New power system time series production simulation method and system based on model decomposition

Abstract

The application belongs to the technical field of power system operation, and provides a novel power system time sequence production simulation method and system based on model decomposition, with the optimal system total operation cost as an objective function, the system power balance constraint and the line transmission overrun constraint being comprehensively considered, and the constraint condition of equipment level being introduced to construct a coordinated time sequence production simulation model; start-up and shutdown secondary compression is carried out; the coordinated time sequence production simulation model is decomposed into a source-load-storage coordinated time sequence output simulation main problem model without line transmission constraint and a line transmission overrun feasibility checking sub-problem model, the unit start-up and shutdown integer variable and the output continuous variable are decoupled to construct a unit start-up and shutdown-output inner decomposition model; optimization solving is carried out until each decomposition model converges, and the optimal solution is obtained. The application comprehensively considers various types of equipment and their coupled operation constraints, adopts an optimality decomposition algorithm to describe the operation process, so as to accurately evaluate the power balance risk.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of new power system operation, and particularly relates to a new power system time sequence production simulation method and system based on model decomposition. BACKGROUND

[0002] The statements in this section merely provide background information related to the present application and do not necessarily constitute prior art.

[0003] At present, the grid-connected scale of renewable energy sources such as wind power and photovoltaic power and the installed capacity of energy storage have both achieved substantial growth, while the demand side flexibility has also been significantly improved, together forming a new pattern of source-load interaction, whose operation characteristics are represented by the interaction and cooperation of large-scale and multi-type devices in multiple time scales.

[0004] However, the traditional production simulation method usually adopts a low time resolution, and weakly time-simplified processes the operation characteristics (such as unit output and load demand) of conventional units and loads, ignoring the time coupling relationship between device operation states, and thus being difficult to accurately depict the randomness and volatility characteristics of renewable energy output. Therefore, the time sequence production simulation for fine simulation of power and energy balance process has become an important development direction of new power system operation analysis.

[0005] However, the existing main methods, such as time sequence production simulation with high time resolution at the hourly level, will lead to a large-scale linear growth of model variables and constraint dimensions in time and space scales. The model contains a large number of constraint conditions describing the heterogeneous operation characteristics of multi-type devices, and especially considers the cross-period coupling constraints, resulting in extremely high model complexity. In addition, the integer variables introduced by a large number of unit start-stop behaviors further aggravate the difficulty of model solving, eventually forming a high-dimensional, strongly coupled mixed integer linear programming problem.

[0006] In order to improve the solving efficiency of time sequence production simulation, two types of simplification methods are currently mainly used: dimension reduction method and decomposition method. The dimension reduction method usually reduces the model solving dimension based on clustering technology, but may simplify or ignore some key unit physical operation constraints (such as minimum output limit, climbing ability, etc.). The decomposition method mainly includes relaxation optimization decomposition, stochastic optimization decomposition and optimality decomposition algorithm. Among them, the relaxation optimization decomposition and stochastic optimization decomposition algorithm have strong adaptability to the problem structure, but it is difficult to strictly guarantee the convergence to the theoretical optimal solution; while the optimality decomposition algorithm can guarantee the theoretical optimality, but usually has higher requirements for the problem form, and its solving efficiency is significantly dependent on the decomposition strategy and cut plane design method used. SUMMARY

[0007] The present application proposes a novel power system time sequence production simulation method and system based on model decomposition to solve the above problems, which realizes comprehensive consideration of various types of equipment and their coupled operation constraints, adopts an optimality decomposition algorithm to describe the operation process as truly as possible, and thus accurately assesses the power balance risk.

[0008] According to some embodiments, the present application adopts the following technical solutions:

[0009] A novel power system time sequence production simulation method based on model decomposition, comprising the following steps:

[0010] Determine the operation boundary of the novel power system, take the optimal system total operation cost as the objective function, comprehensively consider the system power balance constraint and line transmission overrun constraint, and introduce various types of equipment level constraint conditions of source, network, load and storage to construct a coordinated time sequence production simulation model;

[0011] Combine engineering experience, construct an initial compression strategy of start-stop optimization space with economic optimality as the target, solve the optimal system output scheme under the full space-time coordinated optimization framework through relaxation start-stop, and realize secondary compression of start-stop;

[0012] Decompose the coordinated time sequence production simulation model into a source-load-storage coordinated time sequence output simulation main problem model without line transmission constraint and a line transmission overrun feasibility checking sub-problem model, take them as outer decomposition models, decouple the unit start-stop state integer variable and the output continuous variable, and construct a unit start-stop-output inner decomposition model;

[0013] Optimize and solve the inner decomposition model and the outer decomposition model, and in the solving process of the outer decomposition model, adopt an optimality decomposition algorithm to decouple the unit start-stop state integer variable and the output continuous variable until each decomposition model converges to obtain the optimal solution.

[0014] As an optional implementation, the process of determining the operation boundary of the novel power system comprises: determining the operation boundary of the novel power system, calculating the maximum load of each node, combining the typical load daily curve and the annual daily maximum load change curve, reconstructing, and forming an input demand meeting the time sequence production simulation;

[0015] According to the average available density of wind and light resources in each region, calculate the wind and light resource generation boundary of the region, and further construct the typical wind and light power time sequence of each region;

[0016] Determine the spatial distribution, capacity scale and power generation technology parameters of each regional thermal power unit;

[0017] Obtain the rated transmission capacity of the transmission line;

[0018] Obtaining adjustable capacity proportion, adjustable duration and response speed parameters of the demand side adjustable load resource in different time periods.

[0019] As an alternative implementation, the process of taking the total system operation cost as the objective function includes: the total system operation cost includes the thermal power unit output cost, the thermal power unit start-stop cost, the load shedding penalty, the power curtailment penalty and the demand side resource regulation cost.

[0020] As an alternative implementation, the process of comprehensively considering the system power balance constraint and the line transmission over-limit constraint and introducing various device-level constraint conditions of the source, the grid, the load and the storage includes: calculating the total generated power and the total load power to form the source-load power balance constraint of the total generation and the total load being equal;

[0021] According to the node power balance relationship, the active power expression of each node injected into the power transmission network is constructed, the relationship expression between the line power flow and the node injection power is constructed by using the direct current flow model, and the line transmission over-limit constraint expression is constructed according to the line power flow;

[0022] Taking the load of each node and the wind and light unit output, as well as the load shedding power and the power curtailment power as inputs, the standby demand constraint expression of the thermal power unit is constructed;

[0023] Comprehensively considering the active power output range constraint of the thermal power unit, the active power output range constraint considering standby output, the ramp rate constraint and the minimum start / stop time constraint, the operation characteristic constraint expression of the thermal power unit is constructed;

[0024] After the wind and light time series output curve is normalized, the active power output constraint expression of the wind and light generator is constructed by taking the wind and light time series output curve as input;

[0025] Comprehensively considering the charge and discharge power constraint, the state of charge range constraint and the energy balance constraint of the energy storage device, the operation characteristic constraint expression of the energy storage device is constructed;

[0026] Comprehensively considering the operation power range constraint, the ramp rate constraint, the state of charge dynamic recursion constraint and the upper and lower limit boundary constraint of the adjustable load, the operation characteristic constraint expression of the adjustable load is constructed.

