Urban area integrated energy system dynamic aggregation and optimal dispatching method and system

Abstract

The application discloses a method and system for dynamic aggregation and optimal scheduling of an urban area comprehensive energy system, which is based on a polyhedral model, aggregates different original models and constraint conditions of the same type of device into an aggregated model and aggregated constraint condition of the device, generates an original scene based on the aggregated model of each type of device, filters the original scene based on the aggregated constraint condition of each type of device, reduces the filtered scene to obtain a typical scene set according to a similarity index and a representative index of the scene, establishes a day-ahead optimal scheduling target function and an intra-day rolling scheduling target function based on the running parameters of the typical scene generated by the aggregated model and the aggregated constraint condition, and solves the day-ahead optimal scheduling target function and the intra-day rolling scheduling target function to obtain a system scheduling scheme under the conditions of satisfying the running constraints of each type of device and the power balance constraints of the comprehensive energy system, thereby significantly improving the optimal scheduling effect of the comprehensive energy system.

Description

Technical Field

[0001] This invention belongs to the field of integrated energy system optimization and scheduling technology, and particularly relates to a method and system for dynamic aggregation and optimized scheduling of urban regional integrated energy systems that considers the uncertainty of source load. Background Technology

[0002] An integrated energy system (IES) is a system that organically combines and optimizes multiple energy forms such as electricity, heat, and gas. It can realize the conversion and distribution of energy between different energy systems. Its multi-energy synergy effect can significantly improve the system's energy efficiency and stability, and can also greatly promote the application of renewable energy.

[0003] With the rapid development of distributed energy resources, the variety and number of devices in energy systems (IES) have increased significantly, leading to a substantial increase in computational complexity. Furthermore, the inherent uncertainty of renewable energy sources and loads further complicates the energy supply and demand relationship within the system, posing challenges to the optimal operation of IES.

[0004] In existing technologies, the optimized scheduling of integrated energy systems (IES) involves real-time data monitoring of the integrated energy system to obtain integrated energy monitoring data, including power monitoring data, thermal energy monitoring data, renewable energy monitoring data, and energy storage monitoring data. Energy demand analysis is then performed on this integrated energy monitoring data to generate energy demand data. Based on the energy demand data, the integrated energy system is divided into energy nodes, constructing an energy node set, which includes power nodes, thermal energy nodes, renewable energy nodes, and energy storage nodes. Operational constraint analysis is then performed on the energy node set to generate operational constraint data. Finally, energy load calculations are performed on the energy node set based on the operational constraint data to generate energy node load data. While this achieves efficient and accurate energy optimization scheduling, it fails to aggregate models of similar energy devices to address the complex coupling relationships among multiple energy sources in an IES, making it difficult to efficiently optimize the scheduling of these resources. This results in a high computational burden, low system operating efficiency, and a lack of effective and intuitive scheduling optimization methods when dealing with the randomness and volatility of renewable energy. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for dynamic aggregation and optimal scheduling of integrated energy systems (IES) in urban areas. By considering the numerous and varied technical parameters of similar equipment in actual engineering projects, as well as the uncertainty of source and load, the invention aggregates models of similar equipment and uses scenario analysis to account for this uncertainty, aiming to effectively optimize the scheduling of IES across multiple time scales. Aggregating models of similar energy equipment effectively reduces the computational burden, thereby improving system operating efficiency. Furthermore, scenario analysis can better describe the randomness and volatility of renewable energy. Taking all these factors into account, the optimal scheduling effect of IES can be significantly improved.

[0006] The present invention adopts the following technical solution.

[0007] This invention proposes a method for dynamic aggregation and optimal scheduling of integrated energy systems in urban areas, comprising:

[0008] The system acquires operating parameters of various types of devices within the integrated energy system under different operating conditions to establish different original models and constraints for each type of device. Based on the polyhedral model, different original models and constraints of the same type of device are aggregated into an aggregate model and aggregate constraints for that type of device. Original scenarios are generated based on the aggregate models of each type of device, and the original scenarios are filtered based on the aggregate constraints of each type of device. Based on the similarity and representativeness indicators of the scenarios, the filtered scenarios are reduced to obtain a set of typical scenarios. Based on the operating parameters of the typical scenarios generated by the aggregate models and aggregate constraints, the objective function for day-ahead optimization scheduling is to minimize the sum of energy purchase cost, operation and maintenance cost, and penalty cost of all typical scenarios, and the objective function for intraday rolling scheduling is to minimize the sum of changes in energy purchase cost, operation and maintenance cost, and penalty cost of all typical scenarios. The system scheduling scheme is obtained by solving the objective functions for day-ahead optimization scheduling and intraday rolling scheduling under the operating constraints of each type of device and the power balance constraints of the integrated energy system.

[0009] Preferably, the different types of devices in the integrated energy system include: wind power generation devices, photovoltaic power generation devices, electric energy storage devices, gas energy storage devices, thermal energy storage devices, electric-to-gas generator units, CHP generator units, and electric boilers;

[0010] The original model includes a power balance model; constraints include, but are not limited to, capacity constraints, power constraints, and state constraints.

[0011] Preferably, the different original models and constraints of various types of devices, and the power balance constraints of the integrated energy system are used as input data for the compiler; the compiler outputs a set of original polyhedral models. in, Let be the constraint matrix of the i-th primitive polyhedron model of the m-th type of device. Let x be the constraint vector of the i-th primitive polyhedral model of the m-th type of device, x be the operating parameters under different operating conditions, m = 1, 2, ..., M, where M is the total number of device types in the integrated energy system, and i = 1, 2, ..., N, where N is the total number of primitive polyhedral models; primitive polyhedral models The i-th original model and constraints corresponding to the m-th type of device.

[0012] Preferably, the average value of the original polyhedral models of the same type of device is calculated to obtain the basic polyhedral models of each type of device. The following relationship must be satisfied:

[0013]

[0014] In the formula, Let be the constraint matrix of the basic polyhedral model of the m-th type of device. Let be the constraint vector of the basic polyhedral model of the m-th type of device. These are the average values ​​of operating parameters under different working conditions;

[0015] Basic polyhedral model The basic model and basic constraints corresponding to the m-th type of device.

[0016] Preferably, in the same type of device, the scaling factor and translation vector of each original polyhedron model are determined with the optimization objective of having the scaled and translated basic polyhedron model contain the original polyhedron model and having the smallest scaling factor.

[0017] The basic polyhedron model after scaling and translation satisfies the following relationship:

[0018]

[0019] In the formula, Let be the scaling factor for the i-th primitive polyhedron model of the m-th type of device. Let be the translation vector of the i-th primitive polyhedral model of the m-th type of device.

