the stability assessment set are generated; several scheduling periods are generated based on the stability assessment values ​​and output power corresponding to each stability assessment set.

[0031] In one possible implementation, a stability assessment value and output power corresponding to the stability assessment set are generated based on several power output values ​​within the stability assessment set, including: Each power output value is obtained and fitted into a power change curve according to the order in which the predicted environmental data for each predicted power curve terminates. It is understood that environmental conditions at a future time may change at different times, leading to different power prediction results when using corresponding environmental data to predict power at a future time. As the time of prediction approaches the predicted time, the environmental prediction data becomes more accurate, resulting in more accurate power prediction. The power change curve here represents the change in the predicted power value at a future time as of the time of prediction. Theoretically, this power change curve should be a convergent curve, and the convergent predicted power value should be the same as or very close to the actual value at that future time. However, in cases of complex environmental changes, the accuracy of predicting future environmental conditions may decrease, potentially causing the power change curve to diverge. This embodiment avoids the impact of insufficient accuracy of a single prediction result on microgrid scheduling within a set future time period by repeatedly predicting power in the near future, analyzing the predicted power values ​​at each future time, and then constructing the objective function and performing scheduling. When the power change curve converges, the stability evaluation value of the curve is set to 1, and the convergence value of the power change curve is recorded as the output power at the time corresponding to the stability evaluation set. When the power change curve at a certain time converges, it means that as time approaches, the predicted power value at that time is closer to a fixed value, and the prediction result at that time is more stable. Then the stability score corresponding to that time is set to the highest, which is "1". When the power change curve does not converge, the power change curve is substituted into a set stability evaluation function to obtain the stability evaluation value corresponding to the power change curve, and the output power at the time corresponding to the stability evaluation set; one expression of the stability evaluation function includes: ; ; in, for The stability score corresponding to each moment. The curve shows the power change. The set constant power value; This is the starting point corresponding to the power change curve. The endpoint corresponding to the power change curve; for The output power corresponding to each moment; the stability score is set between 0 and 1. When the stability score is closer to 1, it means that the power of multiple predictions is almost the same and the results of multiple predictions are stable.

[0032] This embodiment obtains the stability evaluation value corresponding to each power change curve through the aforementioned stability evaluation function. The larger the difference between the power change curve and the closest constant power line, the more severe the fluctuation of the power change curve, the lower the reliability of the prediction result, and the lower the corresponding stability score is set. The closest constant power line refers to the line whose integral of the difference between the power change curve and the power change curve is minimized. By finding a constant power line that minimizes the integral of the deviation between the power change curve and this constant power line, a stability score is obtained through exponential function mapping. The advantage of this design is that it can effectively identify high-confidence prediction moments, reduce the impact of uncertainty on scheduling, and provide a quantitative basis for subsequent time period division.

