A method for interconnected and coordinated operation of a multi-region integrated energy system

Abstract

The application discloses a kind of interconnection collaborative operation methods of multi-region integrated energy system, specifically related to the field of integrated energy management, by responding to multi-region integrated energy system collaborative scheduling instruction, obtain each regional energy equipment, energy storage and renewable energy related data, construct the industrial mechanism framework containing lagrange multiplier, time delay and multi-energy coupling constraint, coupled internal and external characteristics solve to generate collaborative scheduling instruction.A kind of interconnection collaborative operation methods of multi-region integrated energy system by establishing the unified industrial mechanism framework covering the energy cascade utilization and multi-energy complementary law of whole system, realizes the collaborative promotion of whole system multi-energy complementary and global economy;By partial differential equation to describe heat transfer, the adaptability of scheduling strategy and actual physical working condition is greatly improved;By improved augmented lagrange function, further improve the global optimality and constraint compliance of scheduling strategy.

Description

Technical Field

[0001] This invention relates to the field of integrated energy management technology, and more specifically, to a method for interconnected and coordinated operation of a multi-regional integrated energy system. Background Technology

[0002] With the development of integrated energy management technology, improving the operational efficiency of multi-energy systems and achieving cross-regional energy optimization have become core technological requirements for building new energy systems. Integrated energy management is used to coordinate the production and conversion of multiple energy sources. Its technological implementation highly relies on constructing an industrial mechanism model covering the coupling relationships of multiple energy flows as a decision-making basis, and forming an information processing and collaborative control architecture specifically suited for the purpose of energy system operation and management. Multi-regional integrated energy systems, through the complementarity of multiple energy flows within the region and the coordination between regions, provide operators with digital management and collaborative operation support, and represent a key direction for current energy management technology innovation.

[0003] To address existing management needs, the industry currently primarily adopts distributed optimization-based independent operation control technology as the mainstream solution: treating each region as an independent autonomous management entity, each region constructs an optimization model based on local forecasts with the goal of minimizing its own costs or maximizing local energy efficiency, and iteratively solves the problem until the boundary converges, forming a joint solution acceptable to all regional management entities, thus initially achieving basic energy supply and demand balance and preliminary collaborative management.

[0004] However, in actual use, it still has some shortcomings. For example, the existing scheme breaks down the global problem into independent sub-problems, and each region only focuses on its own interests. It lacks unified management consideration of overall economic efficiency and carbon emissions. The root cause is that a unified industrial mechanism framework reflecting the energy cascade utilization of the whole system has not been established, which makes it impossible to effectively implement the global resource allocation goals. Furthermore, relying solely on pre-defined connection plans for boundary coupling fails to model physical constraints such as cross-regional transmission delays and network capacity in conjunction with regional internal characteristics. The fundamental flaw lies in the lack of a unified mechanistic model to characterize the aforementioned physical processes, significantly reducing the executability of management and scheduling strategies. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a method for interconnected and coordinated operation of a multi-regional integrated energy system, which solves the problems mentioned in the background art through the following scheme.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A method for interconnected and coordinated operation of a multi-regional integrated energy system includes: S1: In response to the coordinated operation and scheduling instructions of the multi-region integrated energy system, obtain the energy conversion equipment model, energy storage dynamic characteristics and renewable energy output prediction curves within each region; S2: Based on the acquired data from various regions, construct an industrial mechanism framework reflecting the energy cascade utilization and multi-energy complementarity of the entire system, including: A set of initial fields of Lagrange multipliers is generated, which are used to quantify the marginal contribution of each region to the global comprehensive benefit at different time sections. Using the initial fields of Lagrange multipliers as weighting coefficients, a set of cross-regional physical constraints containing transmission time delay equations in the form of partial differential equations is constructed. The cross-regional transmission delay time of electrical energy and thermal energy is added as a continuous variable to the optimization model. Combined with the cross-regional multi-energy coupling constraints constructed by the energy hub matrix, an industrial mechanism model containing transmission delay, capacity limitation, and multi-energy flow nonlinearity is formed, and an improved augmented Lagrangian function is output. S3: Couple the internal operating characteristics of each region with the dynamic response characteristics of the cross-regional network to the industrial mechanism framework, and solve to generate collaborative operation scheduling instructions.