[0020] Preferably, in the same type of device, the mapping model and constraint conditions between the constraint matrix of the original polyhedron model and the constraint matrix of the basic polyhedron model satisfy the following relationship:

[0021]

[0022] In the formula, U m Let be the mapping matrix between the constraint matrix of the original polyhedral model and the constraint matrix of the basic polyhedral model in the m-th type of device.

[0023] Preferably, in the same type of device, when the mapping model and constraint conditions are met, the average value of the scaling factor of all original polyhedral models is calculated as the scaling factor of the device; the constraint vector of the basic polyhedral model is updated using the scaling factor of the device and the total number of original polyhedral models.

[0024] The scaling factors for each type of device satisfy the following relationship:

[0025]

[0026] In the formula, Let m be the scaling factor for the m-th type of device;

[0027] The updated constraint vectors of the basic polyhedron model satisfy the following relationship:

[0028]

[0029] In the formula, The updated constraint vectors for the basic polyhedron model.

[0030] Preferably, in the same type of device, the aggregated polyhedron model of the device is constructed using the constraint matrix of the basic polyhedron model and the updated constraint vector. The aggregate polyhedral model set of integrated energy systems is M represents the total number of device types in the integrated energy system;

[0031] The aggregation polyhedron model corresponds to the aggregation model and aggregation constraints of the m-th type of device.

[0032] Preferably, prediction data is generated using an aggregation model of various types of devices; based on the probability distribution of the prediction data, a sampling method is used to obtain prediction samples to construct different original scenes; the original scenes are filtered using aggregation constraints of various types of devices; and the filtered scenes are reduced according to similarity and representativeness indicators, including:

[0033] 1) The similarity index between two scenarios satisfies the following relationship:

[0034] x(i,j)=∑|Ws(:,i)-Ws(:,j)|

[0035] In the formula, x(i,j) is the distance between scene i and typical scene j, and Ws(:,i) and Ws(:,j) are the data of the i-th scene and the j-th scene in the original scene set Ws, respectively;

[0036] 2) Calculate the average similarity index y0(i) between each scene and the other scenes as a representative index for each scene, satisfying the following relationship:

[0037]

[0038] In the formula, x(:,i) is the similarity index between scene i and the other scenes;

[0039] 3) The scenario with the smallest representativeness index is designated as the first scenario, and the scenario with the smallest similarity index to the first scenario is designated as the second scenario. The sum of the probabilities of the first scenario and the second scenario is used as the updated probability of the second scenario, satisfying the following relationship:

[0040] p′ r =p r +p d

[0041] In the formula, p′ r p represents the updated probability for the second scene. r p represents the probability of the second scenario. d The probability of the first scenario;

[0042] 4) Reduce the first scene from the scene after the first reduction to obtain the updated scene set; repeat steps 1) to 4) for the updated scene set until the total number of scenes in the updated scene set reaches the set value, and output the updated scene set as the typical scene set.

[0043] Preferably, the objective function for day-ahead optimization scheduling satisfies the following relationship:

[0044]

[0045] In the formula, F ahead Let P be the objective function for optimizing the current schedule. a Let represent the probability of typical scenario 'a', n represent the total number of typical scenarios, EPC is the energy purchase cost, OMC is the operation and maintenance cost, and WCP is the penalty cost.

[0046] Preferably, based on the aggregation model and aggregation constraints of each type of device, the purchased electricity and gas quantities for each time period are determined, and the energy purchase cost is calculated using the following formula:

[0047]

[0048] In the formula, λ e,t , λ g,t The time-of-use electricity price and gas price for period t are respectively, P g,t G represents the system's power purchase capacity during time period t. g,t Let T be the amount of natural gas purchased by the system during time period t, where T is the total number of time periods.

[0049] Based on the aggregation model and aggregation constraints of various types of devices, the charging and discharging power of energy storage devices and the operating power of energy conversion devices in each time period are determined, and the operation and maintenance costs are calculated using the following formula:

[0050]

[0051] In the formula, β k Let β be the unit charge / discharge cost of the k-th type of energy storage device. l The unit operating cost of the j-th type of energy conversion device, P represents the charging and discharging power of the k-th type of energy storage device during time period t. j，t Let N be the operating power of the j-th type of energy conversion device during time period t. S N represents the total number of energy storage device types. C The total number of energy conversion device types;

[0052] The penalty cost, including the cost of wind curtailment, satisfies the following relationship:

[0053]

[0054] In the formula, k w ΔP is the wind curtailment penalty coefficient. c，t Let T represent the wind curtailment power during time period t, where T is the total number of time periods.

[0055] Preferably, the objective function for intraday rolling scheduling satisfies the following relationship:

[0056]

[0057] In the formula, F inday Let P be the objective function for intraday rolling scheduling. s Let represent the probability of typical scenario s, n represent the total number of typical scenarios, EPC represent energy purchase cost, and ΔOMC represent the change in operation and maintenance cost.

[0058] WCP stands for penalty cost.

[0059] Preferably, based on the aggregation model and aggregation constraints of each type of device, the purchased electricity and gas quantities for each time period are determined, and the energy purchase cost is calculated using the following formula:

[0060]

[0061] In the formula, λ e，t , λ g，t The time-of-use electricity price and gas price for period t are respectively, P g,t G represents the system's power purchase capacity during time period t. g,t The amount of natural gas purchased by the system during time period t, where Tinday is the total number of rolling time periods within the day;

[0062] Based on the aggregation model and aggregation constraints of various types of devices, the changes in charging and discharging power of energy storage devices and the changes in operating power of energy conversion devices are determined for each time period. The changes in operation and maintenance costs are then calculated using the following formula:

[0063]

[0064] In the formula, γ k γ J These are the power change penalty cost coefficients for the k-th type energy storage device and the j-th type energy conversion device, respectively. Let ΔP be the change in charging and discharging power of the k-th type of energy storage device during time period t. j,t Let N be the change in operating power of the j-th type of energy conversion device during time period t. S N represents the total number of energy storage device types. C The total number of energy conversion device types;

[0065] The penalty cost, including the cost of wind curtailment, satisfies the following relationship:

[0066]

[0067] In the formula, k w ΔP is the wind curtailment penalty coefficient. c,t The curtailment power is the wind power during time period t, and Tinday is the total number of rolling time periods within the day.

[0068] This invention also proposes a dynamic aggregation and optimization scheduling system for integrated urban energy systems, comprising:

[0069] The dynamic aggregation module is used to obtain the operating parameters of various types of devices in the integrated energy system under different operating conditions, so as to establish different original models and constraints of each type of device; based on the polyhedral model, different original models and constraints of the same type of device are aggregated into an aggregated model and aggregated constraints of that type of device.