[0033] Please see Figure 2 In conjunction with the first aspect above, in one possible implementation, generating several scheduling periods based on the stability assessment values ​​and output power corresponding to each stability assessment set includes: S1: Obtain several stability assessment values. When the stability assessment value is greater than the set stability threshold, the time corresponding to the stability assessment set is recorded as a stable point. The stable point is the basic anchor point for dividing the scheduling period. Only the time with sufficient predicted stability will be selected as the boundary of the period. The stability threshold can be flexibly set according to the microgrid's security requirements. The specific value is set according to expert experience. S2: The time interval between adjacent stable points is recorded as a stable time interval; and the stable time intervals are numbered according to their chronological order; at this point, the time axis is initially divided into several segments; S3: Select the stable time period with the smallest number as the candidate scheduling time period; the system starts from the time start point and reviews each time period one by one to see if it is suitable as an independent scheduling cycle; S4: Obtain the duration of the candidate scheduling period and the output power corresponding to the two stable points; S5: Determine if the difference between the output power of the two stable points is less than the set power difference threshold. If yes, proceed to S6; otherwise, proceed to S9. This step determines the power fluctuation range within the time period. If the power difference between the two points is too large, it indicates that there is a significant power increase or decrease trend within the time period, making it unsuitable as a constant output scheduling period. It needs to be split into S9 for further processing. S6: Obtain candidate scheduling time periods corresponding to the two stable points, and determine whether the duration of the candidate scheduling time period is greater than the set fluctuation limit time period threshold; if yes, proceed to S7; if no, proceed to S8; if the duration between the two stable points is short, even if there is power fluctuation within this time period, the amplitude of the fluctuation will be limited due to the short duration of the time period; the specific value of the fluctuation limit time period threshold is set according to expert experience. If the duration between the two stable points is long, there may be large power fluctuations within this time period, and it is impossible to guarantee that the power is constant within this time period, so further judgment is required; S7: Set the candidate scheduling period as the scheduling period, and record the average output power of the two stable points as the output power of the scheduling period; proceed to S11; if the fluctuation in the long period is small, it can be used as a period; otherwise, the long period needs to be divided into several small periods with approximately constant power. S8: Obtain the output power corresponding to several moments within the candidate scheduling period, calculate the variance corresponding to the output power, and determine whether the variance is less than the set variance threshold. If yes, proceed to S7; otherwise, proceed to S9. S9: Select several output powers within the selected scheduling period, fit the output powers into a power change curve for the period, and calculate the point with the largest rate of change in the power change curve for the period as the cutoff point, then proceed to S10; when splitting is required, selecting the point with the largest rate of change as the cutoff point can ensure that the cutting is carried out at the position where the power trend changes most drastically, so that the split sub-periods have better internal consistency; this facilitates the rapid determination of small periods with approximately constant power in the future. S10: Use stage points to split the candidate scheduling period into two stable periods, and renumber the stable periods; proceed to S2; S11: Obtain the existing scheduling period and determine whether the total duration of the scheduling period is equal to the total scheduling duration; if yes, proceed to S12; otherwise, renumber the stable period and proceed to S2. S12: Output all scheduling periods and the corresponding output power for each scheduling period.

[0034] This embodiment constructs a "multi-level judgment and dynamic splitting" time period generation mechanism; it not only considers the endpoint differences of power, but also examines the time period length and internal variance in depth, effectively avoiding scheduling strategy lag caused by erroneous merging of time periods with drastic fluctuations; this adaptive partitioning method based on data characteristics effectively avoids scheduling strategy lag caused by erroneous merging of time periods with drastic fluctuations; it provides the optimal time granularity for the subsequent construction of the economic scheduling model, ensures the stability of power within the scheduling time period, and thus improves the accuracy and solution efficiency of the economic scheduling model.

[0035] In one possible implementation, the objective function in the economic dispatch model corresponding to the microgrid is constructed based on the dispatch period corresponding to each power generation unit, including: Obtain the operating cost function and switching cost function corresponding to each power generation unit; obtain the energy storage cost function, electricity purchase cost function, and electricity sales cost function corresponding to each energy storage unit; construct the objective function of the economic dispatch model based on the operating cost function, switching cost function, energy storage cost function, electricity purchase cost function, and electricity sales cost function, wherein the objective function is: ; in, Setting the total operating cost of the microgrid over a future time period represents the objective direction of the optimization problem, namely, pursuing the optimal economic operation of the system. The operating cost of the nth generation unit during the mth scheduling period is usually related to fuel consumption or maintenance costs. This is the operating switching cost caused by the switching of the operating status of the nth power generation unit from the (m-1)th scheduling period to the mth scheduling period. If the operating status does not change from the (m-1)th scheduling period to the mth scheduling period, the corresponding operating switching cost is 0. The introduction of this cost is to quantify the equipment wear and life loss caused by the start-up, shutdown or output adjustment of the unit, and to avoid the scheduling strategy from frequently changing the unit status in pursuit of small electricity price benefits. The energy storage cost of the k-th energy storage device covers losses and maintenance costs during the energy storage process. For electricity purchase costs; The electricity sales cost reflects the economic efficiency of the interaction between the microgrid and the main grid. Through the above construction, the objective function comprehensively covers various economic factors in the microgrid operation process, ensuring the comprehensiveness of the optimization results. It can be understood that the scheduling period corresponding to the fuel power generation unit is obtained by acquiring the time points at both ends of several scheduling periods corresponding to each power generation unit, sorting the time points in chronological order, and recording the time period between adjacent time points as the scheduling period of the fuel power generation unit. It can also be understood that the economic scheduling scheme in this embodiment is an economic scheduling scheme for a future set time period. For example, based on the environmental prediction data for the next hour from each time point in the previous hour, the economic scheduling scheme is the microgrid power scheduling scheme for the corresponding time period in the next hour. Within the next hour, each power generation unit has its own scheduling period division standard. This scheduling scheme is a unified scheduling of equipment within the microgrid within the next hour.