[0027] As an alternative implementation, the process of constructing the start-stop optimization space preliminary compression strategy with economic optimality as the target includes: constructing the unit output economic index expression, comprehensively considering the influence of standby demand and new energy output fluctuation on thermal power output, estimating the upper and lower boundaries of the total thermal power output in each period, sorting all units in ascending order according to the output economic index of the thermal power unit, if the economic index is the same, then performing secondary ascending order sorting according to the minimum technical output, and preferentially selecting the unit with stronger operation flexibility;

[0028] According to the priority, the unit output is set to the maximum technical output in turn and accumulated, and when the cumulative output first exceeds the upper boundary of the total thermal power output, the unit involved is determined as the strong start-up optimization unit;

[0029] For the remaining units, continue to set the output according to the above priority in turn to the average of the maximum and minimum output and accumulate until the lower boundary of the total thermal power output is reached, and the unit involved is determined as the standby start-up optimization unit.

[0030] As an alternative embodiment, the process of realizing the secondary compression of start-up and shut-down by solving the system output optimization scheme under the full space-time coordination optimization framework includes:

[0031] Considering the thermal power unit output cost, load shedding penalty, wind and light power penalty and regulation cost, the objective function expression of the linear relaxation output simulation model is constructed to minimize the source-load-storage coordination operation cost;

[0032] Ignoring all constraints containing only unit start-up and shut-down, the model constraint conditions are only the constraints containing continuous variables and the unit output constraints of the relaxation start-up and shut-down state variables;

[0033] The start-up and shut-down state variables of the strong start-up and standby start-up optimization units are fixed to 1, and the unit output constraint expression of the relaxation start-up and shut-down state variables is constructed;

[0034] For the unit whose output at the target time is greater than the set value, its start-up and shut-down state at that time will be determined as the start-up state, and the corresponding start-up and shut-down state variable value will be fixed to 1 in the subsequent optimization as a constant variable.

[0035] As an alternative embodiment, the process of decomposing the coordinated time sequence production simulation model into the source-load-storage coordinated time sequence output simulation main problem model without line transmission constraints and the line transmission limit feasibility checking sub-problem model includes:

[0036] Considering the thermal power unit output cost, start-up and shut-down cost, load shedding penalty on the load side, wind and light power penalty on the renewable energy side, and system regulation cost, the objective function expression of the source-load-storage coordinated time sequence output simulation main problem model is constructed, which is the same as the objective function expression of the coordinated time sequence production simulation model;

[0037] The constraints related to line transmission limit in the constructed coordinated time sequence production simulation model are decoupled, and the system power balance constraints and the operation characteristics constraints of various types of source, load and storage devices are retained to constitute the constraint conditions of the main problem model;

[0038] The direct current flow model is used to calculate the injection power value of each node in the power transmission network, and the transmission capacity limit of the line is combined to calculate the transmission capacity limit value of each line in each time period, the feasibility of the power flow of all transmission lines in the system is verified, and a transmission limit feasibility checking sub-problem is constructed.

[0039] As an alternative embodiment, the unit start-stop state integer variable is decoupled from the output continuous variable, and the process of constructing the unit start-stop-output inner decomposition model includes:

[0040] The objective function expression of the unit start-stop optimization main problem is constructed, and the objective function expression is the sum of the total start-stop cost of the thermal power unit and the lower bound of the objective function value of the output optimization sub-problem under the target start-stop scheme Minimization;

[0041] The minimum start / stop time constraint of the unit, the start-stop state hard constraint of the must-on / must-off unit, the Initial value constraint in the objective function of the output optimization main problem, and the feasibility / optimality cut plane constraint generated by the sub-problem constitute the constraint conditions of the unit start-stop optimization main problem;

[0042] Under the given unit start-stop state scheme of the main problem, the output optimization sub-problem objective function expression is constructed, which is specifically the sum of the total output cost of the thermal power unit, the total load shedding penalty of the system, the total abandoned electricity penalty of the wind and solar unit, and the total adjustment cost of the demand side resource is minimized;

[0043] The known start-stop state variable value is substituted into the unit output constraint to form the constraint condition of the output optimization sub-problem model, and the unit start-stop optimization main problem and the output optimization sub-problem are solved, and the source-load-storage output scheme of each time period is obtained.

[0044] As an alternative embodiment, the judgment process of the inner decomposition model convergence includes:

[0045] The lower bound value and the upper bound value expression of the objective function of the inner decomposition model in the iteration process are constructed, and in each round of decomposition iteration, the current lower bound value and the upper bound value of the main problem objective function are recorded respectively, and it is judged whether the absolute value of the difference between the lower bound and the upper bound is less than the preset convergence threshold value, if it is satisfied, it is considered that the inner decomposition model has converged to its optimal solution; otherwise, according to the constraint condition of the output optimization sub-problem model, the feasibility / optimality cut plane is generated and fed back to the main problem model, and the unit start-stop scheme is optimized through iteration until the convergence condition is satisfied.

[0046] As an alternative embodiment, the judgment process of the outer decomposition model convergence includes:

[0047] The transmission capacity of each line in all time periods is substituted into the following formula, and the absolute value of the ratio between the transmission capacity and the transmission capacity limit value is determined, if the absolute value is less than the limit threshold value, the current solution is determined as the final solution, otherwise, the feasible cutting plane constraint is generated according to the constructed line transmission limit checking sub-problem, and is added to the constructed source-load-storage co-scheduling output simulation main problem, and the solution of the main problem model is gradually corrected through the feedback of the cutting plane constraint until the convergence condition is met.

[0048] A novel power system time sequence production simulation system based on model decomposition, comprising:

[0049] The co-time sequence production simulation model construction module is configured to determine the operation boundary of the novel power system, take the total operation cost optimization as the objective function, comprehensively consider the system power balance constraint and the line transmission limit constraint, and introduce various device-level constraint conditions of the source, the network, the load and the storage to construct a co-time sequence production simulation model.

[0050] The two-stage compression module is configured to combine engineering experience, construct a start-stop optimization space preliminary compression strategy with economic optimization as the target, solve the system output optimal scheme under the full space-time co-optimization framework through relaxation start-stop, and realize secondary compression of start-stop.

[0051] The inner-outer layer decomposition model construction module is configured to decompose the co-time sequence production simulation model into a source-load-storage co-time sequence output simulation main problem model without line transmission constraint and a line transmission limit feasibility checking sub-problem model, take the main problem model and the sub-problem model as the outer decomposition model, decouple the unit start-stop state integer variable and the output continuous variable, and construct a unit start-stop-output inner decomposition model.

[0052] The solving module is configured to optimize and solve the inner decomposition model and the outer decomposition model, and in the solving process of the outer decomposition model, the optimal decomposition algorithm is used to decouple the unit start-stop state integer variable and the output continuous variable until the convergence of each decomposition model is achieved, and the optimal solution is obtained.

[0053] Compared with the prior art, the beneficial effects of the present application are:

[0054] The present application provides a co-time sequence production simulation method suitable for the characteristics of a high proportion of renewable energy in a novel power system, which finely considers the time sequence operation characteristics of thermal power units, wind and light power sources, energy storage devices, adjustable loads and transmission lines to construct a time sequence production simulation model, further introduces actual engineering experience, constructs a unit start-stop optimization space compression strategy based on target optimality and approximate optimal solution guidance under a co-optimization framework, and on this basis, constructs a double-layer nested decomposition model, realizes reasonable decomposition and rapid iterative solution of the model through an efficient constraint feedback mechanism, and significantly improves the calculation speed of the time sequence production simulation.