[0070] The optimization scheduling module generates original scenarios based on the aggregation model of each type of device, and filters the original scenarios based on the aggregation constraints of each type of device. Based on the similarity and representativeness indicators of the scenarios, it reduces the filtered scenarios to obtain a set of typical scenarios. Based on the typical scenario operating parameters generated by the aggregation model and aggregation constraints, it sets the objective function for day-ahead optimization scheduling as minimizing the sum of energy purchase cost, operation and maintenance cost, and penalty cost of all typical scenarios, and the objective function for intraday rolling scheduling as minimizing the sum of changes in energy purchase cost, operation and maintenance cost, and penalty cost of all typical scenarios. The system scheduling scheme is obtained by solving the objective functions for day-ahead optimization scheduling and intraday rolling scheduling under the operating constraints of each type of device and the power balance constraints of the integrated energy system.

[0071] The present invention is also a terminal, including a processor and a storage medium; the storage medium is used to store instructions; the processor is used to perform operations according to the instructions to execute the steps of the method.

[0072] The present invention is also a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method.

[0073] The beneficial effects of this invention are as follows: compared with the prior art, the method proposed in this invention effectively reduces the complexity of system modeling and calculation by aggregating different models and constraints of the same type of device, while retaining key dynamic characteristics, thus possessing strong practicality; by employing Latin hypercube sampling and scenario reduction, it can reasonably consider the uncertainties of renewable energy and load, ensuring the representativeness of the scheduling scheme and the robustness of the system; and through multi-timescale optimization scheduling methods, it can balance short-term and long-term operational goals, minimizing economic costs while ensuring system safety and stability. Attached Figure Description

[0074] Figure 1 This is a flowchart of the dynamic aggregation and optimal scheduling method for urban area integrated energy systems proposed in this invention;

[0075] Figure 2 This is an architecture diagram of the integrated energy system used in the embodiments of the present invention;

[0076] Figure 3 These are the day-ahead wind and solar power output forecast data in the embodiments of the present invention;

[0077] Figure 4 These are the various day-ahead load forecast data in the embodiments of the present invention;

[0078] Figure 5 These are the day-ahead and intraday dispatch output curves of the electric energy storage device and the gas energy storage device in the embodiments of the present invention;

[0079] Figure 6 These are the day-ahead and intraday dispatch output curves of the thermal energy storage device and the P2G device in this embodiment of the invention;

[0080] Figure 7 These are the day-ahead and intraday dispatch output curves of the CHP unit and electric boiler in this embodiment of the invention. Detailed Implementation

[0081] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The embodiments described in this application are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of this invention.

[0082] This invention proposes a dynamic aggregation and optimal scheduling method for urban regional integrated energy systems that considers source-load uncertainty. It includes three parts: system modeling and model aggregation, scenario analysis, and optimal scheduling. The method achieves adaptive aggregation of different models in the integrated energy system through the scaling and translation method of polyhedral models; it optimizes the construction of typical scenarios by using Latin hypercube sampling and K-means clustering; and it achieves synergy between long-term planning and short-term dynamic optimization through multi-timescale optimal scheduling.

[0083] like Figure 1 As shown, the method includes:

[0084] Step 1: Obtain the operating parameters of various types of devices in the integrated energy system under different operating conditions to establish different original models and constraints for each type of device; based on the polyhedral model, aggregate the different original models of the same type of device into an aggregate model of that type of device.

[0085] Specifically, step 1 includes:

[0086] Step 1.1: Obtain the operating parameters of various types of devices within the integrated energy system under different operating conditions;

[0087] Specifically, different types of devices within an integrated energy system include, but are not limited to: wind power generation devices, photovoltaic power generation devices, electric energy storage devices, gas energy storage devices, thermal energy storage devices, power-to-gas (P2G) units, CHP units, and electric boilers (EB). Devices of the same type refer to devices that have the same function but have different operating parameters, models, or manufacturers. For example, electric energy storage devices with different operating parameters are devices of the same type, and gas energy storage devices of different models are also devices of the same type.

[0088] Step 1.2: Establish different original models and constraints for each type of device using operating parameters under different operating conditions; use the different original models and constraints of each type of device and the power balance constraints of the integrated energy system as input data for the compiler; use the compiler to output the set of original polyhedral models.

[0089] The original models for each type of device include, but are not limited to, the power balance model; the constraints for each type of device include, but are not limited to, capacity constraints, power constraints, and state constraints; in the embodiments, models for electric energy storage devices, gas energy storage devices, and thermal energy storage devices are established respectively, and the limitations of actual hardware conditions are considered to constrain the charging and discharging efficiency, charging and discharging power, and energy storage device state.

[0090] Specifically, the original model and constraints of the energy storage device satisfy the following relationship:

[0091]

[0092] In the formula, This represents the energy storage capacity of the e-th energy storage device during time period t. These represent the charging and discharging power of the e-th energy storage device during time period t, respectively. These represent the charging and discharging efficiencies of the e-th energy storage device, respectively. Let represent the lower limit and upper limit of the energy storage capacity of the e-th energy storage device, respectively. The variables are 0-1, representing the charging and discharging states of the e-th energy storage device during time period t. Let represent the minimum and maximum charging power of the e-th energy storage device, respectively. Let represent the minimum and maximum discharge power of the e-th energy storage device, respectively. These represent the final and initial energy storage capacities of the e-th energy storage device within the scheduling cycle, respectively.

[0093] Similarly, establish the original models and constraints for gas energy storage devices and thermal energy storage devices.

[0094] In this embodiment, the original model and constraints of the energy coupling device are established, including:

[0095] P2G units are devices that convert electrical energy into synthetic natural gas. During peak periods of renewable energy generation, surplus electricity can be converted into natural gas for storage and use. This can effectively improve the economic efficiency of integrated energy system operation and promote wind power consumption. The original model and constraints of P2G units satisfy the following relationship:

[0096]

[0097] In the formula, Let represent the gas power generated and the electrical power consumed by the i-th P2G unit during time period t, respectively. This represents the electrical conversion efficiency of the i-th P2G unit.

[0098] Let represent the lower and upper limits of the power consumption of the i-th P2G unit, respectively. Let represent the maximum downhill ramp rate and the maximum uphill ramp rate of the i-th P2G unit, respectively;

[0099] A CHP unit is a coupled device that converts natural gas into electricity and heat. The original model and constraints of a CHP unit satisfy the following relationship:

[0100]

[0101] In the formula, These represent the electrical power, thermal power, and gas power consumed by the c-th CHP unit during time period t, respectively. These represent the electrothermal power ratio and gas-to-electric conversion efficiency of the c-th CHP unit, respectively.