[0007] Preferably, S2 is the marginal reduction in the overall global benefit resulting from increasing the boundary exchange power of the i-th region at time t by one unit. The quantification of the marginal contribution specifically includes: Solve the single-region optimization problem for each region to obtain the local optimal multipliers for each region. ; Substituting into the collaborative optimization objective function F, the initial field is obtained through one Newton step correction. ,in Step size factor This represents the gradient of the global residual with respect to the multipliers.

[0008] Preferably, step S2, treating the cross-regional transmission delay time of electrical energy as a continuous variable, specifically includes: When the distance between regions exceeds a preset threshold, the power transmission delay time will be increased. As a fixed parameter added to the optimization model, where c is the speed of light, This refers to the length of the power transmission line.

[0009] Preferably, step S2, which involves constructing a cross-regional physical constraint set for the cross-regional thermal energy transfer network, specifically includes: The temperature of the heat transfer medium is described by a one-dimensional convection-diffusion partial differential equation. Dynamic changes along the pipeline at position x and time t: in, For flow rate, For ambient temperature, Where is the thermal diffusivity, For the thermal conductivity of the heat transfer medium, The equivalent heat dissipation coefficient, It is a constant. For pipe diameter, The overall heat transfer coefficient is... For cross-section, For heat medium density, Specific heat capacity of the heat transfer medium; The partial differential equation is expressed in space step size and time step Discretize the above, and transform it into a system of linear algebraic equations describing the relationship between the temperatures of each discrete node in the pipeline; Pipe inlet temperature sequence With outlet temperature sequence The dynamic mapping relationship between them is added as a set of equality constraints to the cross-regional physical constraint set, where Transmission delay time and flow rate As a decision variable, This refers to the length of the pipe. The discretized linear constraints are compared with the upper limit of the pipeline's transmission capacity. Lower limit Together they constitute the physical constraint set of the cross-regional heat transfer network.

[0010] Preferably, S2, the construction of cross-regional multi-energy coupling constraints, specifically includes: For a cross-regional energy conversion device connecting the first and second regions, obtain the mathematical model of the device's energy hub, using matrix equations. Characterization, where the input vector u contains at least the input electrical power and the input gas power, the output vector y contains at least the output gas power and the output thermal power, and H is an energy hub matrix containing the efficiency coefficients and energy distribution coefficients of each energy conversion device; Each component of the matrix equation is expanded into a linear equation constraint and added to the cross-regional physical constraint set; Based on the physical rating of the energy conversion device, upper and lower limit constraints are set for each component in the input vector u and the output vector y, and these upper and lower limit constraints are added to the cross-regional physical constraint set.

[0011] Preferably, in S2, the improved augmented Lagrangian function L is defined as: The collaborative optimization objective function F is integrated with the cross-regional physical constraint set, specifically expressed as: Where de is the set of all decision variables. Let represent the decision variable for the i-th region at time t, and JD be the index set of all constraints. Let jd be the residual function of the jd-th constraint. For Lagrange multipliers, start directly from the initial field of the Lagrange multipliers. Select and map from, This is the penalty parameter.

[0012] Preferably, in step S2, the improved augmented Lagrange function L is solved iteratively, and the local decision variables of each region are updated in parallel in each iteration, the boundary information of adjacent regions is exchanged, and the Lagrange multipliers are updated. During the iteration process, the ratio of the original residual to the dual residual is monitored in real time. ,when Add penalty parameters To strengthen constraint satisfaction, when Time decrease To promote goal optimization, otherwise maintain The original residual and the dual residual remain unchanged until both are less than the preset convergence tolerance. in, For the original residual, Let k be the dual residual and kn be the number of iterations.

[0013] Preferably, S3, the generation of the cooperative operation scheduling instruction, specifically includes: The obtained global optimal solution is decomposed according to time segments and regions. A structured instruction containing a time sequence is generated for each region, which includes at least: region identifier, scheduling start timestamp, time granularity, and instruction sequence arranged in time segment order. The instruction sequence for each time segment includes: the output setting vector of each controllable device in the region, the charging and discharging power setting vector of the energy storage device, the power exchange setting vector with the interconnection line of the adjacent region, and the flow rate setting for the thermal pipeline; each item in the instruction sequence is accompanied by an execution time offset field, which is calculated and determined based on the optimal flow rate and pipeline length obtained by solving the transmission time delay equation, and is used to instruct the edge computing node to execute the instruction in advance or delay by a corresponding time after receiving the instruction.

