Power production simulation and optimal configuration method, equipment and product based on numerical simulation

Abstract

The invention discloses a power production simulation and optimal configuration method, device and product based on numerical simulation, and relates to the technical field of new energy, and the method comprises the steps: constructing an investment cost evolution model according to related historical data and future planning data; constructing a capacity evolution sequence generation model according to unit capacity investment cost evolution prediction data output by the investment cost evolution model; generating a plurality of operation scenes of the power system according to the related historical data; constructing a hydrogen-containing long-time energy storage power system time sequence production simulation model according to a given capacity evolution sequence output by the capacity evolution sequence generation model and multiple operation scenes; constructing a data-model hybrid drive model according to the capacity evolution sequence generation model and the power system time sequence production simulation model; constructing a low-carbon evolution path simulation model of the power system; and obtaining a power system low-carbon evolution path considering hydrogen long-time energy storage according to the power system low-carbon evolution path simulation model. The utilization rate of new energy is improved.

Description

Technical Field

[0001] The present application relates to the field of new energy technology, and in particular to a method, device, medium and product for simulating a low-carbon evolution path of a power system. Background Art

[0002] Under the guidance of the dual carbon goals, the current power system is developing in the direction of clean and low-carbon. The power system should minimize the investment and construction of coal-fired power units and replace the output of traditional power units with clean energy units. This will aggravate the dilemma of unstable power supply. At the same time, large-scale grid connection with a high proportion of new energy will bring more severe new energy consumption problems. In order to improve the utilization rate of new energy, the use of long-term energy storage represented by hydrogen energy systems has excellent prospects and plays an important role in solving the dilemma of new energy consumption and accelerating the low-carbon transformation of power systems. Therefore, in the process of the power system developing into a system that integrates multiple factors of electricity, hydrogen and carbon, how to balance the system's economy, low carbon and cleanliness, and realize the simulation of the evolutionary path driven by multi-dimensional factors of the power system, is the key issue of this application. Summary of the invention

[0003] The purpose of this application is to provide a method, equipment, medium and product for simulating the low-carbon evolution path of an electric power system, which can improve the utilization rate of new energy while balancing the economy, low carbon and cleanliness of the system during the development of the electric power system towards a system that integrates multiple factors of electricity, hydrogen and carbon.

[0004] To achieve the above objectives, this application provides the following solutions:

[0005] In a first aspect, the present application provides a method for simulating a low-carbon evolution path of a power system, comprising:

[0006] Obtain relevant historical data on energy costs in the target area, future planning data, the upper limit of the construction scale of equipment corresponding to each type of energy, the current scale of constructed equipment and the planned retirement time; the future planning data includes the installed capacity, future planned capacity and future planned R&D investment of each type of energy; the relevant historical data includes the installed capacity, R&D investment and unit capacity investment cost of each type of energy in previous years;

[0007] Based on the relevant historical data and the future planning data, construct an investment cost evolution model that takes into account the uncertainty of technological innovation breakthrough rate;

[0008] According to the unit capacity investment cost evolution prediction data output by the investment cost evolution model, the upper limit of the constructible scale of the equipment corresponding to each type of energy, and the scale of the currently constructed equipment and the planned retirement time, a capacity evolution sequence generation model is constructed with minimizing the sum of investment cost and maintenance cost as the objective function and with equipment capacity connection constraints and capacity limit constraints as constraints;

[0009] Based on the relevant historical data, a k-means clustering algorithm is used to generate various operation scenarios of the power system;

[0010] According to the given capacity evolution sequence output by the capacity evolution sequence generation model and the various operation scenarios of the power system, a power system time-series production simulation model containing hydrogen long-term energy storage is constructed with carbon emission limit constraint and clean energy electricity proportion constraint as core constraints and minimizing the sum of operation cost and carbon trading cost as the objective function;

[0011] According to the capacity evolution sequence generation model and the power system time-series production simulation model containing hydrogen long-term energy storage, a data-model hybrid driving model capable of measuring system cleanliness and low carbon is constructed by using a convolutional neural network-bidirectional gated recurrent unit-attention mechanism; the data-model hybrid driving model takes the capacity evolution sequence output by the capacity evolution sequence generation model as input, and takes the corresponding unit electricity share marginal cost and unit carbon limit marginal cost as output;

[0012] According to the marginal cost per unit electricity share and the marginal cost per unit carbon limit, the sum of the minimized investment cost and maintenance cost, and the sum of the minimized operating cost and carbon trading cost, a low-carbon evolution path simulation model of the power system considering hydrogen long-term energy storage involving economic cost driving factors, clean energy electricity share driving factors, and carbon emission limit driving factors is constructed;

[0013] According to the simulation model of the low-carbon evolution path of the power system, with the goal of minimizing the comprehensive benefit cost, a two-layer solution framework including the capacity layer and the operation layer is proposed. With the goal of minimizing the comprehensive benefit cost, the capacity layer adopts the PID search algorithm, and the operation layer adopts the improved differential evolution algorithm, so as to obtain the low-carbon evolution path of the power system considering hydrogen long-term energy storage.

[0014] Optionally, according to the relevant historical data and the future planning data, based on a two-factor learning curve model and taking into account the uncertainty of the technological innovation breakthrough rate, an investment cost evolution model that takes into account the uncertainty of the technological innovation breakthrough rate is constructed.