[0102] Electric boilers convert electrical energy into heat energy through water heating, offering advantages such as zero pollution, high efficiency, and safety. The original model and constraints of an electric boiler satisfy the following relationship:

[0103]

[0104] In the formula, This represents the heat power generated by the i-th EB unit during time period t. μ represents the electrical conversion efficiency of the i-th EB unit. Loss This indicates the heat loss efficiency of an electric boiler.

[0105] Let represent the lower limit and upper limit of the power consumption of the i-th EB unit, respectively. Let represent the maximum downhill ramp rate and the maximum uphill ramp rate of the i-th EB unit, respectively.

[0106] The polyhedral model in a compiler is an efficient program optimization technique that maps complex circular dependencies to a high-dimensional geometric space, thereby achieving parallelization and locality optimization of computational tasks during the compilation phase. By constructing and manipulating polyhedral representations, instructions and data access can be effectively scheduled to reduce resource contention and cache misses, thus improving program execution performance. Therefore, different original models and constraints of various types of devices, as well as the power balance constraints of a comprehensive energy system, are used as input data to the compiler; the compiler outputs a set of original polyhedral models. in, Let be the constraint matrix of the i-th primitive polyhedron model of the m-th type of device. Let x be the constraint vector of the i-th primitive polyhedral model of the m-th type of device, x be the operating parameters under different operating conditions, m = 1, 2, ..., M, where M is the total number of device types in the integrated energy system, and i = 1, 2, ..., N, where N is the total number of primitive polyhedral models; primitive polyhedral models The i-th original model and constraints corresponding to the m-th type of device are used to obtain the polyhedral model set, which includes all the original models of various types of devices in the integrated energy system and has a unified format.

[0107] Step 1.3: Calculate the mean value of the original polyhedral models of the same type of device to obtain the basic polyhedral models of each type of device. The following relationship must be satisfied:

[0108]

[0109] In the formula, Let be the constraint matrix of the basic polyhedral model of the m-th type of device. Let be the constraint vector of the basic polyhedral model of the m-th type of device. These are the average values ​​of operating parameters under different working conditions;

[0110] Basic polyhedral model The basic model and basic constraints corresponding to the m-th type of device;

[0111] By compiling all the original models of various types of devices in the integrated energy system into polyhedral models with the same format, and using the average value of constraint matrices and constraint vectors to calculate the basic models of various types of devices instead of using the average value of operating parameters under different operating conditions, the establishment of basic models can be achieved more simply and quickly.

[0112] Step 1.4: In the same type of device, with the optimization objective of having the scaled and translated base polyhedron model contain the original polyhedron model and have the smallest scaling factor, determine the scaling factor and translation vector of each original polyhedron model.

[0113] The basic polyhedron model after scaling and translation satisfies the following relationship:

[0114]

[0115] In the formula, Let be the scaling factor for the i-th primitive polyhedron model of the m-th type of device. Let be the translation vector of the i-th original polyhedral model of the m-th type of device;

[0116] Translation vectors are used to adjust the position of the polyhedral model to match the dynamic characteristics of the operating state of different types of devices as time changes.

[0117] Step 1.5: In the same type of device, establish a mapping model and constraint conditions between the constraint matrix of the original polyhedron model and the constraint matrix of the basic polyhedron model, satisfying the following relationship:

[0118]

[0119] In the formula, U m It is the mapping matrix between the constraint matrix of the original polyhedral model and the constraint matrix of the basic polyhedral model in the m-th type of device;

[0120] By establishing the above mapping model and constraints, the linear mapping relationship between the original polyhedral model and the basic polyhedral model is guaranteed, ensuring that the final aggregated model can cover all the original models.

[0121] Step 1.6: In the same type of device, when the mapping model and constraint conditions are met, calculate the average value of the scaling factor of all original polyhedral models, and use it as the scaling factor of the device of that type; update the constraint vector of the basic polyhedral model using the scaling factor of the device of that type and the total number of original polyhedral models.

[0122] The scaling factors for each type of device satisfy the following relationship:

[0123]

[0124] In the formula, Let m be the scaling factor for the m-th type of device;

[0125] The updated constraint vectors of the basic polyhedron model satisfy the following relationship:

[0126]

[0127] In the formula, The updated constraint vectors for the basic polyhedron model;

[0128] Step 1.7: For devices of the same type, construct the aggregated polyhedral model of that type of device using the constraint matrix of the basic polyhedral model and the updated constraint vectors. The aggregate polyhedral model corresponds to the aggregate model and aggregate constraints of the m-th type of device; the set of aggregate polyhedral models for the integrated energy system is as follows: M represents the total number of device types in the integrated energy system.

[0129] In this embodiment, taking an electric energy storage device as an example, the original model and constraints of the electric energy storage device are converted into an original polyhedral model. Among the variables This represents the energy storage capacity of the e-th energy storage device during time period t. These represent the charging and discharging power of the e-th energy storage device during time period t, respectively. These are 0-1 variables, representing the charging and discharging states of the e-th energy storage device during time period t, respectively.

[0130] The original polyhedral model of capacity is represented as:

[0131]

[0132] The original polyhedral model of charge and discharge power is represented as:

[0133]

[0134] The original polyhedral model of charge-discharge mutual repulsion is represented as:

[0135]

[0136] The original polyhedral model of the energy storage device In, the following relationship is satisfied:

[0137]

[0138] The aggregated polyhedral model set for integrated energy systems includes dynamic aggregated models of various types of devices within the integrated energy system. It achieves the aggregation of original models for devices of the same type, realizing the dynamic aggregation of the integrated energy system and facilitating scheduling optimization based on the dynamic aggregated model. In existing technologies, the aggregation of multiple models in integrated energy systems involves strategically combining the prediction results of multiple models. However, due to potential differences between models at different management levels within the integrated energy system, the method of aggregating prediction results can affect the prediction accuracy. To avoid the impact of model differences on the prediction accuracy of integrated energy systems, this invention proposes a dynamic aggregation of original models for devices of the same type based on polyhedral model compilation technology. This allows for broader integration and aggregation with models at different management levels, thereby significantly improving the system's prediction accuracy.

[0139] Step 2: Generate original scenes based on the aggregation model of each type of device, and filter the original scenes based on the aggregation constraints of each type of device; reduce the filtered scenes according to the similarity index and representativeness index of the scenes to obtain a set of typical scenes.