[0014] The technical effects and advantages of this invention are as follows: 1. This invention establishes a unified industrial mechanism framework covering the energy cascade utilization and multi-energy complementarity laws of the entire system, breaking down the barriers of independent optimization where each region focuses only on its own interests. It fundamentally solves the core defect of existing technologies that lack a unified mechanism framework, which leads to the ineffective implementation of global resource allocation goals, and achieves a synergistic improvement in multi-energy complementarity and overall economic efficiency across the entire system. 2. This invention describes heat transfer using partial differential equations, achieving deep collaborative modeling of the physical dynamic characteristics of cross-regional networks and the internal operating characteristics of each region. It solves the fundamental defects of existing technologies that rely solely on pre-set tie-line plans for boundary coupling and fail to characterize the physical transmission process, resulting in poor executability of scheduling strategies. This significantly improves the adaptability of scheduling strategies to actual physical conditions. 3. This invention achieves parallel updating of local decision variables in each region and coordinated convergence of global constraints through an improved augmented Lagrangian function, ensuring strict satisfaction of cross-regional physical constraints. It solves the problems of disconnection between independent subproblem optimization and global constraints and poor convergence in existing technologies, further improving the global optimality and constraint compliance of the scheduling strategy. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating a method for interconnected and coordinated operation of a multi-regional integrated energy system according to an embodiment of this application.

[0016] Figure 2 This is a hardware architecture block diagram of an interconnected and coordinated operation method for a multi-region integrated energy system provided according to an embodiment of this application.

[0017] Figure 3 This is a software architecture block diagram of a method for interconnected and coordinated operation of a multi-regional integrated energy system according to an embodiment of this application.

[0018] Figure 4 This is a flowchart illustrating the construction process of a cross-regional physical constraint set according to an embodiment of this application. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to and includes any or all possible combinations of one or more of the listed items. In the description of the embodiments of this application, unless otherwise stated, “a plurality” means two or more.

[0021] Figure 1 This is a flowchart illustrating a method for interconnected and coordinated operation of a multi-regional integrated energy system provided in an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes: S1: In response to the coordinated operation and scheduling instructions of the multi-regional integrated energy system, obtain the energy conversion equipment model, energy storage dynamic characteristics, and renewable energy output prediction curves within each region; the coordinated operation and scheduling instructions may be day-ahead scheduling tasks executed at a pre-set time, or rolling optimization events issued in real time by the power grid dispatch center or energy operator during operation; these instructions include at least the optimization start and end time, scheduling cycle granularity, and the target preference of this optimization. S2: Based on the acquired data from various regions, construct an industrial mechanism framework that reflects the energy cascade utilization and multi-energy complementarity of the entire system; S3: Couple the internal operating characteristics of each region with the dynamic response characteristics of the cross-regional network to the industrial mechanism framework, and solve to generate collaborative operation scheduling instructions.

[0022] To achieve the above-mentioned interconnected and collaborative operation method, such as Figure 2 As shown, the application at the hardware architecture level includes: edge computing nodes deployed at the edge of the integrated energy system in various regions for performing local data acquisition and preprocessing; one or more cloud-based collaborative computing platforms for performing complex computational tasks of global collaborative modeling and unified representation of spatiotemporal coupling constraints; and a communication network architecture connecting the edge computing nodes and the cloud-based collaborative computing platforms to support low-latency data reporting and control command issuance.

[0023] To achieve the above-mentioned interconnected and collaborative operation method, such as Figure 3 As shown, the application at the software architecture level includes: a data acquisition layer, used to acquire real-time operational data from sensors, smart meters, and energy management systems deployed in various regions; a model building layer, used to execute the above S1 and S2 parts to construct the industrial mechanism framework and initialize parameters; an optimization solution layer, used to execute collaborative optimization solutions based on the improved augmented Lagrangian function to generate the global optimal solution; and an instruction execution layer, used to parse the solution results into executable collaborative operation scheduling instructions and distribute them to the energy management systems or specific execution devices in various regions.