[0015] Optionally, the investment cost evolution model is:

[0016] α y =α0A y B y ;

[0017]

[0018] Among them, α yis the unit capacity investment cost in the yth year; α0 is the initial unit capacity investment cost; A y B is the degree of decrease in unit capacity investment cost caused by the growth of installed capacity in year y; y N is the degree of decrease in unit capacity investment cost caused by the growth of R&D investment in year y; y is the cumulative installed capacity scale in the yth year; N0 is the initial installed capacity scale; a is the learning rate exponent of the cumulative installed capacity scale; u y It is a binary variable, which takes 1 when there is a breakthrough in technological innovation, and takes 0 when there is no breakthrough in technological innovation; C y is the cumulative R&D investment in the yth year; C0 is the initial R&D investment; b is the learning rate index of the cumulative R&D investment; B y-1 It is the degree of decrease in unit capacity investment cost caused by the increase in R&D investment in year y-1.

[0019] Optionally, the objective function of the capacity evolution sequence generation model is:

[0020] minF1=min(F inv +F main );

[0021]

[0022] Among them, F1 is the cost of the capacity evolution sequence generation model; F inv is the investment cost; F main For maintenance cost; kn For the newly added device collection; Ω y For planning year collection; is the unit capacity investment cost of each type of energy device; P kn,y is the installed capacity of the new devices added in year y; E is the unit energy investment cost of each type of energy device; kn,y is the energy of the newly added device in year y; F sa The residual value of the equipment that has not been retired within the planning period; Ω k is the collection of online energy devices; ρ k is the residual value recovery factor; is the unit capacity investment cost at the end of the planning period; P k,Y The installed capacity in the final year of the planning period; E is the unit energy investment cost at the end of the planning period; k,Y The energy for the last year of the planning period; H k,Y To plan the remaining life of various types of energy devices at the end of the project; k is the service life of each type of energy device; Y0 is the time when the energy device begins to be put into use; The unit capacity maintenance cost of each type of energy device; is the unit energy maintenance cost of each type of energy device; Y is the last year of the planning period.

[0023] Optionally, the objective function of the power system time-series production simulation model containing hydrogen long-term energy storage is:

[0024] minF2=min(F om +F co );

[0025]

[0026] Among them, F2 is the cost of the time series production simulation model; F om is the operating cost; F co is the carbon trading cost; k is the collection of online energy devices; Ω y For planning year collection;Ω s To run the scene collection; Ω t It is a collection of daily running times; is the unit coal consumption cost; is the coal consumption; is the unit natural gas consumption cost; is the natural gas consumption; Unit cost of hydrogen consumption; is the hydrogen consumption; Δt is the time interval; The startup cost of thermal power units (including coal-fired and gas-fired); For downtime costs; is the carbon trading price in year y; CO y is the total carbon trading volume in year y.

[0027] Optionally, according to the capacity evolution sequence generation model and the power system time series production simulation model containing hydrogen long-term energy storage, a convolutional neural network-bidirectional gated recurrent unit-attention mechanism is used to construct a data-model hybrid driving model capable of measuring system cleanliness and low carbon performance, specifically including:

[0028] Applying a random function to generate a capacity evolution random sequence based on the capacity evolution sequence generation model;

[0029] Inputting the capacity evolution random sequence into the power system time-series production simulation model containing hydrogen long-term energy storage to obtain the minimized operating cost and carbon trading cost;

[0030] According to the minimized operating costs and carbon trading costs, by adjusting the clean energy electricity share or carbon emission limit, the corresponding marginal cost per unit electricity share and the marginal cost per unit carbon limit are calculated;

[0031] With the capacity evolution random sequence as input and the corresponding marginal cost per unit electricity share and the marginal cost per unit carbon limit as output, the convolutional neural network-bidirectional gated recurrent unit-attention mechanism is trained to obtain a data-model hybrid driven model capable of measuring the cleanliness and low-carbon nature of the system.

[0032] Optionally, the simulation model of the low-carbon evolution path of the power system considering long-term hydrogen energy storage is:

[0033] V=ω F V F +ω new V new +ω co V co ;

[0034] Among them, V is the comprehensive benefit cost value; ω F is the weight coefficient driven by economic cost; V F is the benefit cost value driven by economic cost; new is the weight coefficient driven by the proportion of clean energy electricity; V new is the benefit cost value driven by the proportion of clean energy electricity; ω co is the weight coefficient driven by carbon emission limit; V co is the benefit-cost value driven by carbon emission limits.

[0035] In a second aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of any one of the above-mentioned methods for simulating the low-carbon evolution path of a power system.

[0036] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-mentioned methods for simulating the low-carbon evolution path of a power system.

[0037] In a fourth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of any of the above-mentioned methods for simulating the low-carbon evolution path of a power system.

[0038] According to the specific embodiments provided in this application, this application has the following technical effects:

[0039] The present application provides a method, equipment, medium and product for simulating the low-carbon evolution path of a power system. First, historical data related to energy costs in a target area and the degree of future development are obtained; an investment cost evolution model that takes into account the uncertainty of the technological innovation breakthrough rate is constructed; based on historical data related to energy costs, a differential evolution algorithm is used to solve model parameters, and then the investment cost per unit capacity of energy is predicted, and the upper limit of the construction scale of various types of equipment in the target area, as well as the current scale of constructed equipment and the planned retirement time are obtained. A capacity evolution sequence generation model is constructed in combination with the cost prediction data, with the objective function being to minimize the sum of investment cost and maintenance cost, and the constraints being the equipment capacity connection constraint and the capacity limit constraint. Then, the actual operation historical data of the electric load and wind and photovoltaic output in the target area are obtained, and the k-means clustering algorithm is used to generate the basic operation scenario; with the goal of minimizing the sum of the operation cost and the carbon trading cost, a power system time-series production simulation model with long-term hydrogen energy storage is constructed with the carbon emission limit constraint and the clean energy electricity proportion constraint as the core constraints. The objective function of this model is to minimize the sum of the operation cost and the carbon trading cost. The constraints are mainly composed of the electric energy module constraint, the hydrogen energy module constraint, the carbon volume module constraint and the driving factor constraint. Finally, based on the capacity evolution sequence generation model and the power system time-series production simulation model with long-term hydrogen energy storage, a data-model hybrid driving model with the ability to measure the cleanliness and low-carbon nature of the system is constructed; further integrating the economic cost driving factors, the clean energy electricity proportion driving factors and the carbon emission limit driving factors, a low-carbon evolution path simulation model of the power system considering long-term hydrogen energy storage and a solution framework are constructed. For the established model, the hierarchical optimization idea is adopted, and a two-layer solution framework is proposed. The goal is to minimize the comprehensive benefit cost. The capacity layer adopts the PID search algorithm, and the operation layer adopts the improved differential evolution algorithm to output the evolution results, forming a low-carbon evolution path for the power system considering hydrogen long-term energy storage. The simulation results of the low-carbon evolution path of the power system considering hydrogen long-term energy storage can improve the utilization rate of new energy while balancing the system's economy, low carbon and cleanliness during the development of the power system towards a system that integrates multiple factors of electricity, hydrogen and carbon. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 A schematic diagram of a framework of a method for simulating a low-carbon evolution path of a power system taking into account long-term hydrogen energy storage provided in an embodiment of the present application;