[0055] The application is suitable for large-scale new power system time sequence production simulation, and improves the solving efficiency of time sequence production simulation by cooperating three parts of coordinated time sequence production simulation model construction, start-stop optimization space compression, and double-layer nested decomposition model reconstruction and solving based on optimality decomposition algorithm.

[0056] The application forms a coordinated time sequence production simulation model suitable for the operation characteristics of new power system by constructing an optimization objective function covering the cost of thermal power output, start-stop cost, load shedding penalty, electricity abandonment penalty, and demand side regulation cost, then constructing system level constraints based on system power balance and transmission line transmission capacity limit, and constructing device level operation constraints based on the fluctuating output boundary of new energy units and the cross-time period energy balance of energy storage devices.

[0057] The application introduces a large number of 0-1 integer variables caused by unit start-stop behavior and the mixed integer programming problem caused by the 0-1 integer variables, and proposes a two-stage compression strategy: combining engineering experience, constructing a start-stop optimization space preliminary compression strategy based on economic optimality target; relaxing the start-stop to solve the approximate optimal scheme of system output under the full space-time coordinated optimization framework, and constructing a start-stop secondary compression strategy accordingly, so as to realize model dimension reduction.

[0058] The application adopts a logical optimality decomposition algorithm to realize outer layer decomposition of the model according to the principle of avoiding line transmission overrun: the coordinated time sequence production simulation model is decomposed into a source-load-storage coordinated time sequence output simulation main problem model without line transmission constraint and a line transmission overrun feasibility checking sub-problem model. In the outer main problem, the optimality decomposition algorithm is used to decouple the integer variables of unit start-stop state and the continuous variables of output, so as to significantly reduce the solving difficulty of the coordinated time sequence production simulation.

[0059] In order to make the above-mentioned purposes, characteristics and advantages of the application more obvious and easy to understand, the following preferred embodiments are described in detail below, and the accompanying drawings are described as follows. BRIEF DESCRIPTION OF DRAWINGS

[0060] The drawings accompanying the specification of the application form part of the application and are used to provide further understanding of the application, the illustrative embodiments of the application and the description thereof serve to explain the application, and do not constitute an improper limitation on the application.

[0061] Figure 1 A flow chart of a new power system time sequence production simulation based on model decomposition in an embodiment;

[0062] Figure 2 A flow chart of coordinated time sequence production simulation model construction in an embodiment;

[0063] Figure 3 An implementation flow chart of start-stop compression strategy in an embodiment;

[0064] Figure 4 Figure 1 is a flow chart of a process of simultaneous sequence production simulation model decomposition and solution in an embodiment;

[0065] Figure 5 Figure 2 is a schematic diagram of a double-layer nested decomposition model structure in an embodiment;

[0066] Figure 6(a) is a system configuration diagram of a starting point of an application example in an embodiment;

[0067] Figure 6(b) is a typical load curve diagram of an application example in an embodiment;

[0068] Figure 6(c) is a typical wind power time sequence output curve diagram of an application example in an embodiment;

[0069] Figure 6(d) is a typical photovoltaic time sequence output curve diagram of an application example in an embodiment;

[0070] Figure 6(e) is a comparison diagram of start-stop compression strategy effects of an application example in an embodiment;

[0071] Figure 6(f) is a diagram of an outer-layer transmission over-limit line convergence process of an application example in an embodiment. DETAILED DESCRIPTION

[0072] The application will be further described below in conjunction with the drawings and embodiments.

[0073] It should be noted that the following detailed description is illustrative only and is intended to provide further description of the application. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0074] It should be noted that the terms used herein are merely for the purpose of describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, the singular form is intended to include the plural form unless the context clearly indicates otherwise, and it should be further understood that the terms "comprise" and / or "include" as used herein indicate the presence of the features, steps, operations, devices, components and / or combinations thereof.

[0075] The embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0076] Embodiment One

[0077] As described in the background, traditional production simulation is limited to low time resolution, and the operation characteristics of conventional units and loads are usually approximated by weak time series. It is difficult to explicitly represent the time series coupling effect of device state, and further to accurately depict the randomness characteristics of renewable energy output and the time-varying response mechanism of energy storage devices and adjustable loads.

[0078] In contrast, the new type of power system time series production simulation needs to fully consider the heterogeneous operation characteristic constraints of source-grid-load-storage multi-type devices and their complex coupling relationships, especially the cross-period coupling constraints, which makes the model present large-scale, strongly coupled mixed integer programming characteristics, resulting in too long solving time or even unable to obtain results. Therefore, how to design an efficient optimality decomposition framework to improve the solving efficiency of the model has become a key problem that needs to be solved for the new type of power system time series production simulation.

[0079] To solve the above problems, the embodiment provides a new type of power system time series production simulation method based on model decomposition. First, a coordinated time series production simulation model with hourly time resolution is constructed to depict the random fluctuation characteristics of renewable energy output and the dynamic response capability of storage-load devices. On this basis, guided by the target optimality in a single period and the approximate optimal solution in the whole period, a two-stage compression strategy for the unit start-stop optimization space is designed: preliminary compression of the necessary shutdown units and secondary compression of the necessary start-up units. In the compressed start-stop optimization space, based on the highly coupled characteristics of source-grid-load-storage operation constraints and their mixed integer programming structure, outer and inner decomposition models are constructed to significantly improve the solving speed while ensuring accuracy and sacrificing controllability, thereby supporting the time series production simulation of large-scale new type of power system.

[0080] The following will be described in detail.

[0081] As shown in the method, the method comprises the following steps: Figure 1

[0082] Step 1: Coordinated time series production simulation model construction

[0083] The new type of power system time series production simulation simulates the system power balance process period by period with the source-grid-load-storage planning scheme as the boundary. The model construction is further divided into four parts: first, determine the operation boundary of the new type of power system, then construct the objective function of the coordinated time series production simulation model, further construct the system level constraint conditions, and finally form the device level constraint conditions of source, grid, load, storage and the like. The flow chart of the coordinated time series production simulation model construction is shown in Figure 2 , and the specific steps are as follows:

[0084] Step 1.1: Determine the operation boundary parameters of the new type of power system based on investigation or data extrapolation, specifically including: ​

[0085] (1) Collect the economic and population prediction data of each region in the target year from the reports published by social and economic research institutions, and substitute them into formula (1) to form the maximum load of each node:

[0086] (1)

[0087] In the formula: is the maximum load prediction result of the target year, is the corresponding economic aggregate and population prediction data, is the economic and population-related coefficient of the load, which is obtained based on the historical data of the maximum load, economy, and population in previous years;

[0088] After obtaining the annual maximum load of each node, further combine the typical load day curve and annual daily maximum load change curve published by the power department, and use the normalization and interpolation method to reconstruct the hourly load time series of the target year, which meets the input demand in the time series production simulation.

[0089] (2) Investigate the wind and solar resource distribution map published by the land and resources department, and calculate the average available density of wind and solar resources in each region according to the formula to calculate the boundary of wind and solar resources available in the region.