[0024] The cloud-based collaborative computing platform, through the aforementioned communication network architecture, broadcasts a data acquisition command in parallel to all edge computing nodes deployed within the integrated energy systems of various regions. This command encapsulates the time range and periodic granularity of the current scheduling, requiring each node to provide data according to a unified timeframe. Each edge computing node responds to the received data acquisition command by executing data acquisition operations in parallel, and all data is characterized based on industrial mechanism models to ensure the physical accuracy of subsequent modeling. Edge computing nodes, through their local databases or directly connected device controllers (PLCs), read and obtain the operating parameters and characteristic data of key energy conversion equipment within each region, in order to construct steady-state or quasi-steady-state mechanism models: Gas turbine model: based on efficiency curves It means that, among them For output power, The corresponding power generation efficiency is determined; simultaneously, its ramp rate constraint is obtained. , and start-up / shutdown costs , Waste heat boiler model: based on heat exchange efficiency This indicates that it recovers waste heat power from the input flue gas in the exhaust gas of the gas turbine. With the output of steam / hot water effective thermal power The relationship between them: Electric chiller model: based on coefficient of performance This indicates the amount of electrical power it consumes. With output cooling power The relationship between them: Electro-gas conversion equipment model: based on conversion efficiency This indicates the amount of electrical power it consumes. Equivalent calorific value power of produced natural gas The relationship between them: ; Edge computing nodes acquire dynamic time-domain mechanism models of energy storage devices within each region: Battery energy storage model: using discretized difference equations Description, in which, Let be the state of charge at time t. and t represents the charging and discharging power at time t (both are non-negative values). and These are the charging and discharging efficiencies, respectively. For time step, Rated capacity; simultaneously obtain SoC upper and lower limit constraints; thermal storage tank / gas storage tank model: in terms of energy state The difference equation describes: ,in To store energy, , For charging and discharging power, , For charging efficiency and discharging efficiency, This is the heat dissipation loss coefficient, expressed in kW / K. Let t be the average temperature of the energy storage medium inside the thermal storage tank. Let t be the temperature of the environment in which the thermal storage tank is located, and obtain its upper limit capacity. With lower capacity limit ; Edge computing nodes obtain the predicted renewable energy output sequence for each region over a future period from the local renewable energy forecasting system: Photovoltaic output forecast curve: , Where T is the total number of time segments included in the scheduling cycle; Wind power output prediction curve: , .

[0025] In one specific embodiment, the edge computing node invokes a locally deployed time-series forecasting model based on gated recurrent units (GRUs). The input consists of historical power output data from the past 72 hours and features such as irradiance, wind speed, and temperature from numerical weather prediction. The output is a 24-hour forecast sequence for photovoltaic / wind power with a 15-minute time granularity. Each forecast value is accompanied by a 95% confidence interval to characterize the forecast uncertainty. The GRU model is retrained daily at midnight using the latest week's historical data to ensure forecast accuracy.

[0026] Before uploading the aforementioned data to the cloud-based collaborative computing platform, each edge computing node performs a normalization preprocessing operation. Specifically, for any data sequence, the node uses a maximum-minimum normalization formula to uniformly map its numerical range to the interval [0,1].

[0027] In this invention, the industrial mechanism model refers to a mathematical model established based on physical laws such as thermodynamics, electricity, and fluid mechanics, as well as the inherent energy conversion / storage laws of equipment. Its parameters have clear physical meanings (efficiency, thermal resistance, inertial time constant, etc.), which is different from the "black box" data-driven model that only relies on statistical correlation and has no physical connotation.

[0028] In one possible implementation, the construction of the industrial mechanism framework includes: generating a set of initial Lagrange multiplier fields, which are used to quantify the marginal contribution of each region to the global comprehensive benefits at different time sections; using the initial Lagrange multiplier fields as weighting coefficients, constructing a cross-regional physical constraint set containing a transmission time delay equation in the form of a partial differential equation; adding the cross-regional transmission delay time of electrical energy and thermal energy as continuous variables to the optimization model, and combining the cross-regional multi-energy coupling constraints constructed by the energy hub matrix to form an industrial mechanism model containing transmission delay, capacity limitation, and multi-energy flow nonlinearity, and outputting an improved augmented Lagrangian function.