[0140] A high proportion of renewable energy exists in integrated energy systems, and renewable energy, especially wind power, has seen its share of total energy consumption gradually increase in recent years due to its clean and sustainable characteristics. However, wind power output is highly random and intermittent, with significant fluctuations, posing challenges to the security and reliability of the power grid, particularly with the rapid development of distributed energy resources. Therefore, considering the uncertainty of renewable energy output is of great practical significance. This invention proposes an aggregation model and aggregation constraints based on various types of devices, employing Latin hypercube sampling and scenario filtering and reduction methods to obtain a representative set of scenarios to account for the uncertainties of renewable energy and load.

[0141] Specifically, step 2 includes:

[0142] Step 2.1: Generate prediction data using the aggregation model of each type of device; based on the probability distribution of the prediction data, use sampling to obtain prediction samples to construct different original scenes; filter the original scenes using the aggregation constraints of each type of device.

[0143] In this embodiment, a convergence model of the wind power generation device is used to predict wind power generation. Preferably, the predicted wind power generation data follows a normal distribution. The predicted wind power generation data is used as the mean, and one-tenth of the predicted value with fluctuation characteristics is used as the standard deviation σ. To ensure that the generated scenario matches the system's energy storage regulation capability and to avoid excessive fluctuations affecting system stability, the standard deviation of the wind power output prediction is dynamically corrected based on the dispatchability capability of the convergence model. This makes the energy storage and load regulation capabilities match the wind power fluctuations. The adjustment formula is as follows: Where, N agg N chldis The average number of alternating charge and discharge cycles and the standard value of the energy storage device in the aggregation model are respectively used. The adjusted standard deviation σ′ is used as the standard deviation to construct the probability distribution function of wind power output, thereby characterizing the uncertainty of wind power generation. This modeling method based on normal distribution can better reflect the random characteristics of wind power output, providing a foundation for subsequent scenario simulation and optimization.

[0144] Latin hypercube sampling (LHS) is an efficient numerical experimental design method that can uniformly cover the probability distribution interval of random variables. By dividing the distribution of each variable into several sub-intervals and randomly sampling points within each sub-interval, LHS avoids sample clustering and significantly reduces variance and sampling error. Compared with traditional Monte Carlo methods, LHS can achieve high simulation accuracy with a small sample size and is suitable for modeling arbitrary distribution types and multidimensional random variables. After obtaining the probability distribution of wind power output, a large number of samples conforming to the pattern are generated using LHS. The resulting original scenarios effectively reflect the overall characteristics. The original scenario set is Ws (each column represents a scenario, and each row represents a predicted value), and each scenario has the same probability, satisfying the following relationship:

[0145]

[0146] In the formula, p w Let w be the probability of scene w, and W be the total number of original scenes.

[0147] After obtaining the probability distribution of wind power output, a large number of samples that conform to the pattern are generated through LHS. However, considering that there may be invalid scenarios in the samples that exceed the adjustment capacity, invalid data is filtered according to the aggregation constraint conditions after generating a large number of samples to achieve the filtering of invalid scenarios.

[0148] In this embodiment, the wind power generation device uses the following aggregation constraint condition for filtering:

[0149] Where ΔP wind (s, t) represents the fluctuating wind power output during time period t in scenario s. Let be the maximum upward climbing rate in the aggregation constraint, and Δt be the time interval for scheduling optimization. This represents the maximum instantaneous regulated power in the aggregate constraints. The aggregate constraints fully reflect the dynamic behavior of the device, including capacity dynamics, ramp limits, and charging state mutual exclusion, thus effectively filtering out invalid scenarios.

[0150] Step 2.2: Reduce the number of filtered scenes based on scene similarity and representativeness indicators;

[0151] Specifically, step 2.2 includes:

[0152] 1) The similarity index between two scenarios satisfies the following relationship:

[0153] x(i,j)=∑|Ws(i)-Ws(:,j)]

[0154] In the formula, x(i,j) is the distance between scene i and typical scene j, and Ws(:,i) and Ws(:,j) are the data of the i-th scene and the j-th scene in the original scene set Ws, respectively;

[0155] The closer two scenes are, the lower their similarity index.

[0156] 2) Calculate the average similarity index y0(i) between each scene and the other scenes as a representative index for each scene, satisfying the following relationship:

[0157]

[0158] In the formula, x(:,i) is the similarity index between scene i and the other scenes;

[0159] The value of y0(i) reflects the representativeness of scenario i to the other scenarios. The smaller the value, the worse the representativeness of scenario i to the other scenarios.

[0160] 3) The scenario with the smallest representativeness index is designated as the first scenario, and the scenario with the smallest similarity index to the first scenario is designated as the second scenario. The sum of the probabilities of the first scenario and the second scenario is used as the updated probability of the second scenario, satisfying the following relationship:

[0161] p′ r =p r +p d

[0162] In the formula, p′ r p represents the updated probability for the second scene. r p represents the probability of the second scenario. d The probability of the first scenario;

[0163] 4) Reduce the first scene from the scene after the first reduction to obtain the updated scene set; repeat steps 1) to 4) for the updated scene set until the total number of scenes in the updated scene set reaches the set value, and output the updated scene set as the typical scene set.

[0164] The above steps are iterated continuously until the number of remaining scenarios equals the target number. The scenarios obtained in this way are more representative and can well take into account the uncertainty of new energy output. Moreover, the probability distribution is readjusted each time a scenario is reduced, ensuring that the total probability of the reduced scenarios remains unchanged.

[0165] Step 3: Based on the typical scenario operating parameters generated by the aggregation model and aggregation constraints, the objective function for day-ahead optimization scheduling is to minimize the sum of energy purchase cost, operation and maintenance cost, and penalty cost of all typical scenarios; the objective function for intraday rolling scheduling is to minimize the sum of changes in energy purchase cost, operation and maintenance cost, and penalty cost of all typical scenarios. The system scheduling scheme is obtained by solving the objective functions for day-ahead optimization scheduling and intraday rolling scheduling under the operating constraints of each type of device and the power balance constraints of the integrated energy system.

[0166] Specifically, step 3 includes:

[0167] Step 3.1: Based on the operating parameters generated under each typical scenario according to the aggregation model and aggregation constraints, the objective function for day-ahead optimization scheduling is to minimize the sum of energy purchase cost, operation and maintenance cost and penalty cost under all typical scenarios.

[0168] After modeling and aggregating the various devices in the integrated energy system, day-ahead optimization scheduling begins based on the aggregation model. The scheduling cycle is 24 hours with a time interval of 1 hour, aiming at optimal daily operating economy. A typical scenario set is used to characterize the uncertainty of wind power output and load demand. The objective function satisfies the following relationship:

[0169]

[0170] In the formula, F ahead Let P be the objective function for optimizing the current schedule. a Let represent the probability of typical scenario 'a', n represent the total number of typical scenarios, EOC is the energy purchase cost, OMC is the operation and maintenance cost, WCP is the penalty cost, and in the example, it is the wind curtailment penalty cost.