[0036] In one possible implementation, one method for obtaining switching costs includes: obtaining the first operating state code of the nth power generation unit during the (m-1)th scheduling period, and the second operating state code of the nth power generation unit during the mth scheduling period; constructing a state switching code based on the first and second operating states; and retrieving the corresponding switching cost from a switching cost lookup table corresponding to the power generation unit based on the state switching code; the switching cost lookup table includes query items and switching cost items; the query items include several state switching codes; and the switching cost items include several switching costs corresponding to the state switching codes. Specifically, the operating state code is a digital abstraction of the operating state of the power generation unit, for example, "0" can represent the shutdown state, "1" can represent the rated power operating state, and "2" can represent the low power operating state, etc. The state switching code is composed of the state codes of two consecutive time periods, such as "01" representing switching from the shutdown state to the rated operating state. The switching cost lookup table is a database pre-established according to the equipment characteristics, recording the cost values ​​corresponding to various switching states. During the model solving process, the switching cost can be obtained through a simple table lookup operation, avoiding complex real-time calculations and significantly improving the solution efficiency of the optimization algorithm, which is especially important for microgrid dispatching systems that require rapid response.

[0037] Traditional economic dispatch models often only consider explicit costs such as operating costs and fuel costs, neglecting the implicit costs incurred during the switching of power generation unit operating states. This embodiment effectively avoids the problem of dispatch strategies frequently changing unit operating states in pursuit of small electricity price gains by quantifying switching costs, thereby extending equipment lifespan and improving system operational stability. Specifically, switching costs mainly include the following components: First, equipment wear and tear costs; during the start-up, shutdown, or significant output adjustments of power generation units, mechanical components such as bearings, gearboxes, and turbine blades will be subjected to significant thermal and mechanical stresses, leading to accelerated wear. For example, each start-up of a diesel generator causes additional wear to components such as the engine block and piston rings, while frequent speed changes in wind turbine generators cause fatigue damage to the gearbox and bearings. This part of the cost can be quantified through equipment maintenance records, fault statistics, and equipment lifespan degradation models. Second, start-up and shutdown energy costs; power generation units require additional energy to start from a shutdown state and reach a stable operating state. For example, diesel generators require fuel for preheating and idling, while gas turbines require natural gas for ignition and acceleration. This energy consumption does not generate effective electrical output and is considered pure cost expenditure. Start-up and shutdown energy costs can be calculated using equipment start-up curves and fuel consumption rates. Third, lifespan depreciation costs; frequent state switching accelerates equipment aging and shortens its lifespan. For example, photovoltaic inverter power devices generate thermal cycling stress during frequent switching, leading to solder joint fatigue failure; energy storage batteries experience accelerated capacity decay during frequent charge-discharge switching. Lifespan depreciation costs can be estimated using a full life-cycle cost model, allocating the equipment purchase cost to each state switch. Fourth, ancillary service costs; some power generation units require additional ancillary service support during state switching. For example, large synchronous generators require excitation power for grid connection, and gas turbines require compressed air for startup; these ancillary services also incur corresponding costs.