[0029] The cloud-based collaborative computing platform creates a virtual hub node encompassing all regions. Logically, this virtual hub node sits above all physical regions and does not directly correspond to any actual equipment. Specifically, it is a mathematical abstraction containing a global energy efficiency and carbon emission collaborative optimization objective function. The energy conversion equipment models, energy storage dynamic characteristics, and renewable energy output prediction curves for each region are uniformly mapped to the collaborative optimization objective function F of the virtual hub node using a three-dimensional index structure of region, equipment, and time series. Specifically, for the i-th region, the output of its gas turbine... and its costs Mapped as a sum of terms in F; the state of charge of its energy storage device Mapped to dynamic constraints in F; its photovoltaic predicted output It is mapped to the renewable consumption penalty term in F.

[0030] In the initial state, if the boundary exchange power of the i-th region at time t is increased by one unit, the objective function will decrease. One unit. Introduce a set of globally unified Lagrange multiplier initial fields. Ignoring all inter-regional coupling constraints, we solve the single-region optimization problem for each region separately to obtain the local optimal multipliers for each region. And the predicted boundary power values. Substitute the predicted boundary power values ​​for each region into the global coupling constraints to calculate the coupling constraint residual vector. A one-step multiplier update formula is used. ,in The initial field is obtained by using the penalty parameters. ,in This is the step size factor.

[0031] Let G=(V,E) be an undirected graph representing the network topology of cross-regional energy transmission in a multi-regional integrated energy system, where V is the set of regional nodes (including nodes of each regional integrated energy system and external energy market nodes), and E is the set of energy transmission lines connecting the nodes (including power transmission lines, heating pipelines, and natural gas pipelines). Each edge e∈E must include at least: the type of transmission medium (electricity / heat / gas) and the length of the transmission medium. Rated transmission capacity The cloud-based collaborative computing platform obtains the topology information from a pre-configured topology database or through input via a graphical interface.

[0032] like Figure 4 As shown, the power transmission delay time Where c is the speed of light, and under typical inter-regional distances (≤500km), Less than 0.01 seconds, far smaller than the scheduling cycle granularity, it can be ignored in practical optimization and considered as instantaneous transmission. However, for ultra-long distances, this invention still supports modeling it as a fixed delay. For heat transfer, a one-dimensional convection-diffusion partial differential equation is used. Describe the temperature of the heat transfer medium The dynamic changes along the pipe position x and time t, where For flow rate, The ambient temperature ranges from -20℃ to 40℃, depending on the season and geographical location. In optimization models, it is typically taken as the average temperature within the scheduling period or an hourly temperature sequence. Where is the thermal diffusivity, Let be the thermal conductivity of the heat transfer medium. For water, The value ranges from 0.5 to 0.7 W / (m·K), with the specific value varying with temperature and pressure. The equivalent heat dissipation coefficient, It is a constant. For pipe diameter, The overall heat transfer coefficient ranges from 0.5 to 5.0 W / (m²). 2 ·K), depends on the thickness of the pipe insulation layer, For cross-section, For heat medium density, The specific heat capacity of the heat medium is used; both can be calculated from the physical parameters of the heat medium and the pipeline, and this scheme does not impose specific restrictions on them. The implicit finite difference method is used to discretize the equations into a system of linear algebraic equations, and the pipeline length is considered. Divide into M spatial steps The scheduling period T is divided into K time steps. At each spatial node and time nodes The above will include the transmission delay time. As a continuous variable, it is implicitly included in the temperature state variable, and is represented as a linearization constraint: , where f is the thermodynamic transfer function. Simultaneously, the flow velocity... As a decision variable, through pipeline length The ratio of the flow rate to the transmission delay time defines the transmission delay time. In practical engineering optimization, to balance computational accuracy and solution efficiency, the partial differential equation is usually discretized into a low-dimensional system of linear algebraic equations using the implicit finite difference method. The mathematical expressions for the boundary and initial conditions of the one-dimensional convection-diffusion partial differential equation and their corresponding physical meanings are shown in Table 1. Table 1 For any power-to-gas conversion equipment, combined heat and power unit, or multi-energy conversion station connecting different regions, its input vector ,in To input the electrical power of this hub from region i, To input the natural gas power of the hub from region i, the output vector is... ,in, The natural gas output from this hub to region j, For the heat power output by the hub to region j, then the energy hub matrix is... satisfy ,and Limited by region The upper limit of output, Limited by region The upper limit of the receiving capacity. The matrix elements of H are determined by the efficiency and allocation coefficients of each conversion device. Specifically, the matrix of a typical power-to-gas / cogeneration (CHP) coupling hub is: , The heating efficiency of the gas-fired boiler. For the coupled device connecting region i and region j, its input power comes from the output of region i, and the output power is sent to region j.