[0042] Figure 2 A schematic diagram of a specific implementation process of a method for simulating a low-carbon evolution path of a power system provided in one embodiment of the present application;

[0043] Figure 3 A schematic diagram of a flow chart of a method for simulating a low-carbon evolution path of a power system provided in one embodiment of the present application;

[0044] Figure 4 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0046] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0047] The embodiment of the present application provides a method for simulating the low-carbon evolution path of a power system, such as Figures 1 to 3 As shown, including:

[0048] Step S1: Obtain relevant historical data on energy costs in the target area, future planning data, the upper limit of the buildable scale of equipment corresponding to each type of energy, the current scale of constructed equipment and the planned retirement time; the future planning data include the installed capacity, future planned capacity and future planned R&D investment of each type of energy; the relevant historical data include the installed capacity, R&D investment and unit capacity investment cost of each type of energy in previous years.

[0049] Specifically, according to the relevant historical data and the future planning data, based on the two-factor learning curve model and taking into account the uncertainty of the technological innovation breakthrough rate, an investment cost evolution model that takes into account the uncertainty of the technological innovation breakthrough rate is constructed.

[0050] Step S2: Based on the relevant historical data and the future planning data, an investment cost evolution model that takes into account the uncertainty of the technological innovation breakthrough rate is constructed.

[0051] In practical applications, first, obtain historical data related to energy costs in a certain region, including the installed capacity, R&D investment, and unit capacity investment cost of various types of energy in previous years; obtain the future development of the installed capacity of various types of energy in a certain region and the future R&D investment intensity.

[0052] Secondly, part of the reason for the decline in the unit capacity investment cost of various types of energy is the result of accumulated experience, and the accumulated experience is regarded as the result of the growth of energy installed capacity; the other part is the result of accumulated R&D, and the accumulated R&D is regarded as the result of the growth of R&D investment in the energy industry. Based on the above historical data, based on the two-factor learning curve model, and taking into account the uncertainty of the technological innovation breakthrough rate, an investment cost evolution model considering the uncertainty of the technological innovation breakthrough rate is constructed. The model is as follows:

[0053] α y =α0A y B y (1)

[0054]

[0055] In formula (1): A y B is the degree of decrease in unit capacity investment cost caused by the growth of installed capacity in year y, y is the degree of decrease in unit capacity investment cost caused by the growth of R&D investment in year y, α0 is the initial unit capacity investment cost, α y is the unit capacity investment cost in year y. In formula (2): N y is the cumulative installed capacity scale in the yth year, N0 is the initial installed capacity scale, and a is the learning rate exponent of the cumulative installed capacity scale. In formula (3): C y is the cumulative R&D investment in the yth year, C0 is the initial R&D investment, b is the learning rate index of the cumulative R&D investment; y It is a binary variable, which takes the value of 1 when there is a breakthrough in technological innovation, and takes the value of 0 when there is no breakthrough in technological innovation.

[0056] When R&D investment accumulates to a certain level and the technological breakthrough rate reaches the target, R&D investment B y The benefit of reducing unit capacity cost begins to play a role. The relationship between the technology breakthrough rate and the equivalent R&D cost is established, which can generally be represented by a piecewise linear function. At the same time, the envelope uncertainty model is used to describe the uncertain relationship between the technology breakthrough rate and the R&D cost. The time benefit of the R&D cost is further considered, and the specific model is as follows:

[0057]

[0058]

[0059] In formula (4): For equivalent cumulative R&D investment, is the technological innovation breakthrough rate under equivalent cumulative R&D investment. In formula (5), σ is the boundary indicator representing the uncertainty of technological innovation breakthrough rate, π y is the actual technological innovation breakthrough rate under uncertainty. In formula (6), θ is the time benefit parameter of R&D cost, c y The new R&D investment in year y; is the equivalent cumulative R&D investment in year y+1.

[0060] Finally, based on historical data and with the goal of minimizing variance, the differential evolution algorithm is used to solve the model parameters a, b, θ, and σ. According to the future development of the installed capacity of each type of energy and the data on R&D investment, the investment cost per unit capacity of energy is predicted as the equipment cost data for the capacity evolution sequence generation model.

[0061] Step S3: Based on the unit capacity investment cost evolution prediction data output by the investment cost evolution model, the upper limit of the buildable scale of the equipment corresponding to each type of energy, and the current scale of constructed equipment and the planned retirement time, a capacity evolution sequence generation model is constructed with the objective function of minimizing the sum of investment cost and maintenance cost and with equipment capacity connection constraints and capacity limit constraints as constraints.