[0171] Based on the aggregation models and constraints of various types of equipment, the electricity and gas purchase quantities for each time period are determined, and the energy purchase cost is calculated using the following formula:

[0172]

[0173] In the formula, λ e，t , λ g，t The time-of-use electricity price and gas price for period t are respectively, P g,t G represents the system's power purchase capacity during time period t. g,t Let T be the amount of natural gas purchased by the system during time period t, where T is the total number of time periods.

[0174] Based on the aggregation model and aggregation constraints of various types of devices, the charging and discharging power of energy storage devices and the operating power of energy conversion devices in each time period are determined, and the operation and maintenance costs are calculated using the following formula:

[0175]

[0176] In the formula, β k Let β be the unit charge / discharge cost of the k-th type of energy storage device. l The unit operating cost of the j-th type of energy conversion device, P represents the charging and discharging power of the k-th type of energy storage device during time period t. i,t Let N be the operating power of the j-th type of energy conversion device during time period t. S N represents the total number of energy storage device types. C The total number of energy conversion device types;

[0177] The penalty cost, including the cost of wind curtailment, satisfies the following relationship:

[0178]

[0179] In the formula, k w ΔP is the wind curtailment penalty coefficient. c,t The wind curtailment power during time period t.

[0180] Step 3.2: Based on the operating parameters generated under each typical scenario according to the aggregation model and aggregation constraints, the objective function for intraday rolling scheduling is to minimize the sum of changes in energy purchase cost, operation and maintenance cost and penalty cost under all typical scenarios.

[0181] The intraday optimization model characterizes the uncertainties of wind power and load demand at short time scales within a typical intraday scenario set. With the goal of minimizing intraday energy purchase cost, power adjustment cost, and curtailment cost, and with constraints on the operation of each device in the integrated energy system, power balance constraints, curtailment constraints, and energy purchase constraints, the model solves the scheduling scheme of each device with a 4-hour control time domain and a 15-minute time scale, and performs rolling optimization to obtain the intraday optimal scheduling plan.

[0182] The objective function for intraday rolling scheduling satisfies the following relationship:

[0183]

[0184] In the formula, F inday Let P be the objective function for intraday rolling scheduling. s Let represent the probability of typical scenario s, n represent the total number of typical scenarios, EPC represent energy purchase cost, and ΔOMC represent the change in operation and maintenance cost.

[0185] WCP stands for penalty cost, which in this example is the wind curtailment penalty cost.

[0186] Based on the aggregation models and constraints of various types of equipment, the electricity and gas purchase quantities for each time period are determined, and the energy purchase cost is calculated using the following formula:

[0187]

[0188] In the formula, λ e，t , λ g，t The time-of-use electricity price and gas price for period t are respectively, P g，t G represents the system's power purchase capacity during time period t. g，t Tinday represents the amount of natural gas purchased by the system during time period t, where Tinday is the total number of rolling time periods within the day.

[0189] Based on the aggregation model and aggregation constraints of various types of devices, the changes in charging and discharging power of energy storage devices and the changes in operating power of energy conversion devices are determined for each time period. The changes in operation and maintenance costs are then calculated using the following formula:

[0190]

[0191] In the formula, γ k γ l These are the power change penalty cost coefficients for the k-th type energy storage device and the j-th type energy conversion device, respectively. Let ΔP be the change in charging and discharging power of the k-th type of energy storage device during time period t. l，t Let N be the change in operating power of the j-th type of energy conversion device during time period t. S N represents the total number of energy storage device types. C This represents the total number of energy conversion device types, while Tinday represents the total number of rolling periods within the day.

[0192]

[0193] In the formula, k w ΔP is the wind curtailment penalty coefficient. c，t The curtailment power is the wind power during time period t, and Tinday is the total number of rolling time periods within the day.

[0194] Step 3.2: Use the operational constraints of each type of device and the power balance constraints of the integrated energy system as constraints for day-ahead optimization scheduling and intraday rolling scheduling;

[0195] The power balance constraints of the integrated energy system include:

[0196] 1) Power balance constraint:

[0197] P g，t +P chp，t +P wind，t +P pv,t =P p2g，t +P load，t +P es,t +·P eb,t

[0198]

[0199]

[0200] 2) Gas power balance constraint

[0201]

[0202] 3) Thermal power balance constraint

[0203] η eb (1-μ Loss )P eb,t +P chp,t / r chp =P HL,t +H es,t

[0204] In the formula, P g，t Let G be the power purchased from the grid during time period t. g，t H represents the amount of natural gas purchased by the system during time period t. ng P represents the calorific value of natural gas. chp,t P represents the discharge power of the CHP unit during time period t. wind，t P pv，t These represent the wind power and photovoltaic power generation during time period t, respectively; P p2g，t P represents the aggregated power consumption of the P2G unit during time period t. load，t P GL，t P HL，t P represents the electricity, gas, and heat power load demand during time period t. eb，t P represents the power consumption of the polymer electric boiler during time period t. es，t G es，t II es，t These represent the energy storage and release power of electricity, gas, and heat during time period t, respectively. These are the lower and upper limits of the power consumption of the P2G unit, respectively. These are the maximum downward ramp rate and the maximum upward ramp rate of the aggregated P2G unit, respectively. These are the lower and upper limits of the power consumption of the polymer EB unit, respectively; These are the maximum downward ramp rate and maximum upward ramp rate of the polymer EB unit, respectively; η p2g To aggregate the efficiency of P2G units, η chp The gas-to-electric conversion efficiency η of the CHP unit eb For the efficiency of polymer electric boilers, μ Loss r is the heat transfer loss rate. chp The electrothermal power ratio of the CHP polymerization unit.

[0205] Step 3.3: Under the operational constraints of each type of device and the power balance constraints of the integrated energy system, solve the objective functions of day-ahead optimization scheduling and intraday rolling scheduling to obtain the system scheduling scheme;

[0206] In this embodiment, based on the results of day-ahead optimization scheduling and the dynamic parameters of the aggregation model, various loads and powers within the day are updated and predicted. The unified polyhedral constraints generated in the aggregation model are directly used in the day-ahead optimization problem, replacing the original independent constraints of the distributed equipment. The typical scenarios obtained through Latin hypercube sampling and scenario reduction are used to characterize the uncertainty of wind power and load. Model aggregation reduces the scale of the optimization problem while preserving the characteristics of the equipment, making day-ahead scheduling optimization more efficient. Through the day-ahead stochastic optimization model, the optimal scheduling plan for all equipment within the day can be obtained.