[0033] The cloud-based collaborative computing platform integrates the collaborative optimization objective function F with all physical constraints (including network topology constraints, linear constraints after discretization of the transmission delay equation, energy hub matrix constraints, and device upper and lower bound constraints, etc.) to output an improved augmented Lagrangian function: Where de is the set of all decision variables, including the output of equipment in each region, energy storage charging and discharging power, cross-regional exchange power, pipeline flow velocity, etc., and JD is the index set of all constraints. The residual function of the jd-th constraint ( =0 is an equality constraint. ≤0 is an inequality constraint, which can be transformed into an equality by introducing slack variables. For Lagrange multipliers, start directly from the initial field of the Lagrange multipliers. Select and map from, This is a penalty parameter used to balance the weights of the objective function and constraint violations, and it is adaptively adjusted during the iteration process. During the iterative solution process, the penalty parameter... The adaptive adjustments are shown in Table 2: Table 2 in, For the original residual, Let kn be the dual residual, and kn be the number of iterations. The value of is determined based on numerical experimental experience, with a recommended range of 1.5 to 2.0. The specific value can be optimized during debugging based on the system scale. The initial field of the Lagrange multipliers is used as the starting point for iteration, and a penalty parameter is set. =1.0, convergence tolerance is 1e-4, and the maximum number of iterations is defined as 1000. In the kn-th iteration, each region i is solved independently: ,in From the global Lagrange multiplier vector The selected decision variables corresponding to all regions i are selected from the data. The multiplicand subset of the constraint conditions It is the part of the global augmented Lagrangian function L that is relevant to region i, specifically including: the local running cost function of region i. The penalty term for local physical constraints and the coupling constraint term with neighboring region j (fixing the decision variables of neighboring regions to the values ​​of the previous iteration) After each iteration, each region will exchange updated boundary power. It sends information to neighboring areas through a communication network and simultaneously receives boundary information from neighboring areas.

[0034] Update the Lagrange multipliers: After completing the multiplier update, calculate the original residual. and dual residuals For each constrained residual Divide by the typical value of its corresponding constraint boundary (such as the rated power of the equipment, temperature range) to obtain the dimensionless residual. .like ,in If the baseline vector of the decision variables (which can be the rated values ​​of each device) is used, then convergence is determined, iteration stops, and the global optimal solution is output. Otherwise, if Then, the penalty parameters are adaptively adjusted according to Table 2. The process then proceeds to the next iteration. If convergence is not achieved after reaching the maximum number of iterations, the penalty parameter is gradually increased to an upper limit of 1000, triggering a constraint relaxation mechanism—replacing constraint residuals with absolute values ​​greater than a preset threshold with relaxation variables, and adding an L1 penalty term for the relaxation variables to the objective function. Iteration continues until a feasible solution is obtained. Through the above iterative solution process, the optimal set of decision variables satisfying all global and local constraints is output. It includes a complete operation plan for each region at all scheduling times, including equipment output, energy storage capacity, cross-regional exchange capacity, and pipeline flow rate.