[0090] (2)

[0091] In the formula: , , is the wind and solar resource density, available area, and upper limit of wind and solar resource power generation in the region where the node i is located, is the theoretical highest conversion efficiency of wind and photovoltaic power related to the current generation technology level, which is based on the upper limit value of the technical specifications published by the equipment manufacturer.

[0092] To meet the accuracy requirements of time series simulation, construct the typical wind and solar power time series of each region by investigating the historical measured meteorological and power data provided by NREL and other institutions.

[0093] (3) Investigate the installed capacity data and product catalog published by local energy management departments and mainstream thermal power equipment manufacturers to determine the spatial distribution, capacity scale, and power generation technology parameters of thermal power units in each region.

[0094] (4) Investigate the transmission technology standards and engineering cases, and count the commonly used conductor types and cross-sectional areas under 500kV, 1000kV, and other voltage levels, and obtain the rated transmission capacity of the transmission line accordingly.

[0095] (5) Investigate the demand-side adjustable load resources, and focus on the statistics of the adjustable capacity ratio, regulation duration and response speed of typical loads such as air conditioners and electric hydrogen production in different time periods.

[0096] Step 1.2: Build a coordinated sequential production simulation model objective function with the goal of optimizing the total system operating cost. Specifically, it includes:

[0097] (1) Consider the fuel cost, carbon emission cost, and standby cost of thermal power units, and substitute them into the following formula to build the thermal power unit output cost expression:

[0098] (3)

[0099] In the formula: is the total output cost of thermal power units; is the number of system operating cycles; is the set of thermal power sites; , are the unit coal consumption cost and carbon emission cost; is the coal consumption coefficient of thermal power units; , , are the active power output, positive and negative reserve size of thermal power units at node at time ; , are the unit costs of positive and negative reserve of thermal power units.

[0100] (2) Substitute the following formula to build the thermal power unit start-stop cost expression:

[0101] (4)

[0102] In the formula: is the total start-stop cost of thermal power units; , are the single start and stop costs of thermal power units; , are the start and stop indicator variables of thermal power units at node at time .

[0103] (3) Substitute the following formula to build the load shedding penalty and power curtailment penalty expression:

[0104] (5)

[0105] In the formula: is the total load shedding penalty of the system; is the set of load sites; is the unit power penalty for load shedding of the system; For nodes At the moment The load shedding power; Total curtailment penalties for wind and solar power units; For the site selection of wind and solar power units; Fines per unit power for abandoned electricity from wind and solar power units; node At the moment The amount of wind and solar power curtailed. (4) Substitute into the following formula to construct the expression for demand-side resource adjustment costs:

[0106] (6)

[0107] In the formula: The total adjustment cost of demand-side resources; For the location set of adjustable loads; The unit adjustment cost for adjustable load; For nodes At the moment Adjustable load regulation power.

[0108] (5) Total output cost of thermal power units Total start-up and shutdown costs of thermal power units Total system load shedding penalty Total curtailment penalties for wind and solar power units and the total adjustment cost of demand-side resources Add them together to construct the expression for the total system operating cost:

[0109] (7)

[0110] In the formula: This represents the total operating cost of the system.

[0111] Step 1.3 Taking into account both system power balance constraints and line transmission limit exceedance constraints, construct the system-level constraints for the collaborative time-series production simulation model. The specific steps are as follows:

[0112] (1) Substitute into the following formula to calculate the total power generation and total load power, forming a source-load power balance constraint expression where the total power generation equals the total load:

[0113] (8)

[0114] In the formula: This refers to the set of locations for energy storage devices. , They are time points At the node Active power of wind and solar turbines and energy storage equipment; The corresponding power load is determined by the external input power load sequence; For a moment At the node Adjustable load power baseline.

[0115] (2) Based on the power balance relationship of integrated nodes, the DC power flow model, and the transmission capacity limit of the transmission line, construct the transmission limit constraint expression for the transmission line. The specific steps are as follows:

[0116] First, based on the node power balance relationship, substitute the following formula to construct the expression for the active power injected into each node in the transmission network:

[0117] (9)

[0118] In the formula: For a moment Power flows into the node from other nodes via the transmission network. The active power; It is the set of bus nodes.

[0119] Furthermore, using a DC power flow model, the following expression is used to construct the relationship between line power flow and node injected power:

[0120] (10)

[0121] In the formula: For each line t The active power vector is always present. The system power transfer distribution factor matrix; For each node t Inject active power vectors at all times.

[0122] Finally, substituting the line power flow into the following formula, we construct the line transmission limit constraint expression:

[0123] (11)

[0124] In the formula: For the line At any moment The active power transmitted; , These are the corresponding line transmission capacity limits; This is a collection of power transmission lines.

[0125] Step 1.4 Taking into account the operational characteristics of various equipment including power sources, grids, loads, and storage, construct the equipment-level constraints for the collaborative time-series production simulation model, specifically including:

[0126] (1) The load of each node and the output of wind and solar units, and the cut load power and the abandoned power are taken as inputs, substituted into the following formula to construct the reserve demand constraint expression of thermal power units:

[0127] (12)

[0128] (13)

[0129] In the formula: , , , are the reserve requirement coefficients of load and wind and solar unit output respectively.

[0130] (2) The active power output range constraint of thermal power units, the active power output range constraint considering reserve output, the ramp rate constraint and the minimum start / stop time constraint are comprehensively considered to construct the operation characteristic constraint expression of thermal power units:

[0131] Firstly, the start / stop state and the active power output of each node unit are taken as inputs, substituted into the following formula to construct the active power output constraint expression of thermal power units:

[0132] (14)

[0133] In the formula: is the active power output of the thermal power unit at node at time , and is a 0-1 variable; , are the lower and upper limits of the output of thermal power units respectively.

[0134] Then, the active power output and its limit of the unit are taken as inputs, substituted into the following formula to construct the active power output constraint of thermal power units considering reserve output:

[0135] (15)

[0136] Further, the active power output of the unit is taken as input, substituted into the following formula to construct the ramp constraint of the thermal power unit:

[0137] (16)

[0138] In the formula: , are the maximum rising and falling ramp rates of the thermal power unit respectively.

[0139] Finally, the start and stop indication variables of the unit are taken as inputs, substituted into the following formula to construct the minimum start / stop time constraint of the thermal power unit:

[0140] (17)

[0141] wherein: , is the minimum start-up and shut-down time length of the thermal power unit.

[0142] (3) After normalizing the wind-solar time-series output curve, the wind-solar time-series output curve is taken as the input, and substituted into the following formula to construct the active power output constraint expression of the wind-solar generator unit:

[0143] (18)

[0144] wherein: is the normalized maximum output of the th wind-solar unit at the moment , which is determined by the externally input wind-solar maximum output sequence.

[0145] (4) The charge-discharge power constraint, the state of charge range constraint and the energy balance constraint of the energy storage device are integrated to construct the operation characteristic constraint expression of the energy storage device.

[0146] Firstly, the charge-discharge power of the energy storage device is taken as the input, and substituted into the following formula to construct the charge-discharge power constraint expression of the energy storage device:

[0147] (19)

[0148] wherein: are the maximum charge-discharge power of the energy storage, respectively.

[0149] Then, the state of charge constraint expression of the energy storage device is constructed by substituting into the following formula:

[0150] (20)

[0151] wherein: is the state of charge of the energy storage device at the node at the moment .