[0207] This embodiment provides a specific calculation example, through... Figure 2 A simulation analysis was performed on the integrated energy system architecture shown. Current renewable energy output and various load forecasts are available in [link to data]. Figure 3 and Figure 4 The time-of-use electricity price is shown in Table 1, and the purchase price of natural gas is 5 yuan / m3.

[0208] Table 1 Time-of-use electricity prices

[0209] peak 11.00-15.00；18.00-21.00 1.35 flat 8.00-10.00；16.00-17.00；22.00-24.00 0.87 valley 0.00-7.00 0.42

[0210] The parameters of the aggregation models for each type of device in the example are shown in Tables 2 and 3.

[0211] Table 2 Parameters of the aggregation model for various coupling devices

[0212]

[0213] Table 3 Parameters of the aggregation model for various energy storage devices

[0214] Storage battery 200-1000 100 / 100 0.85 / 0.75 gas tank 200-1000 100 / 100 0.98 / 0.98 thermal storage tank 200-1000 100 / 100 0.98 / 0.98

[0215] Furthermore, initial scenes were generated through Latin hypercube sampling, and typical scenes were extracted using scene reduction techniques. Their distribution characteristics provided crucial references for subsequent scheduling optimization. Based on these typical scenes, Figures 5 to 7 The day-ahead and intraday dispatch output curves of various energy devices are displayed. Optimized dispatching must simultaneously meet constraints related to equipment operation, power balance, curtailment, and energy purchase within the integrated energy system. Day-ahead dispatching aims for optimal operational economy by optimizing the overall power allocation of energy devices within a day, while intraday dispatching aims to minimize intraday energy purchase costs, power adjustment costs, and curtailment costs by dynamically optimizing equipment power output.

[0216] This invention also proposes a dynamic aggregation and optimization scheduling system for integrated urban energy systems, comprising:

[0217] The dynamic aggregation module is used to obtain the operating parameters of various types of devices in the integrated energy system under different operating conditions, so as to establish different original models and constraints of each type of device; based on the polyhedral model, different original models and constraints of the same type of device are aggregated into an aggregated model and aggregated constraints of that type of device.

[0218] The optimization scheduling module generates original scenarios based on the aggregation model of each type of device, and filters the original scenarios based on the aggregation constraints of each type of device. Based on the similarity and representativeness indicators of the scenarios, it reduces the filtered scenarios to obtain a set of typical scenarios. Based on the typical scenario operating parameters generated by the aggregation model and aggregation constraints, it sets the objective function for day-ahead optimization scheduling as minimizing the sum of energy purchase cost, operation and maintenance cost, and penalty cost of all typical scenarios, and the objective function for intraday rolling scheduling as minimizing the sum of changes in energy purchase cost, operation and maintenance cost, and penalty cost of all typical scenarios. The system scheduling scheme is obtained by solving the objective functions for day-ahead optimization scheduling and intraday rolling scheduling under the operating constraints of each type of device and the power balance constraints of the integrated energy system.

[0219] This disclosure can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of this disclosure.

[0220] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0221] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0222] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0223] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for dynamic aggregation and optimal scheduling of integrated energy systems in urban areas, characterized in that, include: Obtain the operating parameters of various types of devices in the integrated energy system under different operating conditions, and establish different original models and constraints for each type of device; Based on the polyhedral model, the compiler takes different original models and constraints of various types of devices, as well as the power balance constraints of the integrated energy system, as input data; and outputs a set of original polyhedral models using the compiler. ,in, For the first The first type of device The constraint matrix of the original polyhedral model. For the first The first type of device The constraint vector of the original polyhedral model. Operating parameters under different working conditions, , This represents the total number of device types in an integrated energy system. , The total number of original polyhedral models; original polyhedral models Corresponding to the The first type of device The original model and constraints are determined; the mean value of the set of original polyhedral models of the same type of device is calculated to obtain the first... Basic polyhedral model of the device , For the first The constraint matrix of the basic polyhedral model of the device. For the first The constraint vectors of the basic polyhedral model of the device. The average values ​​of operating parameters under different working conditions; basic polyhedral model It satisfies the following relationship: ， In the formula, For the first The constraint matrix of the basic polyhedral model of the device. For the first The constraint vectors of the basic polyhedral model of the device. The average values ​​of operating parameters under different working conditions; basic polyhedral model Corresponding to the The basic model and basic constraints of the device; In the same type of device, the basic polyhedral model after scaling and translation satisfies the following relationship: In the formula, For the first The first type of device The scaling factor of the original polyhedron model. For the first The first type of device Translation vectors of the original polyhedral model; The optimization objective is to minimize the scaling factor while ensuring that the scaled and translated base polyhedron model contains the original polyhedron model. The scaling factor and translation vector of each original polyhedron model are determined. For devices of the same type, a mapping model and constraints are established between the constraint matrices of the original polyhedron models and the constraint matrices of the base polyhedron models. For devices of the same type, when the mapping model and constraints are satisfied, the average scaling factor of all original polyhedron models is calculated and used as the scaling factor for that type of device. The constraint vector of the base polyhedron model is updated using the scaling factor of that type of device and the total number of original polyhedron models. For devices of the same type, the aggregated polyhedron model of that type of device is constructed using the constraint matrix of the base polyhedron model and the updated constraint vector. The aggregate polyhedral model set of integrated energy systems is as follows: , The updated constraint vectors for the basic polyhedron model; the corresponding constraint vectors for the aggregated polyhedron model. The aggregation model and aggregation constraints of the same type of device; for devices of the same type, establish the mapping model and constraints between the constraint matrix of the original polyhedral model and the constraint matrix of the basic polyhedral model, satisfying the following relationship: In the formula, For the first The mapping matrix between the constraint matrix of the original polyhedral model and the constraint matrix of the basic polyhedral model in the device class; The scaling factors for each type of device satisfy the following relationship: In the formula, For the first Scaling factor for the device class; The updated constraint vectors of the basic polyhedron model satisfy the following relationship: In the formula, The updated constraint vectors for the basic polyhedron model; The original scenarios are generated based on the aggregation model of each type of device. The original scenarios are then filtered based on the aggregation constraints of each type of device. The filtered scenarios are reduced to a set of typical scenarios based on the similarity and representativeness indicators of the scenarios. The objective function for day-ahead optimization scheduling is to minimize the sum of energy purchase cost, operation and maintenance cost, and penalty cost of all typical scenarios. The objective function for intraday rolling scheduling is to minimize the sum of the changes in energy purchase cost, operation and maintenance cost, and penalty cost of all typical scenarios. The system scheduling scheme is obtained by solving the objective functions for day-ahead optimization scheduling and intraday rolling scheduling under the operating constraints of each type of device and the power balance constraints of the integrated energy system.