[0038] To accurately quantify the aforementioned switching costs, this embodiment provides a method for constructing a switching cost lookup table. The construction of the switching cost lookup table needs to comprehensively consider factors such as the type of power generation unit, rated capacity, operating years, and maintenance status. The specific construction steps are as follows: First, determine the coding rules for the operating status of the power generation unit. Different types of power generation units have different operating status classification methods. For diesel generators, the operating status can be coded as follows: "0" indicates shutdown, "1" indicates idling, "2" indicates rated power operation, and "3" indicates partial load operation, such as 50% rated power. For photovoltaic power generation units, the operating status can be coded as follows: "0" indicates shutdown, such as at night or during maintenance, "1" indicates normal operation, and "2" indicates power-limited operation, such as power limitation under grid constraints. For wind turbine generators, the operating status can be coded as follows: "0" indicates shutdown, "1" indicates low wind speed operation, "2" indicates rated wind speed operation, and "3" indicates high wind speed power-limited operation. It should be understood that the granularity of the operational state division can be flexibly adjusted according to actual scheduling needs. The finer the state division, the more accurate the calculation of switching costs, but the model complexity will also increase accordingly. Secondly, a state switching code is constructed based on the operational state code. The state switching code is composed of the operational state codes of two consecutive scheduling periods, typically represented by two digits. For example, "00" indicates a switch from the shutdown state to the shutdown state (i.e., maintaining shutdown); "01" indicates a switch from the shutdown state to the normal operation state (i.e., starting); "10" indicates a switch from the normal operation state to the shutdown state (i.e., shutdown); and "12" indicates a switch from the normal operation state to the low-power operation state (i.e., load reduction).

[0039] The switching costs corresponding to each state switching code are determined through experimental testing or historical data analysis. For newly built microgrids or scenarios lacking historical data, estimations can be made using technical data such as start-stop characteristic curves and lifespan degradation models provided by equipment manufacturers. For example, the start-up cost of a certain model of diesel generator can be calculated as follows: starting fuel consumption (e.g., 5 liters of diesel fuel multiplied by the fuel price per unit), plus the time cost of the start-up process (e.g., 5 minutes multiplied by the unit time maintenance cost), plus the lifespan depreciation cost corresponding to a single start (e.g., total equipment cost divided by the designed number of start-stop cycles). For microgrids with existing operational data, historical maintenance records, fault records, and operation logs can be statistically analyzed to establish a statistical relationship model between the number of switching cycles and maintenance costs and failure rates, thereby inferring the actual cost of each state switching.

[0040] Finally, the aforementioned state switching codes and corresponding switching cost values ​​are entered into a lookup table to form a complete switching cost lookup table. An example of the lookup table structure is as follows: Query items include state switching codes such as "00", "01", "10", "11", "12", and "21"; the switching cost item includes the corresponding cost value, in yuan / time or yuan / kW. For example, for a diesel generator, the switching cost corresponding to "01" starting might be 500 yuan, the switching cost corresponding to "10" stopping might be 100 yuan, and the switching cost corresponding to "12" load reduction might be 50 yuan. In practical applications, the switching cost lookup table can be dynamically updated based on the equipment's operating years and maintenance status. For example, for older equipment with a longer operating history, its switching cost can be appropriately increased to reflect the actual situation of increased equipment wear and tear and increased failure risk. The switching cost lookup table can also be adjusted according to seasonal factors. For example, in low-temperature winter environments, the starting cost of a diesel generator may be higher than in summer because low-temperature starting requires longer preheating time and more fuel consumption.

[0041] In another embodiment, the aforementioned switching costs can also be constrained.

[0042] In one possible implementation, a method for constructing power balance constraints includes: constructing power balance constraints based on the output power of each power generation unit, the energy storage power and output power of each energy storage unit in the operational constraint data, as well as the purchased power, sold power, and load demand power; the power balance constraints are: ; in, The output power of the nth power generation unit must satisfy its corresponding output power constraint condition. This represents the output power corresponding to the k-th energy storage unit. Let k be the energy storage power corresponding to the k-th energy storage unit. For the power purchased by the microgrid, This refers to the electricity sales capacity of the microgrid. This represents the load demand power of the microgrid; this equation constraint is the cornerstone of the safe operation of the microgrid, ensuring that at any given time the sum of power generation, discharge, and purchase equals the sum of load, charging, and sales, thus maintaining the power supply and demand balance of the system. The power constraints include power balance constraints and output power constraints corresponding to each power generation unit.