[0152] Finally, the energy balance constraint expression of the energy storage device is constructed by taking the state of charge and the charge-discharge power of the energy storage device as the input and substituting into the following formula:

[0153] (21)

[0154] wherein: , are the maximum and minimum power storage ratio of the energy storage device corresponding to the node ; is the rated energy of the energy storage; is the production simulation time step.

[0155] (5) The operation power range constraint, the ramp rate constraint, the state of charge dynamic recursive constraint and the upper and lower limit boundary constraint of the adjustable load are integrated to construct the operation characteristic constraint expression of the adjustable load.

[0156] Firstly, the regulation power and the power baseline of the adjustable load are taken as inputs to construct the operation power range constraint expression of the adjustable load by substituting into the following formula:

[0157] (22)

[0158] In the formula: are the operation power limits of the adjustable load.

[0159] Then, the ramp rate constraint expression of the adjustable load is constructed by substituting into the following formula:

[0160] (23)

[0161] In the formula: , are the maximum rising and falling ramp rates of the adjustable load.

[0162] Further, the state of charge dynamic recursive constraint expression of the adjustable load is constructed by substituting into the following formula:

[0163] (24)

[0164] In the formula: is the state of charge of the adjustable load at the node at time ; is the state transition coefficient of the adjustable load at the node .

[0165] Finally, the upper and lower limit boundary constraint expression of the adjustable load is constructed by taking the state of charge of the adjustable load as input:

[0166] (25)

[0167] Step 2: Start-stop optimization space compression

[0168] This part aims at the high-dimensional discrete non-convex optimization problem caused by the unit start-stop behavior in the coordinated sequential production simulation model. The start-stop optimization space is preliminarily compressed and then compressed again to realize the dimension reduction of the model. The implementation process of the start-stop compression strategy is shown in Figure 3 . The specific steps are as follows:

[0169] Step 2.1 To narrow the search space of the unit start-stop optimization, the preliminary compression strategy of the start-stop optimization space based on the economic index of unit output is proposed by referring to the experience principle of giving priority to the economic unit in engineering practice. The specific steps are as follows:

[0170] First, substitute the following formula to construct the expression for the unit's output economic index.

[0171] (26)

[0172] In the formula: For at any time node The unit's output economic indicators.

[0173] Then, taking into account the impact of reserve demand and the fluctuation of new energy output on thermal power output, the upper and lower boundaries of total thermal power output in each period are estimated by substituting into the following formula.

[0174] (27)

[0175] (28)

[0176] In the formula: For time period Estimated upper and lower limits of total thermal power output; The system's reserve and spinning reserve factors are adjusted downwards to address the risks of power surplus and shortage. For time period Maximum wind and solar power curtailment; For time period Maximum load shedding power; For time period Maximum charging and discharging power of energy storage devices.

[0177] Furthermore, based on the output economic indicators of thermal power units All units are sorted in ascending order. If economic indicators are the same, a second ascending order is made based on minimum technical output, prioritizing units with greater operational flexibility. Following this priority, the unit output is sequentially set to its maximum technical output and accumulated. When the accumulated output first exceeds... If the operation stops immediately, the affected units are identified as forced-start optimization units. For the remaining units, continue to prioritize them according to the above priority, and set their output to the average of the maximum and minimum outputs, accumulating the values ​​until a certain threshold is reached. The units involved were identified as units awaiting startup and optimization.

[0178] The start-up and standby start-up optimization units retain their start-up and standby start-up status as 0-1 integer decision variables in subsequent optimizations. The remaining unselected units are determined to be mandatory shutdown units, and their start-up and standby status variables are fixed at 0 in the model and treated as known constants.

[0179] step 2.2 In the full space-time collaborative optimization framework, the approximate optimal solution of the output is obtained by solving the linear relaxation output simulation model to identify the thermal power generating units with obvious output tendency, and the start-stop secondary compression strategy is constructed accordingly, and the specific steps are as follows:

[0180] Firstly, considering the output cost of thermal power generating units, load shedding penalty, wind and light power penalty and regulation cost, the following formula is substituted to minimize the source-load-storage collaborative operation cost as the target, and the objective function expression of the linear relaxation output simulation model is constructed:

[0181] (29)

[0182] In the formula: is the objective function value of the linear relaxation output simulation model.

[0183] Then, ignoring all constraints containing only unit start-stop, the model constraint conditions are divided into the following two parts for construction: constraints containing only continuous variables and unit output constraints relaxing start-stop state variables.

[0184] Considering the system source-load balance, equipment operation boundary and other constraints, the constraint containing only continuous variables is constructed, and the specific constraint expression is referred to formula (8) in step 1.3 above, and formula (12)-(13), formula (18)-(25) in step 1.4.

[0185] Further, the start-stop state variables of the strong start-up and the to-be-started optimization units are fixed to 1, and the following formula is substituted to construct the unit output constraint expression of the relaxed start-stop state variable:

[0186] (30)

[0187] (31)

[0188] (32)

[0189] In the formula: is the set of units other than the fixed necessary shutdown units in step 2.1.

[0190] Finally, for the unit at time t whose output is greater than its minimum technical output times, its start-stop state at this time will be determined as the start-up state, and the corresponding start-stop state variable value will be fixed to 1 in the subsequent optimization as a constant variable.

[0191] step 3: reconstruction and solution of double-nested decomposition model based on optimality decomposition algorithm

[0192] This part mainly adopts Benders optimal decomposition algorithm, and reconstructs the coordinated scheduling production simulation model based on double nested decomposition structure. The model reconstruction and solution flow chart is shown in Figure 4 , and the structure diagram of double nested decomposition model is shown in Figure 5 . The specific steps are as follows:

[0193] Step 3.1 Based on the feasibility logic of avoiding line transmission overrun, decouple the constraints related to line transmission in the original model, and construct the source-load-storage coordinated scheduling output simulation main problem and line transmission overrun checking sub-problem. The specific steps are as follows:

[0194] (1) Construct the source-load-storage coordinated scheduling output simulation main problem model

[0195] Firstly, considering the output cost of thermal power unit, start-stop cost, load side load shedding penalty, renewable energy side wind and light power penalty and system regulation cost, the objective function expression of the source-load-storage coordinated scheduling output simulation main problem model is constructed, which is consistent with formula (7) in step 1.2.

[0196] Then, decouple the constraints related to line transmission overrun in the coordinated scheduling production simulation model constructed in step 1 (i.e. formula (9)-(11) constructed in step 1.3), and keep the system power balance constraints and the operation characteristics constraints of source, load and storage devices, which constitute the constraint conditions of the main problem model.

[0197] (2) Construct the line transmission overrun feasibility checking sub-problem model

[0198] Based on the output scheme provided by the main problem, construct the transmission overrun feasibility checking sub-problem. Adopt the direct current flow model, calculate the injection power value of each node in the transmission network, and combine with the limitation of line transmission capacity, substitute into the following formula, calculate the transmission capacity overrun value of each line in each period, and verify the feasibility of the power flow of all transmission lines in the system.

[0199] (33)

[0200] In the formula: , are the overrun powers of line transmission in positive and negative directions respectively.

[0201] When or , it indicates that the current output scheme leads to transmission overrun of the line, which needs to be suppressed in the next iteration.