2. The method for dynamic aggregation and optimized scheduling of urban area integrated energy systems according to claim 1, characterized in that, Different types of devices within an integrated energy system include: wind power generation devices, photovoltaic power generation devices, electric energy storage devices, gas energy storage devices, thermal energy storage devices, electric-to-gas generator units, CHP units, and electric boilers; The original model includes a power balance model; constraints include, but are not limited to, capacity constraints, power constraints, and state constraints.

3. The method for dynamic aggregation and optimized scheduling of urban area integrated energy systems according to claim 1, characterized in that, Predictive data is generated using aggregated models of various device types; based on the probability distribution of the predicted data, a sampling method is used to obtain predicted samples to construct different original scenarios; the original scenarios are filtered using aggregated constraints of various device types. Based on scene similarity and representativeness metrics, the filtered scenes are further reduced, including: 1) The similarity index between two scenarios satisfies the following relationship: In the formula, Let be the distance between scene i and typical scene j. , These are the data for the i-th scene and the j-th scene in the original scene set Ws, respectively. 2) Calculate the average similarity index between each scene and the other scenes. As representative indicators for various scenarios, they satisfy the following relationship: In the formula, This is a similarity metric between scene i and the other scenes; This represents the total number of original scenes; 3) The scenario with the smallest representativeness index is designated as the first scenario, and the scenario with the smallest similarity index to the first scenario is designated as the second scenario. The sum of the probabilities of the first scenario and the second scenario is used as the updated probability of the second scenario, satisfying the following relationship: In the formula, This represents the updated probability for the second scenario. Let be the probability of the second scenario. The probability of the first scenario; 4) Reduce the first scene from the scene after the first reduction to obtain the updated scene set; repeat steps 1) to 4) for the updated scene set until the total number of scenes in the updated scene set reaches the set value, and output the updated scene set as the typical scene set.

4. The method for dynamic aggregation and optimized scheduling of urban area integrated energy systems according to claim 1, characterized in that, The objective function for optimizing scheduling currently satisfies the following relationship: In the formula, The objective function for optimizing the current schedule is... Let represent the probability of typical scenario 'a', and let n represent the total number of typical scenarios. To purchase energy costs, For maintenance costs, To incur penalties and costs.

5. The method for dynamic aggregation and optimized scheduling of urban area integrated energy systems according to claim 4, characterized in that, Based on the aggregation models and constraints of various types of equipment, the electricity and gas purchase quantities for each time period are determined, and the energy purchase cost is calculated using the following formula: In the formula, , These represent the time-of-use electricity price and gas price for period t, respectively. The system's power purchase capacity during time period t. Let T be the amount of natural gas purchased by the system during time period t, where T is the total number of time periods. Based on the aggregation model and aggregation constraints of various types of devices, the charging and discharging power of energy storage devices and the operating power of energy conversion devices in each time period are determined, and the operation and maintenance costs are calculated using the following formula: In the formula, The unit charge / discharge cost of the k-th type of energy storage device, The unit operating cost of the j-th type of energy conversion device, , Let be the charging and discharging power of the k-th type of energy storage device during time period t. Let be the operating power of the j-th type of energy conversion device during time period t. This represents the total number of energy storage device types. The total number of energy conversion device types; The penalty cost, including the cost of wind curtailment, satisfies the following relationship: In the formula, This is the wind curtailment penalty coefficient. Let T represent the wind curtailment power during time period t, where T is the total number of time periods.

6. The method for dynamic aggregation and optimized scheduling of urban area integrated energy systems according to claim 1, characterized in that, The objective function for intraday rolling scheduling satisfies the following relationship: In the formula, The objective function for intraday rolling scheduling is... Let represent the probability of typical scenario s, and n represent the total number of typical scenarios. To purchase energy costs, This represents the change in operation and maintenance costs. To incur penalties and costs.

7. The method for dynamic aggregation and optimal scheduling of urban area integrated energy systems according to claim 6, characterized in that, Based on the aggregation models and constraints of various types of equipment, the electricity and gas purchase quantities for each time period are determined, and the energy purchase cost is calculated using the following formula: In the formula, , These represent the time-of-use electricity price and gas price for period t, respectively. The system's power purchase capacity during time period t. T represents the amount of natural gas purchased by the system during time period t. inday This represents the total number of rolling periods within the day; Based on the aggregation model and aggregation constraints of various types of devices, the changes in charging and discharging power of energy storage devices and the changes in operating power of energy conversion devices are determined for each time period. The changes in operation and maintenance costs are then calculated using the following formula: In the formula, , These are the power change penalty cost coefficients for the k-th type energy storage device and the j-th type energy conversion device, respectively. , Let be the changes in charging and discharging power of the k-th type of energy storage device during time period t. Let be the change in operating power of the j-th type of energy conversion device during time period t. This represents the total number of energy storage device types. The total number of energy conversion device types; The penalty cost, including the cost of wind curtailment, satisfies the following relationship: In the formula, This is the wind curtailment penalty coefficient. Let T be the wind curtailment power during time period t. inday This represents the total number of rolling periods within the day.

8. A dynamic aggregation and optimal scheduling system for an urban area integrated energy system, used to implement the dynamic aggregation and optimal scheduling method for an urban area integrated energy system as described in any one of claims 1 to 7, characterized in that, include: The dynamic aggregation module is used to obtain the operating parameters of various types of devices in the integrated energy system under different operating conditions, so as to establish different original models and constraints for each type of device. Based on the polyhedral model, different original models and constraints of the same type of device are aggregated into an aggregated model and aggregated constraints of that type of device. The optimization scheduling module generates original scenarios based on the aggregation model of each type of device, and filters the original scenarios based on the aggregation constraints of each type of device. Based on the similarity and representativeness indicators of the scenarios, it reduces the filtered scenarios to obtain a set of typical scenarios. Based on the typical scenario operating parameters generated by the aggregation model and aggregation constraints, it sets the objective function for day-ahead optimization scheduling as minimizing the sum of energy purchase cost, operation and maintenance cost, and penalty cost of all typical scenarios, and the objective function for intraday rolling scheduling as minimizing the sum of changes in energy purchase cost, operation and maintenance cost, and penalty cost of all typical scenarios. The system scheduling scheme is obtained by solving the objective functions for day-ahead optimization scheduling and intraday rolling scheduling under the operating constraints of each type of device and the power balance constraints of the integrated energy system.

9. A terminal, comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1-7.

Images