Power prediction and dynamic scheduling decision-making system for wind-light-gas storage system

Abstract

The invention relates to the technical field of renewable energy sources, and discloses a wind-light-gas storage system power prediction and dynamic scheduling decision-making system, which comprises a data acquisition module for acquiring wind-light power generation data, meteorological data, energy storage state data and load demand data of a wind-light-gas storage system; the power prediction module is used for outputting a power prediction result through an integrated learning model based on the historical power data, the wind-solar power generation data, the meteorological data and the load demand data; the dynamic scheduling decision module is used for generating a scheduling instruction through a multi-objective optimization model based on the power prediction result, the real-time operation data, the energy storage state data and the load demand data; and the control execution module is used for feeding back and updating the adjusted energy storage state data according to the scheduling instruction. According to the method, the power prediction precision of the wind-light-gas storage system can be improved, the scheduling decision is flexibly optimized, the intermittency and volatility of renewable energy sources are effectively handled, stable power supply is ensured, and efficient operation and cost control of the wind-light-gas storage system are realized.

Description

Technical Field

[0001] This invention relates to the field of renewable energy technology, and in particular to a power prediction and dynamic scheduling decision system for wind, solar, gas and storage systems. Background Technology

[0002] Against the backdrop of energy transition, wind, solar, gas, and storage systems have emerged. However, current power forecasting accuracy for these systems is insufficient. The intermittency and volatility of renewable energy sources make forecasting difficult, and existing methods struggle to accurately predict power output at different times. Furthermore, dynamic dispatching decisions are not flexible enough, failing to make rapid, optimal adjustments based on real-time power changes and demand. This hinders the ability to ensure stable power supply while achieving efficient system operation and cost control, thus limiting the large-scale application of wind, solar, gas, and storage systems.

[0003] Therefore, there is an urgent need to provide a technical solution to address the above problems. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a power prediction and dynamic scheduling decision-making system for wind, solar, gas, and storage systems.

[0005] In a first aspect, the present invention provides a power prediction and dynamic scheduling decision-making system for wind, solar, gas, and storage systems, the technical solution of which is as follows: The data acquisition module is used to collect real-time data on wind and solar power generation, meteorological data, energy storage status data, and load demand data of the wind-solar-gas-storage system. The power prediction module is used to process historical power data, wind and solar power generation data, meteorological data and load demand data through an integrated learning model, and output power prediction results at a specified time scale. The dynamic scheduling decision module is used to generate real-time scheduling instructions based on the power prediction results, the real-time operating data, the energy storage status data, and the load demand data, through a multi-objective optimization model. The control execution module is used to adjust the power output and energy storage status of the wind, solar, gas and storage system according to the real-time scheduling instructions, and at the same time feed back the adjusted energy storage status data to the data acquisition module.

[0006] The beneficial effects of the wind-solar-gas-storage system power prediction and dynamic scheduling decision-making system of the present invention are as follows: The system of this invention can improve the accuracy of power prediction for wind, solar, gas and storage systems, optimize scheduling decisions flexibly, effectively cope with the intermittency and volatility of renewable energy, ensure stable power supply, achieve efficient operation and cost control of wind, solar and gas and storage systems, and promote the large-scale application of wind, solar and gas and storage systems.

[0007] Based on the above scheme, the wind, solar and gas storage system power prediction and dynamic scheduling decision system of the present invention can be further improved as follows.

[0008] In one alternative embodiment, the data acquisition module is used for: The wind and solar power generation data is acquired in real time by sensors connecting the wind power unit and the photovoltaic unit of the wind-solar-gas-storage system. The meteorological data is synchronized from meteorological monitoring equipment and cloud-based meteorological service interfaces; Read the battery management system of the energy storage unit in the wind, solar, gas and energy storage system to obtain the energy storage status data; The load demand data is collected through the power grid dispatch center interface and user-side smart meters.

[0009] In one alternative approach, the power prediction module is specifically used for: The historical power data, the wind and solar power generation data, the meteorological data, and the load demand data are standardized and feature constructed to generate a time-series feature vector; Based on the time-series feature vectors, a heterogeneous base prediction model composed of base models with multiple different algorithm architectures is trained to obtain the trained heterogeneous base prediction model; wherein, each base model is modeled differently for the short-term fluctuation characteristics of wind power and the solar radiation variation characteristics of photovoltaic power. The time-series feature vector is input into the trained heterogeneous base prediction model, and each base model processes the time-series feature vector to generate the corresponding base model prediction result. A dynamic weighted fusion strategy is used to aggregate the prediction results of each base model to generate preliminary power prediction values ​​at a specified time scale; The preliminary power prediction value is subjected to quantile regression processing to output the power prediction result containing a confidence interval; wherein, the confidence interval is used to characterize the prediction uncertainty boundary.