[0202] step 3.2 In view of the strong coupling relationship between the unit start-stop state and the output decision in the source-load-storage co-scheduling output simulation main problem, and the significant mixed integer characteristics of the model due to a large number of 0-1 start-stop decision variables, the unit start-stop state integer variable and the output continuous variable are decoupled to construct a unit start-stop-output inner decomposition model. The specific steps are as follows:

[0203] (1) Construct the unit start-stop optimization main problem model

[0204] First, substitute the following formula to construct the objective function expression of the unit start-stop optimization main problem:

[0205] (34)

[0206] In the formula: is the objective function value of the unit start-stop optimization main problem; is the lower bound of the output optimization sub-problem objective function value under the start-stop scheme.

[0207] Then, considering the unit minimum start / stop time constraint, the start-stop state hard constraint of the must-on / must-off unit, the initial value constraint in the output optimization main problem objective function, and the feasibility / optimal cutting plane constraint generated by the sub-problem, substitute the following formula to constitute the constraint conditions of the unit start-stop optimization main problem:

[0208] (35)

[0209] In the formula: is the dual variable value vector obtained by dualizing the sub-problem; is the constant term vector in the sub-problem constraint; is the coefficient matrix of the coupling variables between the main-sub problems in the sub-problem constraint; is the coupling variable vector between the main-sub problems; , are the fixed must-off unit set in step 2.1 and the fixed must-on unit set in step 2.2, respectively.

[0210] (2) Construct the output optimization sub-problem model

[0211] First, under the unit start-stop state scheme given by the main problem, substitute the following formula to construct the output optimization sub-problem objective function expression:

[0212] (36)

[0213] In the formula: is the output optimization sub-problem objective function value.

[0214] Then, considering the constraints of system source-load balance, equipment operation boundary, etc., the specific constraint expressions refer to the above formula (8) in step 1.3, and formula (12)-(13), formula (18)-(25) in step 1.4. At the same time, the known start-stop state variable values are substituted into the unit output constraint (i.e. formula (14)-(16)) to form the output optimization sub-problem constraint. Further, the following formula is substituted to construct the feasibility / optimality cut plane constraint expression. In summary, the constraint conditions of the output optimization sub-problem model are formed.

[0215] (37)

[0216] Finally, solving the unit start-stop optimization main problem and the output optimization sub-problem, the source-load-storage output scheme of each period is obtained.

[0217] Step 3.3: Construct the convergence condition of the inner decomposition model to determine whether the inner decomposition model has reached convergence.

[0218] First, the following formula is substituted to construct the lower bound value and upper bound value expression of the objective function of the inner decomposition model in the iteration process:

[0219] (38)

[0220] In the formula: , are the lower bound and upper bound of the overall objective function of the inner decomposition model, respectively.

[0221] In each round of decomposition iteration, the current lower bound value and upper bound value of the main problem objective function are recorded. Then, the following formula is substituted to determine whether the absolute value of the difference between the lower bound and the upper bound is less than the preset convergence threshold, which is used as the convergence condition of the inner decomposition model.

[0222] (39)

[0223] In the formula: is the convergence threshold between the upper bound and the lower bound of the objective function.

[0224] Finally, if the above convergence condition is met, it is considered that the inner decomposition model has converged to its optimal solution. If the above convergence condition is not passed, the feasibility / optimality cut plane is generated according to the form of formula (37) in the sub-problem of step 3.2 step (2) and is fed back to the main problem model of step 3.2 step (1). Through iterative optimization of the unit start-stop scheme, the convergence condition is met.

[0225] Step 3.4: Construct the convergence condition of the outer decomposition model to determine whether the outer decomposition model has reached convergence.

[0226] Firstly, the transmission capacity of each line in all time periods is substituted into the following formula to determine whether the absolute value of the ratio between the transmission capacity limit value and the transmission capacity is less than the limit threshold value, as the convergence condition of the outer decomposition model.

[0227] (40)

[0228] In the formula: is the line transmission limit threshold value.

[0229] Then, it is determined whether the convergence condition is met. If the convergence condition is met, it is determined that the current solution is the final solution. Otherwise, the line transmission limit check sub-problem generated by step (2) in step 3.1 is substituted into the following formula to generate a feasible cut plane constraint, which is added to the source-load-storage co-simulation power output simulation main problem constructed in step 3.1:

[0230] (41)

[0231] In the formula: are the forward and reverse limit power vectors of the limit line t at the moment; is the active power injection vector of each node in the last generation in the iteration process; is the set of moments when the line occurs limit.

[0232] Finally, the solution of the main problem in step 3.1 is gradually corrected through the feedback of the cut plane constraint until the convergence condition is met.

[0233] Application Example

[0234] A simplified system of a provincial power grid is taken as an example to verify the effect of the method of the application. The topological structure of the starting system is shown in FIG. 6(a). The system has 90 nodes and 200 lines. In terms of power supply structure, 77.98 GW of thermal power plants, 30.22 GW of wind power plants, 65.34 GW of photovoltaic power plants, and 2 GW of energy storage stations are configured, and include 500 kV and 100 kV voltage levels. The load side includes 84.80 GW of basic load, 9.58 GW of air conditioning load, and 15.57 GW of electric hydrogen load.

[0235] Firstly, the co-simulation production simulation model is constructed according to step 1.

[0236] The population and economic data of the region were investigated, and a typical load curve was constructed as shown in FIG. 6(b). In combination with the resource distribution map of the province, the upper limit of wind power and photovoltaic power capacity of each node was calculated as shown in Table 1, and the typical wind power and photovoltaic power time series output curves were as shown in FIG. 6(c) and FIG. 6(d). Considering the current technical level of mainstream equipment manufacturers, the upper limit of power generation, regulation power and transmission capacity of thermal power, energy storage equipment, adjustable load and transmission line in each region was obtained through investigation, as shown in Table 1 and Table 2. The above data was substituted into the aforementioned equations (3)-(25) to complete the construction of the coordinated time series production simulation model.

[0237] Table 1 Upper limit of power generation of fire, wind, light, load and storage resources of each node

[0238]

[0239] Table 2 Transmission line capacity parameters

[0240]

[0241] The time scales of 4 days, 7 days, 30 days, 90 days and 180 days were selected for time series production simulation, covering short-term and medium-long-term demand. Then, in step 2, the start-stop optimization space compression strategy was adopted: in the preliminary compression stage, the upper and lower limit estimated values of the total output of thermal power units were set based on the actual value, and the upper and lower margin range was , according to which the units not in operation in the optimization interval were identified as necessary shutdown units and the start-stop state value was fixed at 0; in the secondary compression stage, the units with approximate output greater than the minimum technical output were identified as necessary start-up units, and the start-stop state value was fixed at 1. Taking the 7-day simulation scenario as an example, the effect of this start-stop compression strategy is shown in FIG. 6(e), where each cell represents a unit, and the dark blue square represents the necessary start-up unit and the light blue square represents the necessary shutdown unit.