[0062] In practical applications, first, we need to obtain the upper limit of the construction scale of each type of equipment in a certain area, as well as the current scale of constructed equipment and the planned retirement time.

[0063] Secondly, a capacity evolution sequence generation model is constructed, with the objective function of minimizing the sum of investment cost and maintenance cost, and the constraints are equipment capacity connection constraint and capacity limit constraint. In addition to the historical data in step S1, the model input data also includes the equipment cost data in step S2.

[0064] Objective function:

[0065] min F1=min(F inv +F main ) (9)

[0066]

[0067] In formula (9), F1 is the cost of the capacity evolution sequence generation model, which mainly includes the investment cost F inv and maintenance cost F main In formula (10): is the unit capacity investment cost of each type of energy device, is the unit energy investment cost of each type of energy device, P kn,y and E kn,y is the installed capacity and energy of the new devices in year y, Fsa is the residual value of the equipment that has not been retired within the planning period. In formula (11): k is the residual value recovery coefficient, and is the unit capacity investment cost and unit energy investment cost at the end of the planning period, P k,Y and E k,Y is the installed capacity and energy at the end of the planning period, H k,Y To plan the remaining life of various types of energy devices at the end of the project, T k is the service life of each type of energy device. In formula (12): Y0 is the time when the energy device starts to be put into use. In formula (13): and It is the unit capacity maintenance cost and unit energy maintenance cost of each type of energy device.

[0068] The constraints include equipment capacity connection constraints and equipment capacity limit constraints. Among them, the equipment capacity connection constraints are:

[0069] Initial equipment will be decommissioned after reaching its operating life. At the same time, energy devices installed in a certain stage will continue to operate in subsequent stages and will be decommissioned after reaching the end of their life.

[0070] P k,y =P k,y-1 +P kn,y -P kr,y (14)

[0071]

[0072] In formula (14) to formula (15): P k,y is the installed capacity online in year y, P k,y-1 is the installed capacity online in year y-1; P kn,y is the newly added installed capacity at the beginning of year y, P kr,y is the installed capacity retired at the beginning of year y; For the yT k The installed capacity online in the year.

[0073] The equipment capacity limit constraints are:

[0074] B k,y ≤P k,y ≤A k,y (16)

[0075] In formula (16): B k,y and A k,y They are the lower and upper limits of installed capacity of energy devices in year y respectively.

[0076] Step S4: Based on the relevant historical data, a k-means clustering algorithm is used to generate multiple operation scenarios of the power system.

[0077] In practical applications, the actual operation historical data of the power load and wind and photovoltaic output in a certain area are obtained. For the historical data of wind and photovoltaic output and power load, the historical data are divided into four parts: spring, summer, autumn and winter according to the weather characteristics of the region. The k-means clustering algorithm is used to cluster the multi-day data in each quarter into one cluster, and the four basic scenarios of power system operation are obtained. Based on this, the power load forecast data for future years can also be obtained according to the power growth ratio.

[0078] Step S5: Based on the given capacity evolution sequence output by the capacity evolution sequence generation model and the various operating scenarios of the power system, a time-series production simulation model of a power system containing hydrogen long-term energy storage is constructed with carbon emission limit constraints and clean energy electricity proportion constraints as core constraints and minimizing the sum of operating costs and carbon trading costs as the objective function.

[0079] In practical applications, under a given capacity evolution sequence, a power system time-series production simulation model with long-term hydrogen energy storage is constructed with carbon emission limit constraints and clean energy electricity proportion constraints as core constraints. The objective function of the model is to minimize the sum of operating costs and carbon trading costs. The constraints are mainly composed of power module constraints, hydrogen module constraints, carbon module constraints, and driving factor constraints.

[0080] The model objective function is:

[0081] min F2=min(F om +F co ) (17)

[0082]

[0083] In formula (17): F2 is the cost of the time series production simulation model, which mainly includes the operating cost F om and carbon trading costs F co In formula (18): and They are coal consumption, natural gas consumption and hydrogen consumption; and They are unit coal consumption cost, unit natural gas consumption cost and unit hydrogen consumption cost respectively; and is the startup cost and shutdown cost of thermal power units (including coal-fired and gas-fired). In formula (19): is the carbon trading price in year y, CO yis the total carbon trading volume in year y. In addition, Ω k is the collection of online energy devices, Ω kn For the newly added device set, Ω kr is the decommissioned device collection, Ω y is the planning year set, Ω s To run the scene set, Ω t is the daily operation time set, Δt is the time interval. In this model, 9 energy devices are considered, Ω k-g It is a collection of online energy devices for thermal power units, consisting of coal-fired thermal power units Ω k-coal Gas-fired power plant k-gas Composition, Ω k-pv For photovoltaic units online energy device collection, Ω k-wind For wind turbine online energy device collection, Ω k-ess Online energy device collection for electrochemical energy storage devices, Ω k-el Online energy device assembly for electrolyzer, Ω k-fc Fuel cell online energy device collection, Ω k-ht Online energy device assembly for hydrogen storage tanks and Ω k-ccs Online energy device assembly for carbon capture and storage.

[0084] The constraints of the electric energy module in the model include electrochemical energy storage operation constraints, coal-fired units and gas-fired units operation constraints, and wind turbine units and photovoltaic units operation constraints.