[0010] In one alternative approach, the dynamic scheduling decision module is specifically used for: A multi-objective function is constructed with the objectives of minimizing system operating costs, minimizing power supply stability deviation, and minimizing energy storage lifetime loss, and power fluctuation constraints are set based on the confidence interval in the power prediction results. By integrating the real-time operation data, the energy storage status data, and the load demand data, a dynamic scheduling input vector is generated that includes the real-time grid electricity price, equipment status parameters, and load change trends. The multi-objective function, the power fluctuation constraint, and the dynamic scheduling input vector are input into the multi-objective optimization model, and the optimal scheduling solution set is calculated using the Pareto front algorithm. Based on a preset weighting strategy, the optimal real-time solution is selected from the optimal scheduling solution set and transformed into an executable real-time scheduling instruction.

[0011] In one alternative approach, the expression for the multi-objective function is: ;in, Indicates system operating costs. Indicates power supply stability deviation. Indicates energy storage lifespan loss; The expression for the power fluctuation constraint is: ;in, This represents the lower limit of the total wind and solar power during time period t. This represents the upper limit of the total wind and solar power during time period t. The predicted wind power for time period t. The photovoltaic power forecast for time period t; The equipment status parameters include: gas turbine efficiency coefficient and energy storage state of charge; the load change trend includes: load change rate; the expression for the dynamic scheduling input vector is: ;in, This represents the dynamic scheduling input vector for time period t. The real-time electricity price of the power grid during time period t. The efficiency coefficient of the gas turbine unit. The state of charge of the energy storage during time period t. Let t be the load change rate during the time period.

[0012] In one alternative approach, the expression for the multi-objective optimization model is: ; ; ; In the formula, For the scheduling decision variable vector, and These are the lower and upper limits of the safe state of charge for energy storage, respectively. The energy storage discharge power during time period t is... The energy storage charging power during time period t. The power purchased from the grid during time period t. The output power of the gas turbine unit during time period t. and These are the minimum and maximum output power of the gas turbine unit, respectively.

[0013] In one alternative approach, the system operating cost is expressed as: ;in, The total number of time periods in the scheduling cycle. The cost coefficient for gas-fired power generation during time period t; The expression for the power supply stability deviation is: ;

[0014] The expression for the energy storage lifetime loss is: ;in, The energy storage state-of-charge offset loss coefficient. The state of charge of the energy storage during time period t. This is the optimal operating point for energy storage.

[0015] In one alternative approach, the dynamic scheduling decision module is specifically used for: The corresponding weight configuration mode is activated based on the current operating scenario type of the wind, solar, gas and storage system; wherein, the operating scenario type includes high-fluctuation wind and solar periods, high-load demand periods and normal operating periods; Extract the objective function value of each solution in the optimal scheduling solution set, normalize the objective function value, and calculate the composite weight of the system operating cost target, power supply stability deviation target, and energy storage lifetime loss target; The optimal scheduling solution set is weighted and ranked based on the composite weights, and the solution with the highest score is selected as the real-time optimal solution. The decision variable values ​​in the real-time optimal solution are mapped to the set power values ​​of the gas turbine units, the energy storage charging and discharging commands, and the power limits of the grid interaction of the wind, solar, gas and storage system, so as to generate the real-time dispatch command.

[0016] In one alternative approach, the control execution module is specifically used for: The gas turbine set power value in the real-time dispatch command is analyzed, and the gas turbine of the wind, solar and gas storage system is controlled to gradually adjust its output according to the set power value. According to the energy storage charging and discharging instructions in the real-time scheduling instructions, a charging and discharging enable signal and a target power value are sent to the battery management system of the energy storage unit to ensure that the charging and discharging operations are executed mutually exclusively. Read the grid interaction power limit in the real-time dispatch instruction, and adjust the grid interaction power to the target range through the grid-connected inverter of the wind, solar, gas and storage system; After the energy storage unit completes the state adjustment, the updated energy storage state data is collected in real time and fed back to the data acquisition module; wherein, the updated energy storage state data includes: the updated energy storage state of charge and charging and discharging power.

[0017] Secondly, this invention provides a method for power prediction and dynamic scheduling decision-making of wind, solar, gas, and storage systems. The technical solution of this method is as follows: Real-time collection of wind and solar power generation data, meteorological data, energy storage status data, and load demand data from the wind, solar, gas, and storage system; Based on historical power data, wind and solar power generation data, meteorological data, and load demand data, the data is processed through an integrated learning model to output power prediction results at a specified time scale. Based on the power prediction results, the real-time operation data, the energy storage status data, and the load demand data, a multi-objective optimization model is used to calculate and generate real-time scheduling instructions. The power output and energy storage status of the wind, solar, gas and energy storage system are adjusted according to the real-time scheduling instructions, and the adjusted energy storage status data is fed back and updated.