[0242] Further, based on the compressed start-stop optimization space, the double-layer nested structure decomposition model is reconstructed and solved in step 3. First, in the outer main problem of source-load-storage coordinated scheduling output simulation, the inner decomposition model of unit start-stop-output optimization is constructed, and the convergence threshold is set to 0.01% of the initial lower bound of the objective function of the inner decomposition model, so as to ensure that the output scheme passed to the outer model has been highly approximated to the optimal solution in the compressed space. After the inner decomposition model is iteratively converged, the obtained source-load-storage output scheme is passed to the line transmission overrun feasibility check outer sub-problem, and the convergence threshold of the outer decomposition model is set to be less than the transmission capacity limit value of each line in all periods, so as to ensure the hard constraint requirement for safe operation of the power grid. Finally, after the outer decomposition model is iteratively converged, the final source-load-storage coordinated scheduling output simulation scheme satisfying all constraints is obtained. Taking the transmission overrun convergence process of line No. 46 in a 7-day simulation scenario as an example, the convergence curve of the outer decomposition model is shown in Fig. 6(f).

[0243] The method of solving the coordinated scheduling production simulation model as a whole (hereinafter referred to as the "original method") is used as a comparison, and the comparison results of the solving results of the original method and the method of the application are shown in Tables 3 and 4. The results show that in the simulation scenarios of 4 days, 7 days, 30 days and 90 days, the calculation time of the method of the application is 15.7%, 7.5%, 5.9% and 5.3% of that of the original method, and the optimal comprehensive cost difference is only 0.1%, 1.4%, 2.8% and 3.7%. In the long-time simulation scenario of 180 days, the original method is interrupted due to memory overflow, and the method of the application only needs 10.2% of the execution time of the original method, and can successfully obtain a feasible solution. This shows that the method of the application has achieved significant acceleration compared with the original method, and the calculation speed has been improved by 5-20 times. Under the premise of ensuring that the accuracy loss is less than 5%, the method effectively reduces the calculation complexity. In addition, the method can support time sequence simulation for up to 180 days, breaking through the memory limitation bottleneck faced by the original method. In summary, the method of the application significantly improves the calculation efficiency, obtains a similar time sequence production simulation scheme as the original method, greatly shortens the solving time and expands the scale of the solvable problem.

[0244] Table 3: Comparison of solving time

[0245]

[0246] Table 4: Comparison of optimal target values

[0247]

[0248] Example Two

[0249] A novel power system time sequence production simulation system based on model decomposition, comprising:

[0250] The co-scheduling production simulation model construction module is configured to determine a new power system operation boundary, take the system total operation cost optimization as an objective function, comprehensively consider system power balance constraints and line transmission overrun constraints, and introduce various types of device-level constraint conditions of sources, networks, loads and storages to construct a co-scheduling production simulation model;

[0251] The two-stage compression module is configured to combine engineering experience, construct a start-stop optimization space preliminary compression strategy with economic optimization as a target, solve the system output optimal scheme under the full space-time co-optimization framework through relaxation of start-stop, and realize secondary compression of start-stop.

[0252] The inner-outer layer decomposition model construction module is configured to decompose the co-scheduling production simulation model into a source-load-storage co-scheduling output simulation main problem model without line transmission constraints and a line transmission overrun feasibility checking sub-problem model, take the two as outer layer decomposition models, decouple unit start-stop state integer variables and output continuous variables, and construct a unit start-stop-output inner layer decomposition model.

[0253] The solving module is configured to optimize and solve the inner layer decomposition model and the outer layer decomposition model, and in the solving process of the outer layer decomposition model, an optimality decomposition algorithm is used to decouple unit start-stop state integer variables and output continuous variables until each decomposition model converges, and an optimal solution is obtained.

[0254] The processing flow of each model refers to the method provided in Embodiment 1.

[0255] Those skilled in the art should understand that embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, etc.) containing computer-usable program code. CD - ROM

[0256] The present application is described with reference to flowcharts and / or block diagrams according to the method, device (system), and computer program product of the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices produce a device that implements the functions described in the flowcharts and / or block diagrams. Figure 1 one flow or multiple flows and / or blocks Figure 1 ​means for performing the function specified by the block or blocks.

[0257] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the flow Figure 1 flow or flows and / or blocks Figure 1 means for performing the function specified by the block or blocks.

[0258] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the flow Figure 1 flow or flows and / or blocks Figure 1 steps of means for performing the function specified by the block or blocks.

[0259] The preferred embodiments herein disclosed merely by way of examples that can embody the principles of the application. Without departing from the spirit and principle of the application, one skilled in the art can make various changes and modifications of the application to adapt it to various usage and conditions. Therefore, the present application is not limited to the described embodiments but instead is defined by the appended claims in which:

Claims

1. A novel power system time series production simulation method based on model decomposition, characterized by, The method comprises the following steps: Determine the new power system operation boundary, take the system total operation cost optimization as the objective function, comprehensively consider the system power balance constraint and line transmission overrun constraint, and introduce various types of device level constraint conditions of source, network, load and storage to construct a coordinated time sequence production simulation model; Combine engineering experience, construct an initial compression strategy of start-stop optimization space with economic optimization as the target, solve the optimal system output scheme under the full space-time coordinated optimization framework by relaxing start-stop, and realize secondary compression of start-stop; Decompose the coordinated time sequence production simulation model into a source-load-storage coordinated time sequence output simulation main problem model without line transmission constraint and a line transmission overrun feasibility checking sub-problem model, take the two as outer decomposition models, decouple the unit start-stop state integer variable and output continuous variable, and construct a unit start-stop-output inner decomposition model; Optimize and solve the inner decomposition model and the outer decomposition model, and in the process of solving the outer decomposition model, decouple the unit start-stop state integer variable and the output continuous variable by using an optimality decomposition algorithm until the convergence of each decomposition model is achieved, and an optimal solution is obtained; The process of combining engineering experience to construct an initial compression strategy of start-stop optimization space with economic optimization as the target comprises the following steps: constructing an expression of unit output economic index, comprehensively considering the influence of reserve demand and new energy output fluctuation on thermal power output, estimating the upper and lower boundaries of total thermal power output in each period, sorting all units in ascending order according to the output economic index of the thermal power unit, if the economic indexes are the same, performing secondary ascending sorting according to the minimum technical output, and preferentially selecting units with stronger operation flexibility; According to the priority, the unit output is set to the maximum technical output and is accumulated, and when the cumulative output first exceeds the upper boundary of the total thermal power output, the involved units are determined as strong start-up optimization units; For the remaining units, continue to sort them according to the above priority, set the output to the average of the maximum and minimum output and accumulate it, until the lower boundary of the total thermal power output is reached, and the involved units are determined as standby start-up optimization units; The process of solving the optimal system output scheme under the full space-time coordinated optimization framework by relaxing start-stop to realize secondary compression of start-stop comprises the following steps: Comprehensively consider the output cost of the thermal power unit, the load shedding penalty, the wind and light power penalty and the regulation cost, construct an expression of the objective function of the linear relaxed output simulation model with the minimum source-load-storage coordinated operation cost as the target; Ignore all constraints containing only unit start-stop, and take the constraints containing only continuous variables and the unit output constraints with relaxed start-stop state variables as the model constraint conditions; Fix the start-stop state variables of the strong start-up and standby start-up optimization units to 1, and construct an expression of the unit output constraint of the relaxed start-stop state variable; For the unit whose output at the target time is greater than the set value, the start-stop state at the target time will be determined as the start-up state, and the corresponding start-stop state variable value will be fixed to 1 in the subsequent optimization and treated as a constant variable.