[0035] Global Lagrange multiplier vector Divide into several sub-vectors according to constraint type: The local constraints corresponding to region i, This corresponds to the coupling constraint between regions i and j. In the subproblem of region i, only... and all (j is adjacent to i), denoted as For the cross-regional power exchange constraints connecting regions i and j Its corresponding multiplier Initialized to This refers to the average marginal contribution of adjacent regions. The mapping relationship between the decision variable *de* and the components of the augmented Lagrangian function *L* is shown in Table 3. Table 3 The operational characteristics within each region, including but not limited to the efficiency curves of energy conversion equipment, the dynamic state-of-charge constraints of energy storage equipment, the uncertainty range of renewable energy output, and the rigid / flexible demand of local loads, are mapped to the collaborative optimization objective function F. Decision variables... and its associated constraints The form ≤0 is embedded into each term of the augmented Lagrangian function L. Specifically, for each region i, its local running cost function is directly summed to F, and its local physical constraints are used as... Part of it, through Lagrange multipliers and penalty items Perform a relaxation process.

[0036] Once the solution converges, the global optimal solution is obtained, which includes all decision variable values ​​(equipment output, energy storage power, exchange power, pipeline flow rate, etc.) for all regions at each scheduling time t. An independent collaborative operation scheduling instruction set is generated for each region, which contains a time-series structured data packet in JSON format. Specific fields include: region identifier, each time segment within the scheduling cycle, output setpoint for each controllable device (gas turbine, energy storage, electric chiller, electric-to-gas equipment, etc.) within the region, planned exchange power for the interconnection line with adjacent regions, and for instructions involving heat transmission, an advance / lag time offset is added to indicate whether the executing equipment needs to act earlier or later to compensate for transmission delay.

[0037] The cloud-based collaborative computing platform uses a communication network architecture to distribute collaborative operation scheduling instruction sets from various regions to the corresponding edge computing nodes in parallel. Upon receiving the instructions, the edge computing nodes parse them into low-level control signals recognizable by their local Energy Management System (EMS) or Power Supply Controller (PLC). Based on the time delay compensation parameters in the instructions, they set the offset for the execution time to ensure that, after considering transmission delays, energy arrives at the target area at the correct time. They then send instruction reception confirmation information back to the cloud, including a local timestamp and execution status.

[0038] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other. In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for interconnected and coordinated operation of a multi-regional integrated energy system, characterized in that, include: S1: In response to the coordinated operation and scheduling instructions of the multi-region integrated energy system, obtain the energy conversion equipment model, energy storage dynamic characteristics and renewable energy output prediction curves within each region; S2: Based on the acquired data from various regions, construct an industrial mechanism framework reflecting the energy cascade utilization and multi-energy complementarity of the entire system, including: A set of initial fields of Lagrange multipliers is generated, which are used to quantify the marginal contribution of each region to the overall global benefits at different time sections; Using the initial field of the Lagrange multipliers as weighting coefficients, a cross-regional physical constraint set containing a transmission time delay equation in the form of a partial differential equation is constructed; the cross-regional transmission delay time of electrical energy and thermal energy is added as a continuous variable to the optimization model, and the cross-regional multi-energy coupling constraint constructed by the energy hub matrix is ​​combined to form an industrial mechanism model containing transmission delay, capacity limitation, and multi-energy flow nonlinearity, and outputs an improved augmented Lagrange function; S3: Couple the internal operating characteristics of each region with the dynamic response characteristics of the cross-regional network to the industrial mechanism framework, and solve to generate collaborative operation scheduling instructions.

2. The interconnected and coordinated operation method of a multi-regional integrated energy system according to claim 1, characterized in that: S2 represents the marginal reduction in the overall global benefit resulting from increasing the boundary exchange power of the i-th region by one unit at time t. The quantification of the marginal contribution specifically includes: Solve the single-region optimization problem for each region to obtain the local optimal multipliers for each region. ; Substituting into the collaborative optimization objective function F, the initial field is obtained through one Newton step correction. ,in Step size factor This represents the gradient of the global residual with respect to the multipliers.

3. The interconnected and coordinated operation method of a multi-regional integrated energy system according to claim 1, characterized in that: S2, which treats the inter-regional transmission delay time of electrical energy as a continuous variable, specifically includes: When the distance between regions exceeds a preset threshold, the power transmission delay time will be increased. As a fixed parameter added to the optimization model, where c is the speed of light, This refers to the length of the power transmission line.