[0018] The beneficial effects of the power prediction and dynamic scheduling decision-making method for wind, solar, gas and storage systems of the present invention are as follows: The method of this invention can improve the accuracy of power prediction for wind, solar, gas and storage systems, optimize scheduling decisions flexibly, effectively cope with the intermittency and volatility of renewable energy, ensure stable power supply, achieve efficient operation and cost control of wind, solar and gas storage systems, and promote the large-scale application of wind, solar and gas storage systems.

[0019] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention 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 the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a schematic diagram of an embodiment of the power prediction and dynamic scheduling decision-making system for wind, solar, gas and storage systems according to the present invention; Figure 2 This is a flowchart illustrating an embodiment of the power prediction and dynamic scheduling decision-making method for a wind, solar, gas, and storage system according to the present invention. Detailed Implementation

[0022] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. Although exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein.

[0023] Figure 1 This diagram illustrates a structural schematic of an embodiment of a power prediction and dynamic scheduling decision-making system for wind, solar, gas, and storage systems provided by the present invention. Figure 1 As shown, the system includes: The data acquisition module 110 is used to collect real-time data on wind and solar power generation, meteorological data, energy storage status data, and load demand data of the wind-solar-gas-storage system.

[0024] The wind-solar-gas-storage system refers to a hybrid energy system comprising wind power generation units, photovoltaic power generation units, gas turbine generator sets, and energy storage units, used to collaboratively provide power supply. Wind and solar power generation data refers to time-series data of wind power generation and photovoltaic power generation collected in real time by sensors connecting the wind and photovoltaic units. Meteorological data refers to environmental parameters such as solar radiation intensity, wind speed, temperature, and humidity synchronized from meteorological monitoring equipment and cloud-based meteorological service interfaces. Energy storage status data refers to the state of charge (SOC) and real-time charge / discharge power values ​​read from the battery management system of the energy storage unit. Load demand data refers to time-series information on power load demand collected through the grid dispatch center interface and user-side smart meters.

[0025] The power prediction module 120 is used to process historical power data, wind and solar power generation data, meteorological data and load demand data through an integrated learning model, and output power prediction results at a specified time scale.

[0026] Historical power data refers to the historical values ​​of wind power, photovoltaic power, and total output power recorded by the wind-solar-gas-storage system during past operating cycles. An ensemble learning model is a predictive model composed of base models with various algorithmic architectures, used to handle the fluctuating characteristics of renewable energy. The specified time scale refers to the time range covered by the prediction results, including short-term (15 minutes to 4 hours) and ultra-short-term (5 to 15 minutes) predictions. The power prediction results refer to the predicted total wind and solar power values ​​including confidence intervals, used to characterize the expected range and uncertainty boundaries of wind and solar power generation in future periods.

[0027] The dynamic scheduling decision module 130 is used to generate real-time scheduling instructions based on the power prediction results, the real-time operating data, the energy storage status data and the load demand data through a multi-objective optimization model.

[0028] The multi-objective optimization model refers to a mathematical optimization model that aims to minimize system operating costs, power supply stability deviations, and energy storage lifespan losses, while satisfying power fluctuation constraints. Real-time dispatch instructions refer to executable control instructions that include set power values ​​for gas turbine units, energy storage charging and discharging instructions, and grid interaction power limits.

[0029] The control execution module 140 is used to adjust the power output and energy storage status of the wind, solar, gas and storage system according to the real-time scheduling instructions, and at the same time, feed back the adjusted energy storage status data to the data acquisition module.

[0030] Power output refers to the real-time electrical power provided by the gas turbine, wind power unit, and photovoltaic unit of the wind-solar-gas-storage system to the grid or load. Energy storage status refers to the state of charge (SOC) and charging / discharging operation status (charging / discharging / idle) of the energy storage unit.

[0031] The technical solution of this embodiment can improve the accuracy of power prediction and optimize the scheduling decision of wind, solar and gas storage systems, effectively cope with the intermittency and volatility of renewable energy, ensure stable power supply, realize efficient operation and cost control of wind, solar and gas storage systems, and promote the large-scale application of wind, solar and gas storage systems.

[0032] In an alternative embodiment, the data acquisition module 110 is used for: The wind and solar power generation data are acquired in real time by sensors connecting the wind power unit and the photovoltaic unit of the wind-solar-gas-storage system.

[0033] Among them, a wind power unit refers to a wind power generation device consisting of a wind turbine and a converter. A photovoltaic unit refers to a photovoltaic power generation device consisting of photovoltaic modules and an inverter.

[0034] The meteorological data is synchronized from meteorological monitoring equipment and cloud-based meteorological service interfaces.