2. A novel power system time series production simulation method based on model decomposition according to claim 1, characterized in that, The process of determining the operation boundary of the new power system includes: determining the operation boundary of the new power system, calculating the maximum load of each node, combining the typical load day curve with the annual day maximum load change curve, reconstructing, and forming the input demand in the time sequence production simulation; According to the average available density of wind and light resources in each region, the available generation boundary of wind and light resources in the region is calculated, and the typical wind and light power time sequence of each region is constructed; Determine the spatial distribution, capacity scale and power generation technology parameters of each regional thermal power unit; Obtain the rated transmission capacity of the transmission line; Obtain the adjustable capacity proportion, regulation duration and response speed parameters of the demand side adjustable load resource in different time periods.

3. A novel power system time series production simulation method based on model decomposition according to claim 1, characterized in that, The process of taking the optimal total system operation cost as the objective function includes: the total system operation cost includes thermal power unit output cost, thermal power unit start-stop cost, load shedding penalty, abandoned power penalty and demand side resource regulation cost.

4. A novel power system time series production simulation method based on model decomposition according to claim 1, characterized in that, Comprehensively consider the system power balance constraint and line transmission overrun constraint, and introduce the constraint conditions of various types of source, network, load and storage device level, the process includes: calculating the total power generation and total load power, forming the source and load power balance constraint that the total power generation is equal to the total load; According to the node power balance relationship, the active power expression of each node injected into the transmission network is constructed, the relationship expression between line power flow and node injection power is constructed by using the direct current flow model, and the line transmission overrun constraint expression is constructed according to the line power flow; Taking the load of each node and the output of wind and light units, as well as the load shedding power and the abandoned power as inputs, the standby demand constraint expression of thermal power unit is constructed; Comprehensively consider the active power output range constraint of thermal power unit, the active power output range constraint considering standby output, the ramp rate constraint and the minimum start / stop time constraint, and construct the operation characteristic constraint expression of thermal power unit; After normalizing the wind and light time sequence output curve, taking the wind and light time sequence output curve as input, the active power output constraint expression of wind and light generator is constructed; Comprehensively consider the charge and discharge power constraint, the state of charge range constraint and the energy balance constraint of the energy storage device, and construct the operation characteristic constraint expression of the energy storage device; Comprehensively consider the operation power range constraint, the ramp rate constraint, the state of charge dynamic recursion constraint and the upper and lower limit boundary constraint of the adjustable load, and construct the operation characteristic constraint expression of the adjustable load.

5. A novel power system time series production simulation method based on model decomposition according to claim 1, characterized in that, The process of decomposing the coordinated time sequence production simulation model into a source-load-storage coordinated time sequence output simulation main problem model without line transmission constraint and a line transmission overrun feasibility checking sub-problem model includes: Comprehensively consider the output cost of thermal power unit, start-stop cost, load shedding penalty of load side, wind and light power penalty of renewable energy side and system regulation cost, construct the objective function expression of the source-load-storage coordinated time sequence output simulation main problem model, and the objective function expression of the coordinated time sequence production simulation model is the same; Decouple the constraints related to line transmission overrun in the constructed coordinated time sequence production simulation model, retain the system power balance constraint and the operation characteristic constraint of source, load and storage devices, and jointly constitute the constraint conditions of the main problem model; The transmission capacity limit value of each line in each time period is calculated by using a direct current power flow model, combining with the line transmission capacity limit, and the transmission capacity limit value of each line in each time period is calculated, the transmission capacity limit value of each line in each time period is calculated, and the transmission capacity limit value of each line in each time period is calculated.

6. A novel power system time series production simulation method based on model decomposition according to claim 1, characterized in that, The process of decoupling the integer variable of unit start-stop state and the continuous variable of output includes: Constructing an objective function expression of the main problem of the unit start-stop optimization, which is the minimization of the sum of the total start-stop cost of the thermal power unit and the lower bound of the objective function value of the output optimization sub-problem under the target start-stop scheme ​ The minimum start / stop time constraint of the unit, the start / stop state hard constraint of the must-on / must-off unit, the initial value constraint, and the feasibility / optimality cut plane constraint generated by the sub-problem constitute the constraint conditions of the unit start / stop optimization main problem. initial value constraint, and the feasibility / optimality cut plane constraint generated by the sub-problem constitute the constraint conditions of the unit start / stop optimization main problem. Under the given unit start-stop state scheme of the main problem, the objective function expression of the output optimization sub-problem is constructed, which is specifically to minimize the sum of the total output cost of thermal power units, the total load shedding penalty of the system, the total abandoned electricity penalty of wind and solar units, and the total adjustment cost of demand side resources; Considering the system source-load balance and equipment operation boundary constraints, the known start-stop state variable value is substituted into the unit output constraint to form the constraint condition of the output optimization sub-problem model, and the source-load-storage output scheme of each time period is obtained by solving the unit start-stop optimization main problem and the output optimization sub-problem.

7. A novel power system time series production simulation method based on model decomposition according to claim 1, characterized in that, The process of judging the convergence of the inner layer decomposition model includes: The lower bound and upper bound expressions of the objective function of the inner layer decomposition model in the iteration process are constructed, and the lower bound and upper bound of the objective function of the main problem are recorded in each round of decomposition iteration. It is judged whether the absolute value of the difference between the lower bound and the upper bound is less than the preset convergence threshold. If it is satisfied, it is considered that the inner layer decomposition model has converged to its optimal solution; otherwise, according to the constraint condition of the output optimization sub-problem model, a feasibility / optimality cut plane is generated and fed back to the main problem model, and the unit start-stop scheme is iteratively optimized until the convergence condition is met.

8. A novel power system time series production simulation method based on model decomposition according to claim 1, characterized in that, The process of judging the convergence of the outer layer decomposition model includes: The transmission capacity of all lines in all time periods is substituted into the following formula to judge whether the absolute value of the ratio between the two is less than the limit threshold. If so, the current solution is the final solution; otherwise, a feasible cut plane constraint is generated according to the constructed line transmission limit checking sub-problem, and it is added to the constructed source-load-storage coordinated time sequence output simulation main problem. Through the feedback of the cut plane constraint, the solution of the main problem model is gradually modified until the convergence condition is met.

9. A novel power system time series production simulation system based on model decomposition, applying the method of any one of claims 1-8, characterized in that, It includes: The coordinated time sequence production simulation model construction module is configured to determine the operation boundary of the new power system, taking the optimal total operation cost of the system as the objective function, considering the system power balance constraint and line transmission limit constraint, and introducing various device-level constraint conditions of source, grid, load and storage, to construct the coordinated time sequence production simulation model; The two-stage compression module is configured to combine engineering experience to construct a start-stop optimization space preliminary compression strategy with economic optimality as the target, and to realize secondary compression of start-stop by relaxing start-stop to solve the optimal system output scheme under the full space-time coordination optimization framework; The inner and outer layer decomposition model construction module is configured to decompose the coordinated time sequence production simulation model into a source-load-storage coordinated time sequence output simulation main problem model without line transmission constraint and a line transmission limit feasibility checking sub-problem model, to take them as the outer layer decomposition model, to decouple the integer variable of unit start-stop state and the continuous variable of output, and to construct the unit start-stop-output inner layer decomposition model; The solving module is configured to solve the inner-layer decomposition model and the outer-layer decomposition model, and in the solving process of the outer-layer decomposition model, an optimality decomposition algorithm is used to decouple the integer variable of the start-stop state of the unit and the continuous variable of the output until each decomposition model converges, and an optimal solution is obtained.

Images