4. The interconnected and coordinated operation method of a multi-regional integrated energy system according to claim 1, characterized in that: S2 involves constructing a cross-regional physical constraint set for the cross-regional heat transfer network, specifically including: The temperature of the heat transfer medium is described by a one-dimensional convection-diffusion partial differential equation. Dynamic changes along the pipeline at position x and time t: in, For flow rate, For ambient temperature, Where is the thermal diffusivity, For the thermal conductivity of the heat transfer medium, The equivalent heat dissipation coefficient, It is a constant. For pipe diameter, The overall heat transfer coefficient is... For cross-section, For heat medium density, Specific heat capacity of the heat transfer medium; The partial differential equation is expressed in space step size and time step Discretize the above, and transform it into a system of linear algebraic equations describing the relationship between the temperatures of each discrete node in the pipeline; Pipe inlet temperature sequence With outlet temperature sequence The dynamic mapping relationship between them is added as a set of equality constraints to the cross-regional physical constraint set, where Transmission delay time and flow rate As a decision variable, This refers to the length of the pipe. The discretized linear constraints are compared with the upper limit of the pipeline's transmission capacity. Lower limit Together they constitute the physical constraint set of the cross-regional heat transfer network.

5. The interconnected and coordinated operation method of a multi-regional integrated energy system according to claim 1, characterized in that: The construction of S2, the cross-regional multi-energy coupling constraint, specifically includes: For a cross-regional energy conversion device connecting the first and second regions, obtain the mathematical model of the device's energy hub, using matrix equations. Characterization, where the input vector u contains at least the input electrical power and the input gas power, the output vector y contains at least the output gas power and the output thermal power, and H is an energy hub matrix containing the efficiency coefficients and energy distribution coefficients of each energy conversion device; Each component of the matrix equation is expanded into a linear equation constraint and added to the cross-regional physical constraint set; Based on the physical rating of the energy conversion device, upper and lower limit constraints are set for each component in the input vector u and the output vector y, and these upper and lower limit constraints are added to the cross-regional physical constraint set.

6. The interconnected and coordinated operation method of a multi-regional integrated energy system according to claim 1, characterized in that: In S2, the improved augmented Lagrangian function L is defined as: The collaborative optimization objective function F is integrated with the cross-regional physical constraint set, specifically expressed as: Where de is the set of all decision variables. Let represent the decision variable for the i-th region at time t, and JD be the index set of all constraints. Let jd be the residual function of the jd-th constraint. For Lagrange multipliers, start directly from the initial field of the Lagrange multipliers. Select and map from, This is the penalty parameter.

7. The interconnected and coordinated operation method of a multi-regional integrated energy system according to claim 6, characterized in that: S2 iteratively solves the improved augmented Lagrange function L, updating the local decision variables of each region in parallel in each iteration, exchanging the boundary information of adjacent regions, and updating the Lagrange multipliers. During the iteration process, the ratio of the original residual to the dual residual is monitored in real time. ,when Add penalty parameters To strengthen constraint satisfaction, when Time decrease To promote goal optimization, otherwise maintain The original residual and the dual residual remain unchanged until both are less than the preset convergence tolerance. in, For the original residual, Let k be the dual residual and kn be the number of iterations.

8. The interconnected and coordinated operation method of a multi-regional integrated energy system according to claim 1, characterized in that: S3, the generation of coordinated operation scheduling instructions, specifically includes: The obtained global optimal solution is decomposed according to time segments and regions. A structured instruction containing a time sequence is generated for each region, which includes at least: region identifier, scheduling start timestamp, time granularity, and instruction sequence arranged in time segment order. The instruction sequence for each time segment includes: the output setting vector of each controllable device in the region, the charging and discharging power setting vector of the energy storage device, the power exchange setting vector with the interconnection line of the adjacent region, and the flow rate setting for the thermal pipeline; each item in the instruction sequence is accompanied by an execution time offset field, which is calculated and determined based on the optimal flow rate and pipeline length obtained by solving the transmission time delay equation, and is used to instruct the edge computing node to execute the instruction in advance or delay by a corresponding time after receiving the instruction.

Images