[0085] Electrochemical energy storage operation constraints:

[0086]

[0087]

[0088] In formula (20): The external output power of electrochemical energy storage is determined by the charging power of electrochemical energy storage. And electrochemical energy storage discharge power Formula (21) is the upper and lower power limit constraints of electrochemical energy storage, and is the lower and upper limits of the electrochemical energy storage charge and discharge power. In formula (22): is the state of charge of the electrochemical energy storage at the tth moment, is the rated energy of electrochemical energy storage, η ess,in With η ess,out It is the charging efficiency and discharging efficiency of electrochemical energy storage; is the state of charge of the electrochemical energy storage at the t-1th moment; k, y, s, t-1 represent the kth device, the yth year, the sth operating scenario and the t-1th moment respectively; tN is the end time; t0 is the initial time. In formula (23): and is the minimum state of charge and maximum state of charge of electrochemical energy storage. In formula (24): and It is the charge state of the electrochemical energy storage device at the initial moment and the charge state at the final moment in the daily scheduling cycle.

[0089] Operation constraints of coal-fired / gas-fired power units:

[0090]

[0091]

[0092]

[0093]

[0094] In formula (25) to formula (26): and is the output power of coal-fired power units and gas-fired power units, and is the unit consumption output power conversion coefficient between coal-fired power units and gas-fired power units, The amount of coal input to the coal-fired power unit; is the input natural gas volume of the gas-fired power plant. In equations (27) to (30): u k,y,s,t-1 is the on / off status of the thermal power unit at time t-1; is the startup cost of the thermal power unit at time t; is the shutdown cost of the thermal power unit at the tth moment; u k,y,s,t is the on / off state of the thermal power unit at time t, TS and TO are the minimum shutdown time and minimum operation time of the thermal power unit; U k and D k is the single startup cost and single shutdown cost of the thermal power unit. In formula (31) to formula (33): is the output power of the thermal power unit at the tth moment, is the installed capacity of thermal power units, is the output power of the thermal power unit at time t-1; and is the minimum and maximum technical output of the thermal power unit, r k,y and d k,y It is the upward climbing rate and downward descending rate of the thermal power unit.

[0095] Wind power / photovoltaic unit operation constraints:

[0096]

[0097]

[0098] In formula (34) to formula (35): and is the actual output power of wind turbines and photovoltaic units, and is the predicted electric power of wind turbines and photovoltaic units, and It is the installed capacity of wind turbines and photovoltaic units.

[0099] The hydrogen energy storage system has the same energy storage function as other energy storage devices such as batteries. When the wind and solar power output power is greater than the load demand, the electrolyzer consumes the surplus electricity, generates hydrogen by electrolyzing water, and stores the hydrogen in the hydrogen storage tank, which is equivalent to increasing the electrical load; when the wind and solar power output power is less than the load demand, the fuel cell generates electricity to meet the load demand, improving the reliability of the system. The hydrogen energy module constraints include the electrolyzer operation model, the fuel cell operation model, and the hydrogen storage tank operation model.

[0100] Electrolyzer operation model:

[0101]

[0102] In formula (36): is the output hydrogen of the electrolyzer device, is the input electrical power of the electrolyzer device, is the electric-hydrogen conversion coefficient of the electrolyzer device. In formula (37): The maximum input power of the electrolyzer device depends on the installed capacity of the electrolyzer device. and the situation of hydrogen storage tanks. is the maximum hydrogen storage capacity of the hydrogen storage tank, is the actual hydrogen storage situation of the hydrogen storage tank. In formula (39): is the maximum state of charge of the hydrogen storage tank, is the rated energy of the hydrogen storage tank.

[0103] Fuel cell operation model:

[0104]

[0105] In formula (40): is the output power of the fuel cell, is the input hydrogen consumption of the fuel cell, is the hydrogen-to-electricity conversion coefficient of the fuel cell. In formula (41): The maximum output power of the fuel cell, which depends on the installed capacity of the fuel cell and the situation of hydrogen storage tanks, is the minimum hydrogen storage capacity of the hydrogen storage tank. In formula (43): It is the minimum state of charge of the hydrogen storage tank.

[0106] Hydrogen storage tank model running model:

[0107]

[0108]

[0109] Formula (44) is the relationship when the hydrogen storage tank stores hydrogen, is the hydrogen storage capacity of the hydrogen storage tank at time t-1; is the output hydrogen amount of the electrolyzer device at the time t-1; Δt is the time interval; is the hydrogen consumption at time t-1; Formula (45) is the relationship when the hydrogen storage tank releases hydrogen, It is the efficiency of hydrogen filling and discharging of the hydrogen storage tank. is the maximum hydrogen storage capacity of the hydrogen storage tank. In formula (47): and are the initial and final state of charge of the hydrogen storage tank during the annual scheduling cycle. s0 is the initial scenario of the hydrogen storage tank operation; s N This is the final scenario of the operation of the hydrogen storage tank. In order to ensure that the hydrogen energy storage system can work continuously and effectively, the charge state of the hydrogen storage tank should be equal at the beginning and end of the scheduling cycle.

[0110] The carbon volume module consists of carbon trading costs and carbon capture and storage (CCS) devices.

[0111] The carbon trading cost is explained by formula (19). Carbon trading is essentially a trading mechanism to achieve carbon emission reduction through the purchase and sale of carbon emission quotas.

[0112] The process of CCS device capturing carbon dioxide emitted by thermal power units is shown as follows, where set g includes coal-fired and gas-fired thermal power units.

[0113]

[0114]

[0115] In formula (48): is the energy consumption of thermal power units, is the carbon emission per unit of energy consumption, is the carbon emissions generated by thermal power units. In formula (49): is the capture efficiency of the CCS device, is the actual capture capacity of the CCS device, is the amount of dissipation during the capture process of the CCS device. In equations (50) and (51): is the actual power consumption of the CCS device, is the installed capacity of the CCS device, is the power consumption coefficient of the CCS device.

[0116] The driving factor constraints consist of clean energy electricity share constraints, carbon emission limit constraints and power balance constraints.

[0117] Constraints on the proportion of clean energy electricity:

[0118]

[0119] In formula (52), is the electrical load in the area, is the minimum clean energy electricity proportion parameter.