[0035] Meteorological monitoring equipment refers to: solar meters, anemometers, and temperature and humidity sensors deployed at wind and solar power stations. Cloud-based meteorological service interfaces refer to: web application programming interfaces (APIs) used to obtain refined regional meteorological forecast data.

[0036] The energy storage status data is obtained by reading the battery management system of the energy storage unit in the wind, solar, gas and energy storage system.

[0037] An energy storage unit refers to an electrical energy storage device consisting of a battery pack, a battery management system (BMS), and a power conversion system (PCS). The battery management system is a hardware control unit that monitors and controls the state of charge, temperature, and charging / discharging power of the energy storage unit.

[0038] The load demand data is collected through the power grid dispatch center interface and user-side smart meters.

[0039] The power grid dispatch center interface refers to the data communication interface for exchanging real-time electricity prices and dispatch instructions with the power grid energy management system (EMS). The user-side smart meter refers to the metering device that collects load power data from end users.

[0040] Among the above-mentioned optional methods, the data acquisition method of the data acquisition module is further clarified to ensure that the collected data on wind and solar power generation, meteorology, energy storage status and load demand are comprehensive and accurate, so as to provide reliable data support for subsequent power prediction and scheduling decisions.

[0041] In an alternative embodiment, the power prediction module 120 is specifically used for: The historical power data, wind and solar power generation data, meteorological data, and load demand data are standardized and feature constructed to generate a time-series feature vector.

[0042] Among them, the time series feature vector refers to the multi-dimensional time series matrix of historical power, meteorological and load data generated through standardization and feature construction processing.

[0043] Based on the aforementioned time-series feature vectors, a heterogeneous basis prediction model composed of base models with multiple different algorithm architectures is trained to obtain the trained heterogeneous basis prediction model.

[0044] Among them, each base model is used to model the short-term fluctuation characteristics of wind power and the solar radiation variation characteristics of photovoltaic power. The heterogeneous base prediction model refers to a combination of differentiated base models composed of decision trees, neural networks and support vector machines, which are used to model the fluctuation characteristics of wind power and the solar radiation characteristics of photovoltaic power respectively.

[0045] The time-series feature vector is input into the trained heterogeneous base prediction model, and each base model processes the time-series feature vector to generate the corresponding base model prediction result.

[0046] Among them, the prediction results of the basic models refer to the wind and solar power prediction values ​​independently output by each basic model in the heterogeneous basis prediction model.

[0047] A dynamic weighted fusion strategy is used to aggregate the prediction results of each base model to generate preliminary power prediction values ​​at a specified time scale.

[0048] The dynamic weighted fusion strategy refers to a strategy that dynamically adjusts the weights based on the real-time accuracy of the base model predictions, aggregating the outputs of each base model. The preliminary power prediction value refers to the total wind and solar power prediction value generated by the dynamic weighted fusion strategy before uncertainty quantification. It should be noted that the weight adjustment principle of the dynamic weighted fusion strategy is based on the recent prediction errors of the base models. Calculate dynamic weights It optimizes the model contribution in real time, and its function is to automatically increase the decision weight of the high-precision model during periods of severe wind and solar fluctuations, thereby suppressing prediction jumps.

[0049] The preliminary power prediction value is subjected to quantile regression processing to output the power prediction result containing the confidence interval.

[0050] The confidence interval is used to characterize the boundary of prediction uncertainty.

[0051] It should be noted that the base model for wind power uses a long short-term memory network (LSTM) to handle the temporal fluctuation characteristics of wind speed, while the base model for photovoltaics uses support vector regression (SVR) to fit the nonlinear relationship of solar radiation intensity.

[0052] Among the above-mentioned optional methods, further utilizing time-series feature vectors and heterogeneous basis prediction models, combined with dynamic weighted fusion strategies and quantile regression processing, the power prediction results are accurately output, the prediction uncertainty boundary is quantified, and more accurate basis is provided for scheduling decisions.

[0053] In an alternative embodiment, the dynamic scheduling decision module 130 is specifically used for: A multi-objective function is constructed with the objectives of minimizing system operating costs, minimizing power supply stability deviation, and minimizing energy storage lifetime loss, and power fluctuation constraints are set based on the confidence interval in the power prediction results.

[0054] Here, system operating cost refers to the sum of grid power purchase costs and gas-fired power generation fuel costs for the wind, solar, gas, and energy storage system during the dispatch cycle. Power supply stability deviation refers to the sum of the absolute differences between load demand power and the total system output (net output of wind, solar, gas, and energy storage). Energy storage lifetime loss refers to the equivalent lifetime decay caused by deep charging and discharging cycles and state of charge shifts in energy storage. Confidence interval refers to the fluctuation range of the power prediction value generated through quantile regression processing, characterizing the uncertainty boundary of the prediction result. Power fluctuation constraints refer to the lower and upper limits of the allowable fluctuation range of the total wind and solar power, set based on the confidence interval.