[0043] In one possible implementation, one way to construct the various output power constraints includes: Obtain the output power of each power generation unit for each scheduling period, and construct output power constraints for the corresponding power generation unit based on the output power; one expression of the output power constraints for the power generation unit includes: ; in, This represents the output power corresponding to the nth power generation unit. This represents the maximum output power corresponding to the nth power generation unit; one way to express the maximum output power includes: ; in, This represents the output power of the nth power generation unit in the mth scheduling period. It should be understood that during periods of high volatility, the scheduling period is divided into finer segments, and the constraint update frequency for the maximum output power is higher, enabling timely capture of power changes. In contrast, during stable periods, the scheduling period is longer, and the constraints are relatively stable. This dynamic coupling mechanism between constraints and scheduling periods allows the mathematical model to accurately adapt to the volatile characteristics of new energy power generation, avoiding model distortion caused by fixed constraints, thereby maximizing the economic operation potential of the microgrid while ensuring safety.

[0044] In conjunction with the first aspect above, in one possible implementation, the other constraints include power generation unit ramp-up constraints and energy storage unit constraints, etc. The ramp-up constraint condition for the power generation unit is: ; in, This represents the output power of the nth power generation unit during the mth scheduling period. This represents the output power of the nth power generation unit during the (m-1)th scheduling period. For the maximum gradeability, For switching time; The constraints of the energy storage unit are: ; ; ; ; The first equation represents the dynamic constraint condition for the state of charge of the energy storage unit; where: Let be the state of charge of the k-th energy storage unit at time t; Let be the charging efficiency of the k-th energy storage unit; Let be the discharge efficiency of the k-th energy storage unit; Let be the charging power of the k-th energy storage unit at time t; Let be the discharge power of the k-th energy storage unit at time t; For scheduling time; The second equation represents the uplink and downlink constraints for the energy storage unit; where: This represents the minimum permissible state of charge for the energy storage unit. This represents the maximum permissible state of charge of the energy storage unit. The third equation represents the charging and discharging power constraint condition for the energy storage unit; where: This represents the upper limit of the energy storage power of the k-th energy storage unit; This represents the upper limit of the output power of the k-th energy storage unit; The fourth equation is the mutual exclusion constraint condition for charging and discharging of the energy storage unit.

[0045] Please see Figure 3 Secondly, a microgrid economic dispatch device is provided, comprising: a data acquisition module, a power prediction module, a data processing module, and a dispatch module; The data acquisition module is used to acquire the predicted environmental data corresponding to the set time of each power generation unit in the microgrid, as well as the operating constraint data of the microgrid; The power prediction module is used to input the predicted environmental data into the power prediction model corresponding to the power generation unit to obtain the corresponding predicted power curve. It can be understood that each power generation unit has its own corresponding power prediction model. Multiple lightweight models are used instead of a powerful copy model, which saves data processing time. The data processing module includes an economic scheduling model construction unit and an economic scheduling model solution unit; The economic dispatch model construction unit is used to construct the objective function in the economic dispatch model corresponding to the microgrid based on the dispatch period corresponding to each power generation unit; and to construct several constraint conditions in the economic dispatch model corresponding to the microgrid based on the operational constraint data; the constraint conditions include power constraint conditions and basic constraint conditions. The economic scheduling model solving unit is used to solve the objective function in the economic scheduling model to obtain the optimal scheduling scheme; The scheduling module is used to schedule the power resources in the microgrid based on the scheduling scheme.

[0046] Some of the data in the above formula are calculated by removing dimensions and taking their numerical values. The formula is the closest to the real situation obtained by software simulation of a large amount of collected data. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0047] The above embodiments are only used to illustrate the technical methods of this application and are not intended to limit it. Although this application has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of this application without departing from the spirit and scope of the technical methods of this application.