[0120] Carbon emission limit constraints: There are two main paths for carbon emission limit constraints. The first is steady restrictions year by year, and the other is total amount restrictions.

[0121]

[0122] X y ≤X base σ y (54)

[0123]

[0124] In formula (53) to formula (55): X y is the equivalent carbon emissions of the region in year y, X base is the basic carbon emission per year, σ y is the carbon emission limit parameter for year y, X base,a is the total basic carbon emissions during the planning period, σ a It is the total carbon emission limit parameter during the planning period.

[0125] Power balance constraint: In any scenario and at any time, the sum of the power output should be equal to the total power load.

[0126]

[0127] Finally, the objective function is redefined according to the weights of the basic scenarios and is re-expressed as:

[0128]

[0129] Formula (57): τ s is the weight coefficient of each operating scenario, and are the operating cost and carbon trading cost under the sth scenario, respectively, s It is the sum of the operating cost and carbon trading cost in the sth scenario.

[0130] Step S6: Based on the capacity evolution sequence generation model and the power system timing production simulation model containing hydrogen long-term energy storage, a convolutional neural network-bidirectional gated recurrent unit-attention mechanism is used to construct a data-model hybrid driven model capable of measuring the cleanliness and low-carbon nature of the system; the data-model hybrid driven model takes the capacity evolution sequence output by the capacity evolution sequence generation model as input, and takes the corresponding unit electricity share marginal cost and unit carbon limit marginal cost as output.

[0131] S6 specifically includes:

[0132] Step S61: applying a random function to generate a capacity evolution random sequence based on the capacity evolution sequence generation model.

[0133] Step S62: Input the capacity evolution random sequence into the power system time-series production simulation model containing hydrogen long-term energy storage to obtain the minimized operating cost and carbon trading cost.

[0134] Step S63: According to the minimized operating cost and carbon trading cost, by adjusting the clean energy electricity proportion or carbon emission limit, the corresponding unit electricity proportion marginal cost and unit carbon limit marginal cost are calculated.

[0135] Step S64: Taking the capacity evolution random sequence as input and the corresponding marginal cost per unit electricity share and the marginal cost per unit carbon limit as output, the convolutional neural network-bidirectional gated recurrent unit-attention mechanism is trained to obtain a data-model hybrid driven model capable of measuring the cleanliness and low-carbon nature of the system.

[0136] In practical applications, first of all, the proportion of clean energy electricity and carbon emission limits are used as specific indicators to measure the cleanliness and low-carbon nature of the power system considering long-term hydrogen energy storage. Based on the expected clean energy electricity proportion path and the expected annual carbon emission limit path, the relationship between economic costs and the path is studied by appropriately increasing the proportion of clean energy electricity and reducing the carbon emission limit.

[0137]

[0138] In equations (58) and (59), ΔF2 is the change in operating cost and carbon trading cost, is the change in the proportion of clean energy electricity, Δσ y is the change in carbon emission limit, is the marginal cost per unit of electricity, is the marginal cost per unit of carbon limit.

[0139] Secondly, based on the capacity evolution sequence generation model and the power system time-series production simulation model with hydrogen long-term energy storage, a convolutional neural network-bidirectional gated recurrent unit-attention mechanism (CNN-BiGRU-AM) is used to build a data-model hybrid driven model capable of measuring the cleanliness and low-carbon nature of the system. The CNN-BiGRU-AM model helps to extract the hidden spatiotemporal features in the capacity evolution sequence data. The attention mechanism guides the model to prioritize the relevant input data features, enhancing the robustness of the model.

[0140] The specific process is as follows: the first step is to use random functions to continuously generate capacity evolution random sequences based on the capacity evolution sequence generation model; the second step is to input the capacity evolution random sequence into the power system time series production simulation model containing hydrogen long-term energy storage to obtain the minimized operating cost and carbon trading cost, and further appropriately increase the proportion of clean energy electricity or reduce the carbon emission limit to calculate the marginal cost per unit electricity share and the marginal cost per unit carbon limit; the third step is to build a massive data set, input the capacity evolution random sequence, the marginal cost per unit electricity share and the marginal cost per unit carbon limit into the CNN-BiGRU-AM neural network model, and continuously train. The fourth step is that after the training is completed, after inputting the capacity evolution sequence into the CNN-BiGRU-AM model, the corresponding marginal cost per unit electricity share and marginal cost per unit carbon limit can be quickly obtained.

[0141] Step S7: Based on the marginal cost per unit electricity share and the marginal cost per unit carbon limit, the sum of the minimized investment cost and maintenance cost, and the sum of the minimized operating cost and carbon trading cost, a low-carbon evolution path simulation model of the power system taking into account hydrogen long-term energy storage is constructed, involving economic cost driving factors, clean energy electricity share driving factors, and carbon emission limit driving factors.

[0142] In practical applications, the corresponding driving model is constructed by taking into account the economic cost driving factors, the clean energy electricity proportion driving factors and the carbon emission limit driving factors.

[0143] Constructing an economic cost-driven model:

[0144]

[0145] v + (F s )=-(F0-F s -F1),F0>F s +F1 (61)

[0146] v - (F s )=ζ(F s+F1-F0),F s +F1≥F0 (62)

[0147] In formula (60) to formula (62), is the economic cost probability coefficient; F s is the sum of the operating cost and carbon trading cost in the sth scenario; in step S3, the minimized investment cost and maintenance cost are output; in step S6, the minimized operating cost and carbon trading cost are output based on the expected clean energy electricity proportion path and the expected annual carbon emission limit path; F0 is the sum of the above minimized investment cost and maintenance cost, and minimized operating cost and carbon trading cost. ζ is the driving coefficient.

[0148] Construct a clean energy electricity share driven model and a carbon emission limit driven model.