[0055] By integrating the real-time operation data, the energy storage status data, and the load demand data, a dynamic scheduling input vector is generated that includes the real-time grid electricity price, equipment status parameters, and load change trends.

[0056] Among them, the real-time electricity price refers to the time-of-use price signal obtained from the grid dispatch center interface. Equipment status parameters refer to the real-time values ​​of the gas turbine unit efficiency coefficient and the energy storage state of charge. Load change trend refers to the rate of change of load power per unit time (kW / min) calculated from historical load data. The dynamic dispatch input vector refers to the real-time optimized input matrix integrating the real-time electricity price, equipment status parameters, and load change trend.

[0057] The multi-objective function, the power fluctuation constraint, and the dynamic scheduling input vector are input into the multi-objective optimization model, and the optimal scheduling solution set is calculated using the Pareto front algorithm.

[0058] The optimal scheduling solution set refers to the set of non-dominated solutions generated by the Pareto front algorithm in a multi-objective optimization model.

[0059] Based on a preset weighting strategy, the optimal real-time solution is selected from the optimal scheduling solution set and transformed into an executable real-time scheduling instruction.

[0060] The preset weighting strategy refers to the target priority rules configured based on periods of high volatility in wind and solar power, periods of high load demand, and regular operating scenarios. The real-time optimal solution refers to the scheduling scheme with the highest score selected from the optimal scheduling solution set.

[0061] Among the above-mentioned optional methods, a multi-objective function is further constructed, the dynamic scheduling input vector is integrated, the Pareto front algorithm is used to solve for the optimal scheduling solution set, the real-time optimal solution is selected and transformed into scheduling instructions, thereby improving the scientific nature and flexibility of scheduling decisions.

[0062] In one alternative approach, the expression for the multi-objective function is: ;in, This represents the system operating cost (unit: yuan). Indicates power supply stability deviation (unit: kW). This represents the energy storage lifespan loss (unit: dimensionless loss coefficient).

[0063] It should be noted that the system operating cost By accumulating the electricity purchase cost from the grid ( ) and the cost of gas-fired power generation ( This constitutes the economic objective. Power supply stability deviation. Using the absolute value summation form Quantify the instantaneous deviation between power supply and demand. Energy storage lifespan loss. Combined charge-discharge cycle loss ( ) and SOC offset loss ( This comprehensively characterizes battery health degradation. The multi-objective function in this embodiment transforms the three main objectives of economy, reliability, and equipment lifespan into optimizable quantities, supporting multi-objective collaborative decision-making.

[0064] The expression for the power fluctuation constraint is: ;in, This represents the lower limit of the total wind and solar power during time period t (unit: kW). This represents the upper limit of the total wind and solar power during time period t (unit: kW). The predicted wind power output for time period t is expressed in kW. Let be the predicted photovoltaic power (in kW) for time period t. It should be noted that the power fluctuation constraint is based on the confidence interval of the power prediction results. Set total power constraints for wind and solar power. The purpose of power fluctuation constraints is to quantify prediction uncertainty into scheduling boundaries, preventing actual wind and solar power from exceeding the prediction range and causing scheduling instability. The predicted wind and solar power for time period t (unit: kW) The confidence interval half-width for time period t is given (unit: kW).

[0065] The equipment status parameters include: gas turbine unit efficiency coefficient and energy storage state of charge; the load change trend includes: load change rate; the expression for the dynamic scheduling input vector is: ;in, This represents the dynamic scheduling input vector for time period t. The real-time electricity price of the power grid during time period t (unit: yuan / kWh). The efficiency coefficient of the gas turbine unit (dimensionless). State of charge of energy storage during time period t (unit: %) Let t be the load change rate (kW / min) during the time period t.

[0066] Among the above-mentioned optional methods, a multi-objective optimization model is further used to comprehensively consider system operating costs, power supply stability deviations, and energy storage lifespan losses, thereby optimizing the scheduling of the wind, solar, gas, and energy storage system and ensuring efficient and stable system operation.

[0067] In one alternative approach, the expression for the multi-objective optimization model is: ; ; ; In the formula, For the scheduling decision variable vector, and These represent the lower and upper limits of the safe state of charge for energy storage (unit: %). The energy storage discharge power during time period t (unit: kW) The energy storage charging power during time period t (unit: kW). The power purchased from the grid during time period t (unit: kW). The output power of the gas turbine unit during time period t (unit: kW). and These are the minimum and maximum output power of the gas turbine unit (unit: kW).

[0068] It should be noted that, This represents the power fluctuation constraint condition. This indicates the safety constraints of the energy storage SOC. This indicates the output constraints of the gas turbine unit. This represents the mutual exclusion constraint condition for energy storage charging and discharging.