Claims

1. A microgrid economic dispatch method, characterized in that, include: The system acquires predicted environmental data corresponding to a set time for each power generation unit in the microgrid, the predicted environmental data including parameter curves corresponding to several environmental parameters; inputs the predicted environmental data into the power prediction model corresponding to the power generation unit to obtain the corresponding predicted power curve; generates several scheduling periods based on the several predicted power curves; the power prediction model is obtained through training an artificial intelligence model. The objective function in the economic dispatch model corresponding to the microgrid is constructed based on the dispatch period corresponding to each power generation unit. Obtain the operational constraint data of the microgrid; construct several constraint conditions in the economic dispatch model corresponding to the microgrid based on the operational constraint data; the constraint conditions include power constraint conditions and basic constraint conditions; The objective function in the economic dispatch model is solved to obtain the optimal dispatch scheme; the power resources in the microgrid are dispatched based on the dispatch scheme.

2. The microgrid economic dispatch method according to claim 1, characterized in that, One training method for the power prediction model includes: Acquire several historical environmental data and the power change curve of the power generation unit under the environmental data; the environmental data includes parameter curves corresponding to several environmental parameters; extract the parameter values ​​of the parameter curves at each time point and the power values ​​of the power change curves at each time point; integrate several parameter values ​​and power values ​​at the same time point into a set of training data or test data; thereby construct several training data and test data. The artificial intelligence model is trained using training data and tested using test data to obtain an artificial intelligence model with several parameter values ​​as input and the power value of the corresponding power generation unit under the environmental data as output. The power values ​​at different times are integrated into a power rate change curve, and the power change curve is used as the output of the predicted power curve. Finally, a power prediction model is obtained with several parameter change curves as input and the predicted power curve as output.

3. The microgrid economic dispatch method according to claim 1, characterized in that, The generation of several scheduling periods based on several predicted power curves includes: Obtain several predicted power curves; extract the power output values ​​corresponding to each time point from each predicted power curve, and integrate several power output values ​​at the same time point into a stability evaluation set; generate the stability evaluation value and output power corresponding to the stability evaluation set based on several power output values ​​within the stability evaluation set; generate several scheduling periods based on the stability evaluation value and output power corresponding to each stability evaluation set.

4. The microgrid economic dispatch method according to claim 3, characterized in that, Based on several power output values ​​within the stability assessment set, the stability assessment value and output power corresponding to the stability assessment set are generated, including: Each power output value is obtained and fitted into a power change curve according to the order of the termination time of the corresponding predicted environmental data and the predicted power curve. When the power change curve converges, the stability evaluation value of the curve is set to 1, and the convergence value of the power change curve is recorded as the output power at the time corresponding to the stability evaluation set. When the power change curve does not converge, the power change curve is substituted into the set stability evaluation function to obtain the stability evaluation value corresponding to the power change curve, and the output power at the time corresponding to the stability evaluation set.

5. The microgrid economic dispatch method according to claim 4, characterized in that, The generation of several scheduling periods based on the stability assessment values ​​and output power corresponding to each stability assessment set includes: S1: Obtain several stability evaluation values. When the stability evaluation value is greater than the set stability threshold, record the time corresponding to the stability evaluation set as a stable point. S2: The time interval between adjacent stable points is recorded as a stable time interval; and the stable time intervals are numbered according to their chronological order. S3: Select the stable time period with the smallest number as the candidate scheduling time period; S4: Obtain the duration of the candidate scheduling period and the output power corresponding to the two stable points; S5: Determine if the difference between the output power of the two stable points is less than the set power difference threshold. If yes, proceed to S6; otherwise, proceed to S9. S6: Obtain the candidate scheduling time periods corresponding to the two stable points, and determine whether the duration of the candidate scheduling time period is greater than the set fluctuation limit time period threshold; if yes, proceed to S7; if no, proceed to S8. S7: Set the candidate scheduling period as the scheduling period, and record the average output power of the two stable points as the output power of the scheduling period; proceed to S11; S8: Obtain the output power corresponding to several moments within the candidate scheduling period, calculate the variance corresponding to the output power, and determine whether the variance is less than the set variance threshold. If yes, proceed to S7; otherwise, proceed to S9. S9: Select several output powers within the selected scheduling period, fit the output powers into a power change curve for the period, calculate the point with the largest rate of change in the power change curve for the period as the cutoff point, and proceed to S10. S10: Use stage points to split the candidate scheduling period into two stable periods, and renumber the stable periods; proceed to S2; S11: Obtain the existing scheduling period and determine whether the total duration of the scheduling period is equal to the total scheduling duration; if yes, proceed to S12; otherwise, renumber the stable period and proceed to S2. S12: Output all scheduling periods and the corresponding output power for each scheduling period.