[0149]

[0150] v + (γ y )=-(γ y -γ y,0 ),γ y >γ y,0 (64)

[0151] v - (γ y )=ζ(γ y,0 -γ y ),γ y,0 ≥γ y (65)

[0152] In formula (64) to formula (65): is the probability coefficient of clean energy electricity proportion; γ y is the proportion of clean energy electricity in year y; γ y,0 It is the reference parameter for the proportion of clean energy electricity in year y.

[0153]

[0154] v + (σ y )=-(σ y,0 -σ y ),σ y,0 >σ y (67)

[0155] v - (σ y )=ζ(σ y -σ y,0 ),σ y ≥σ y,0 (68)

[0156] In formula (67) to formula (68): is the carbon emission limit probability coefficient; σ y is the carbon emission limit for year y; y,0 It is the reference parameter of carbon emission limit in year y.

[0157] Secondly, the objective function of the simulation model of the low-carbon evolution path of the power system considering long-term hydrogen energy storage is expressed as:

[0158] V=ω F V F +ω new V new +ω co V co (69)

[0159] In formula (69), V is the comprehensive benefit cost value, V F 、V new and V co are the benefit cost values ​​driven by economic cost, clean energy power ratio and carbon emission limit respectively, ω F ,ω new and ω co It is the weight coefficient driven by economic cost, clean energy electricity proportion and carbon emission limit.

[0160] Step S8: Based on the simulation model of the low-carbon evolution path of the power system, with the goal of minimizing the comprehensive benefit cost, a two-layer solution framework including a capacity layer and an operation layer is proposed. With the goal of minimizing the comprehensive benefit cost, the capacity layer adopts a PID search algorithm, and the operation layer adopts an improved differential evolution algorithm to obtain a low-carbon evolution path of the power system considering long-term hydrogen energy storage.

[0161] In practical applications, the low-carbon evolution path simulation model of the power system considering hydrogen long-term energy storage relaxes the clean energy electricity share constraint and the carbon emission limit constraint, that is, through the unit electricity share marginal cost and the unit carbon limit marginal cost, the clean energy electricity share constraint and the carbon emission limit constraint are converted into the clean energy electricity share driven model and the carbon emission limit driven model, and the constraint conditions are converted to the objective function part, realizing the relaxation of the model. Therefore, the constraints of the model do not include the two, and the other constraints are consistent with the constraints of the capacity evolution sequence generation model and the power system timing production simulation model containing hydrogen long-term energy storage.

[0162] For the low-carbon evolution path simulation model of the power system considering long-term hydrogen energy storage, the hierarchical optimization idea is adopted, and a two-layer solution framework is proposed. The hierarchical solution is carried out with the goal of minimizing the comprehensive benefit cost. The model can be divided into two parts: the capacity layer and the operation layer. The capacity layer adopts the PID search algorithm, and the operation layer adopts the improved differential evolution algorithm to solve.

[0163] At the capacity layer, a PID search algorithm is used to solve the problem. The PID search algorithm is based on the incremental PID algorithm, which continuously adjusts the system deviation to make the entire population converge to the optimal state.

[0164] e k (t) = x * (t-1)-x(t-1) (70)

[0165] Δu(t)=K p r2[e k (t)-e k-1 (t)]+K i r3e k (t)+K d r4[e k (t)-2e k-1 (t)+e k-2 (t)] (71)

[0166] x(t+1)=x(t)+ξΔu(t)+(1-ξ)o(t) (72)

[0167] In formula (70) to formula (72): e k (t) is the population deviation at the tth iteration, x(t) is the population state at the tth iteration, and x * (t) is the best state at the iteration number t, which is the state of the overall historical minimum; x * (t-1) is the optimal state at iteration t-1; x(t-1) is the population state at iteration t-1, x(t+1) is the population state at iteration t+1, e k-1 (t) is the total deviation of the previous iteration when the number of iterations is t, e k-2 (t) is the total deviation of the first two iterations when the number of iterations is t, r2, r3 and r4 are random numbers, K p , K i and K d are the proportional, integral and differential coefficients respectively; Δu(t) is the output value of PID regulation, o(t) is the zero output condition factor, and ξ is a random number.

[0168] At the operation layer, an improved differential evolution algorithm is used to solve the problem. The entire population is randomly divided into two sub-populations, and different mutation strategies are used to speed up the algorithm optimization and improve the algorithm optimization performance.

[0169]

[0170]

[0171] In formula (73) to formula (76): x i,G is the i-th individual in the population, G is the population algebra, v i,G is the population individual after mutation, and are three different random individuals in the population, x pbest,G is to randomly select one from the first p individuals with outstanding performance in the current population, F is the scaling factor; CR is the crossover probability, rand(j) is a random number, v i,j,G is the jth element of the population individual after mutation, x i,j,G is the jth element of the i-th population individual; u i,G is the population individual obtained after crossover, f(u i,G ) and f(x i,G ) are u i,G and x i,G The fitness of i,j,G is the jth element of the population individual obtained after crossover; v i,j,G is the jth element of the population individual after mutation; x i,G+1 is the i-th individual in the population of the G+1th generation.

[0172] When running the simulation, firstly, the initial parameters are set according to the capacity configuration results transmitted from the capacity layer, the population is initialized, the individual fitness values ​​are calculated, and the optimal value is recorded. Then, the entire population is randomly divided into two sub-populations, and different mutation strategies are adopted, namely, equations (73) to (74), and corresponding mutations are performed on each sub-population. According to equation (75), the crossover operation is performed. According to equation (76), the test individual is compared with the target individual, and the optimal value is recorded to determine whether the convergence accuracy is met or the number of iterations is reached. If the termination condition is not met, the population is randomly divided, the mutation operation and the crossover operation are performed again. When it stops, the output x i,G+1 , that is, the optimal results of the operation simulation part under the current energy device capacity configuration can be obtained.