[0069] The design stems from the physical limitations of the energy storage unit, which prevents the battery management system from performing charge and discharge operations simultaneously. This constraint avoids equipment damage caused by command conflicts, ensuring hardware safety.

[0070] Among the above-mentioned optional methods, the scheduling decision variable vector and its range are further defined, the output power constraints of the gas turbine units are clarified, and the system is ensured to meet the equipment operation requirements when optimizing the scheduling, thereby improving the reliability of the system operation.

[0071] In one alternative approach, the system operating cost is expressed as: ;in, The total number of time periods in the scheduling cycle. The cost coefficient for gas-fired power generation during time period t (unit: yuan / kWh); The expression for the power supply stability deviation is: ; The expression for the energy storage lifetime loss is: ;in, The energy storage state of charge offset loss coefficient (unit: %⁻¹). State of charge of energy storage during time period t (unit: %) The optimal operating point for energy storage (unit: %).

[0072] Among the above-mentioned optional methods, the system operating cost, power supply stability deviation and energy storage life loss are further expressed separately, and each objective is quantified, which makes it easier to balance multiple objectives in dynamic scheduling decisions and achieve optimal comprehensive benefits.

[0073] In an alternative embodiment, the dynamic scheduling decision module 130 is specifically used for: The corresponding weight configuration mode is activated based on the current operating scenario type of the wind, solar, gas, and storage system.

[0074] The current operating scenario type refers to the status label of the system in real time, identifying periods of high volatility in wind and solar power, periods of high load demand, or periods of normal operation. The weight configuration mode refers to the weight allocation template for cost targets, stability targets, and lifespan loss targets preset for different operating scenarios. Operating scenario types include periods of high volatility in wind and solar power, periods of high load demand, and periods of normal operation. Periods of high volatility in wind and solar power refer to the time period when the half-width of the predicted confidence interval for wind and solar power exceeds a threshold. Periods of high load demand refer to the time period when the load power exceeds 80% of the system's rated capacity. Periods of normal operation refer to the time period when high volatility or high load conditions are not met.

[0075] Extract the objective function value of each solution in the optimal scheduling solution set, normalize the objective function value, and calculate the composite weights of the system operating cost target, power supply stability deviation target, and energy storage lifetime loss target.

[0076] Here, the objective function value refers to the system operating cost, power supply stability deviation, and energy storage lifetime loss corresponding to each solution in the optimal scheduling solution set. The composite weight refers to the multi-objective weighting coefficients calculated by combining the normalized objective function value with the weight configuration mode.

[0077] The optimal scheduling solution set is weighted and ranked based on the composite weights, and the solution with the highest score is selected as the real-time optimal solution.

[0078] The decision variable values ​​in the real-time optimal solution are mapped to the set power values ​​of the gas turbine units, the energy storage charging and discharging commands, and the power limits of the grid interaction of the wind, solar, gas and storage system, so as to generate the real-time dispatch command.

[0079] Among them, the set power value of the gas turbine unit refers to the target output power value of the gas turbine unit in the real-time optimal solution. The energy storage charge and discharge command refers to the charge and discharge enable signal (0 / 1) and target power value of the energy storage unit in the real-time optimal solution. The grid interaction power limit refers to the power boundary value that is allowed to be purchased or sold from the grid in the real-time optimal solution.

[0080] In the above-mentioned optional methods, different weight configuration modes are further activated according to the type of operation scenario, the optimal scheduling solution set is weighted and ranked, the real-time optimal solution is selected and a scheduling instruction is generated, so as to achieve precise and flexible scheduling control.

[0081] In an alternative embodiment, the control execution module 140 is specifically used for: The gas turbine set power value in the real-time dispatch command is analyzed, and the gas turbine of the wind, solar and gas storage system is controlled to gradually adjust its output according to the set power value.

[0082] According to the energy storage charging and discharging instructions in the real-time scheduling instructions, a charging and discharging enable signal and a target power value are sent to the battery management system of the energy storage unit to ensure that the charging and discharging operations are executed mutually exclusively.

[0083] The charge / discharge enable signal refers to the Boolean instruction (1: charge enable, 0: discharge enable) that controls the switching of the charge / discharge state of the energy storage unit. The target power value refers to the absolute value of the charge / discharge power set in the energy storage charge / discharge instruction.

[0084] Read the grid interaction power limit from the real-time dispatch command, and adjust the grid interaction power to the target range through the grid-connected inverter of the wind, solar, gas and storage system.

[0085] The target range refers to the power range defined by the power limit of the power grid interaction.

[0086] After the energy storage unit completes the state adjustment, the updated energy storage state data is collected in real time and fed back to the data acquisition module 110.

[0087] The updated energy storage status data includes: updated energy storage state of charge (SOC) and charge / discharge power.