6. The microgrid economic dispatch method according to claim 1, characterized in that, The objective function in the economic dispatch model for the microgrid, constructed based on the dispatch time periods corresponding to each power generation unit, includes: Obtain the operating cost function and switching cost function corresponding to each power generation unit; obtain the energy storage cost function, electricity purchase cost function, and electricity sales cost function corresponding to each energy storage unit; construct the objective function of the economic dispatch model based on the operating cost function, switching cost function, energy storage cost function, electricity purchase cost function, and electricity sales cost function, wherein the objective function is: ； in, Set the total operating cost of the microgrid over a future time period. Let $\frac{m}{n}$ be the operating cost of the $n$-th power generation unit during the $m$-th scheduling period. The operating switching cost is incurred when the operating status of the nth power generation unit changes from the (m-1)th scheduling period to the mth scheduling period. Let $ be the energy storage cost of the k-th energy storage device. For electricity purchase costs; This refers to the cost of selling electricity.

7. The microgrid economic dispatch method according to claim 6, characterized in that, One method for obtaining the switching cost includes: Obtain the first operating status code of the nth power generation unit during the (m-1)th scheduling period, and the second operating status code of the nth power generation unit during the mth scheduling period; construct a state switching code based on the first and second operating status surfaces, and obtain the corresponding switching cost by querying the switching cost lookup table corresponding to the power generation unit based on the state switching code; the switching cost lookup table includes query items and switching cost items; the query items include several state switching codes; the switching cost items include several switching costs corresponding to the state switching codes.

8. The microgrid economic dispatch method according to claim 1, characterized in that, One construction method based on power constraints includes: Based on the output power of each power generation unit, the energy storage power and output power of each energy storage unit in the operational constraint data, as well as the purchased power, sold power, and load demand power, power balance constraints are constructed; the power balance constraints are as follows: ； in, The output power of the nth power generation unit must satisfy its corresponding output power constraint condition. This represents the output power corresponding to the k-th energy storage unit. Let k be the energy storage power corresponding to the k-th energy storage unit. For the power purchased by the microgrid, This refers to the electricity sales capacity of the microgrid. The load demand power of the microgrid; The power constraints include power balance constraints and output power constraints corresponding to each power generation unit.

9. A microgrid economic dispatch method according to claim 8, characterized in that, One method for constructing the output power constraint conditions includes: Obtain the output power of each power generation unit for each scheduling period, and construct output power constraints for the corresponding power generation unit based on the output power; one expression of the output power constraints for the power generation unit includes: ； in, This represents the output power corresponding to the nth power generation unit. This represents the maximum output power corresponding to the nth power generation unit; one way to express the maximum output power includes: ； in, This represents the output power of the nth power generation unit during the mth scheduling period.

10. A microgrid economic dispatch device, based on the application of the microgrid economic dispatch method according to any one of claims 1 to 9, characterized in that, include: The system includes a data acquisition module, a power prediction module, a data processing module, and a scheduling module. The data acquisition module is used to acquire the predicted environmental data corresponding to the set time of each power generation unit in the microgrid, as well as the operating constraint data of the microgrid. The power prediction module is used to input the predicted environmental data into the power prediction model corresponding to the power generation unit to obtain the corresponding predicted power curve. The data processing module includes an economic scheduling model construction unit and an economic scheduling model solving unit; The economic dispatch model construction unit is used to construct the objective function in the economic dispatch model corresponding to the microgrid based on the dispatch period corresponding to each power generation unit; and to construct several constraint conditions in the economic dispatch model corresponding to the microgrid based on the operational constraint data; the constraint conditions include power constraint conditions and basic constraint conditions. The economic scheduling model solving unit is used to solve the objective function in the economic scheduling model to obtain the optimal scheduling scheme; The scheduling module is used to schedule the power resources in the microgrid based on the scheduling scheme.

Images