[0088] Among the above-mentioned optional methods, the dispatch instructions are further analyzed more precisely to regulate the output of the gas turbine unit, control the charging and discharging of energy storage, adjust the power grid interaction, ensure the feedback and updating of energy storage status data, and guarantee the coordinated operation of all parts of the system.

[0089] Figure 2 This diagram illustrates a flowchart of an embodiment of a power prediction and dynamic scheduling decision-making method for wind, solar, gas, and storage systems provided by the present invention. Figure 2 As shown, it includes the following steps: S1. Real-time acquisition of wind and solar power generation data, meteorological data, energy storage status data, and load demand data of the wind, solar, gas, and storage system; S2. Based on historical power data, wind and solar power generation data, meteorological data, and load demand data, the data is processed through an integrated learning model to output power prediction results at a specified time scale. S3. Based on the power prediction results, the real-time operation data, the energy storage status data, and the load demand data, a real-time scheduling command is generated by performing calculations through a multi-objective optimization model. S4. Adjust the power output and energy storage status of the wind, solar, gas and storage system according to the real-time scheduling command, and update the adjusted energy storage status data.

[0090] The technical solution of this embodiment can improve the accuracy of power prediction and optimize the scheduling decision of wind, solar and gas storage systems, effectively cope with the intermittency and volatility of renewable energy, ensure stable power supply, realize efficient operation and cost control of wind, solar and gas storage systems, and promote the large-scale application of wind, solar and gas storage systems.

[0091] Furthermore, the system provided in the above embodiments is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the system can be divided into different functional modules according to the actual situation to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiments, which will not be repeated here.

[0092] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this invention.

[0093] It should be noted that the terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and represent a limitation on a specific order or sequence. Where appropriate, the order of use for similar objects can be interchanged so that the embodiments of this application described herein can be implemented in an order other than that shown or described.

[0094] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A power prediction and dynamic scheduling decision-making system for wind, solar, gas, and storage systems, characterized in that, The system includes: The data acquisition module is used to collect real-time data on wind and solar power generation, meteorological data, energy storage status data, and load demand data of the wind-solar-gas-storage system. The power prediction module is used to process historical power data, wind and solar power generation data, meteorological data and load demand data through an integrated learning model, and output power prediction results at a specified time scale. The dynamic scheduling decision module is used to generate real-time scheduling instructions based on the power prediction results, the real-time operating data, the energy storage status data, and the load demand data, through a multi-objective optimization model. The control execution module is used to adjust the power output and energy storage status of the wind, solar, gas and storage system according to the real-time scheduling instructions, and at the same time feed back the adjusted energy storage status data to the data acquisition module.

2. The power prediction and dynamic scheduling decision-making system for wind, solar, gas, and storage systems according to claim 1, characterized in that, The data acquisition module is used for: The wind and solar power generation data is acquired in real time by sensors connecting the wind power unit and the photovoltaic unit of the wind-solar-gas-storage system. The meteorological data is synchronized from meteorological monitoring equipment and cloud-based meteorological service interfaces; Read the battery management system of the energy storage unit in the wind, solar, gas and energy storage system to obtain the energy storage status data; The load demand data is collected through the power grid dispatch center interface and user-side smart meters.

3. The power prediction and dynamic scheduling decision-making system for wind, solar, gas, and storage systems according to claim 1, characterized in that, The power prediction module is specifically used for: The historical power data, the wind and solar power generation data, the meteorological data, and the load demand data are standardized and feature constructed to generate a time-series feature vector; Based on the time-series feature vectors, a heterogeneous base prediction model composed of base models with multiple different algorithm architectures is trained to obtain the trained heterogeneous base prediction model; wherein, each base model is modeled differently for the short-term fluctuation characteristics of wind power and the solar radiation variation characteristics of photovoltaic power. The time-series feature vector is input into the trained heterogeneous base prediction model, and each base model processes the time-series feature vector to generate the corresponding base model prediction result. A dynamic weighted fusion strategy is used to aggregate the prediction results of each base model to generate preliminary power prediction values ​​at a specified time scale; The preliminary power prediction value is subjected to quantile regression processing to output the power prediction result containing a confidence interval; wherein, the confidence interval is used to characterize the prediction uncertainty boundary.

4. The power prediction and dynamic scheduling decision-making system for wind, solar, gas, and storage systems according to claim 3, characterized in that, The dynamic scheduling decision module is specifically used for: A multi-objective function is constructed with the objectives of minimizing system operating costs, minimizing power supply stability deviation, and minimizing energy storage lifetime loss, and power fluctuation constraints are set based on the confidence interval in the power prediction results. By integrating the real-time operation data, the energy storage status data, and the load demand data, a dynamic scheduling input vector is generated that includes the real-time grid electricity price, equipment status parameters, and load change trends. The multi-objective function, the power fluctuation constraint, and the dynamic scheduling input vector are input into the multi-objective optimization model, and the optimal scheduling solution set is calculated using the Pareto front algorithm. Based on a preset weighting strategy, the optimal real-time solution is selected from the optimal scheduling solution set and transformed into an executable real-time scheduling instruction.

5. The power prediction and dynamic scheduling decision-making system for wind, solar, gas, and storage systems according to claim 4, characterized in that, The expression for the multi-objective function is: ;in, Indicates system operating costs. Indicates power supply stability deviation. Indicates energy storage lifespan loss; The expression for the power fluctuation constraint is: ;in, This represents the lower limit of the total wind and solar power during time period t. This represents the upper limit of the total wind and solar power during time period t. The predicted wind power for time period t. The photovoltaic power forecast for time period t; The equipment status parameters include: gas turbine efficiency coefficient and energy storage state of charge; the load change trend includes: load change rate; the expression for the dynamic scheduling input vector is: ;in, This represents the dynamic scheduling input vector for time period t. The real-time electricity price of the power grid during time period t. The efficiency coefficient of the gas turbine unit. The state of charge of the energy storage during time period t. Let t be the load change rate during the time period.

6. The power prediction and dynamic scheduling decision-making system for wind, solar, gas, and storage systems according to claim 5, characterized in that, The expression for the multi-objective optimization model is: ； ； ； In the formula, For the scheduling decision variable vector, and These are the lower and upper limits of the safe state of charge for energy storage, respectively. The energy storage discharge power during time period t. The energy storage charging power during time period t. The power purchased from the grid during time period t. The output power of the gas turbine unit during time period t. and These are the minimum and maximum output power of the gas turbine unit, respectively.

7. The power prediction and dynamic scheduling decision-making system for wind, solar, gas, and storage systems according to claim 6, characterized in that, The expression for the system operating cost is: ;in, The total number of time periods in the scheduling cycle. The cost coefficient for gas-fired power generation during time period t; The expression for the power supply stability deviation is: ; The expression for the energy storage lifetime loss is: ;in, The energy storage state-of-charge offset loss coefficient. The state of charge of the energy storage during time period t. This is the optimal operating point for energy storage.

8. The power prediction and dynamic scheduling decision-making system for wind, solar, gas, and storage systems according to any one of claims 5 to 7, characterized in that, The dynamic scheduling decision module is specifically used for: The corresponding weight configuration mode is activated based on the current operating scenario type of the wind, solar, gas and storage system; wherein, the operating scenario type includes high-fluctuation wind and solar periods, high-load demand periods and normal operating periods; Extract the objective function value of each solution in the optimal scheduling solution set, normalize the objective function value, and calculate the composite weight of the system operating cost target, power supply stability deviation target, and energy storage lifetime loss target; The optimal scheduling solution set is weighted and ranked based on the composite weights, and the solution with the highest score is selected as the real-time optimal solution. The decision variable values ​​in the real-time optimal solution are mapped to the set power values ​​of the gas turbine units, the energy storage charging and discharging commands, and the power limits of the grid interaction of the wind, solar, gas and storage system, so as to generate the real-time dispatch command.

9. The power prediction and dynamic scheduling decision-making system for wind, solar, gas, and storage systems according to claim 8, characterized in that, The control execution module is specifically used for: The gas turbine set power value in the real-time dispatch command is analyzed, and the gas turbine of the wind, solar and gas storage system is controlled to gradually adjust its output according to the set power value. According to the energy storage charging and discharging instructions in the real-time scheduling instructions, a charging and discharging enable signal and a target power value are sent to the battery management system of the energy storage unit to ensure that the charging and discharging operations are executed mutually exclusively. Read the grid interaction power limit in the real-time dispatch instruction, and adjust the grid interaction power to the target range through the grid-connected inverter of the wind, solar, gas and storage system; After the energy storage unit completes the state adjustment, the updated energy storage state data is collected in real time and fed back to the data acquisition module; wherein, the updated energy storage state data includes: updated energy storage state of charge and charging / discharging power.

10. A method for power prediction and dynamic scheduling decision-making in a wind-solar-gas-storage system, characterized in that, The method includes: Real-time collection of wind and solar power generation data, meteorological data, energy storage status data, and load demand data from the wind, solar, gas, and storage system; Based on historical power data, wind and solar power generation data, meteorological data, and load demand data, the data is processed through an integrated learning model to output power prediction results at a specified time scale. Based on the power prediction results, the real-time operation data, the energy storage status data, and the load demand data, a multi-objective optimization model is used to calculate and generate real-time scheduling instructions. The power output and energy storage status of the wind, solar, gas and energy storage system are adjusted according to the real-time scheduling instructions, and the adjusted energy storage status data is fed back and updated.

Images