[0173] Based on the above two algorithms, firstly, on the basis of the initial installed capacity of each type of energy device, the PID search algorithm is used to initialize the population of the installed capacity of the online energy device; the next step is to calculate the system error. At this time, the goal of calculating the system error is to minimize the comprehensive benefit cost. The investment cost and maintenance cost can be calculated according to the installed capacity of the online device. The operation layer needs to be called to perform time-series production simulation, and the optimal operating cost, carbon trading cost, clean energy electricity ratio and carbon emission limit of the operation layer are returned; finally, the PID search algorithm continuously adjusts the system capacity evolution sequence to make the entire population converge to the optimal state, and outputs the evolution plan of the clean energy electricity ratio path and the carbon emission limit path.

[0174] This application has the following innovative features:

[0175] (1) The investment cost evolution model disclosed in this application, which takes into account the uncertainty of technological innovation breakthrough rate, can predict the unit capacity investment cost of various types of typical equipment.

[0176] (2) This application uses a data-model hybrid driven method based on a convolutional neural network-bidirectional gated recurrent unit-attention mechanism (CNN-BiGRU-AM) to achieve rapid measurement of system cleanliness and low carbon performance.

[0177] (3) This application determines a simulation model of the low-carbon evolution path of the power system taking into account hydrogen long-term energy storage based on the capacity evolution sequence generation model and the power system time-series production simulation model containing hydrogen long-term energy storage, and relaxes the core constraints.

[0178] (4) This application uses a PID search algorithm and an improved differential evolution algorithm to solve a simulation model of the low-carbon evolution path of the power system.

[0179] This application has the following advantages:

[0180] (1) This application constructs an energy device investment cost evolution model that takes into account the uncertainty of technological innovation breakthrough rate. The model has an excellent fitting effect on the historical data of investment costs and can improve the prediction accuracy of the investment cost per unit capacity of energy devices.

[0181] (2) This application constructs a data-model hybrid driven model based on the CNN-BiGRU-AM neural network model that has the ability to measure the cleanliness and low-carbon performance of the system, and can quickly obtain the equivalent benefits of cleanliness and low-carbon performance under different capacity evolution sequences.

[0182] (3) This application constructs a capacity evolution sequence generation model and a power system time-series production simulation model containing hydrogen long-term energy storage, accurately models the electric energy, hydrogen energy and carbon content, and effectively simplifies the operating scenarios, providing an effective means for the capacity evolution of energy devices in the power system and enhancing the solvability of the model. At the same time, considering the role of hydrogen long-term energy storage, it can effectively reduce the economic cost of energy device configuration; further constructs a low-carbon evolution path simulation model for the power system considering hydrogen long-term energy storage, providing a model basis for the evolution path optimization simulation.

[0183] (4) This application adopts a two-layer solution framework of PID search algorithm and improved differential evolution algorithm to solve the low-carbon evolution path simulation model of the power system considering hydrogen long-term energy storage, which effectively improves the global search capability and convergence speed of the algorithm, and has the characteristics of high search efficiency and fast solution speed.

[0184] (5) This application provides an effective means for the evolution of the power system power structure and the arrangement of power capacity construction, taking into account economy, low carbon and cleanliness, and effectively improving the comprehensive benefits of the power system. The evolutionary path formed takes into account economy, low carbon and cleanliness, provides directional guidance for the power system power construction, effectively reduces the capacity configuration cost and improves the utilization rate of new energy and the degree of clean and low carbonization of the power system.

[0185] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store simulation data of the low-carbon evolution path of the power system. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for simulating the low-carbon evolution path of the power system is implemented.

[0186] Those skilled in the art will understand that Figure 4The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0187] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0188] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0189] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0190] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0191] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0192] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0193] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for simulating a low-carbon evolution path of a power system, characterized in that: The power system low-carbon evolution path simulation method comprises: Obtain relevant historical data on energy costs in the target area, future planning data, the upper limit of the construction scale of equipment corresponding to each type of energy, the current scale of constructed equipment and the planned retirement time; the future planning data includes the installed capacity, future planned capacity and future planned R&D investment of each type of energy; the relevant historical data includes the installed capacity, R&D investment and unit capacity investment cost of each type of energy in previous years; Based on the relevant historical data and the future planning data, construct an investment cost evolution model that takes into account the uncertainty of technological innovation breakthrough rate; According to the unit capacity investment cost evolution prediction data output by the investment cost evolution model, the upper limit of the constructible scale of the equipment corresponding to each type of energy, and the scale of the currently constructed equipment and the planned retirement time, a capacity evolution sequence generation model is constructed with minimizing the sum of investment cost and maintenance cost as the objective function and with equipment capacity connection constraints and capacity limit constraints as constraints; Based on the relevant historical data, a k-means clustering algorithm is used to generate various operation scenarios of the power system; According to the given capacity evolution sequence output by the capacity evolution sequence generation model and the various operation scenarios of the power system, a power system time-series production simulation model containing hydrogen long-term energy storage is constructed with carbon emission limit constraint and clean energy electricity proportion constraint as core constraints and minimizing the sum of operation cost and carbon trading cost as the objective function; According to the capacity evolution sequence generation model and the power system time-series production simulation model containing hydrogen long-term energy storage, a data-model hybrid driving model capable of measuring system cleanliness and low carbon is constructed by using a convolutional neural network-bidirectional gated recurrent unit-attention mechanism; the data-model hybrid driving model takes the capacity evolution sequence output by the capacity evolution sequence generation model as input, and takes the corresponding unit electricity share marginal cost and unit carbon limit marginal cost as output; According to the marginal cost per unit electricity share and the marginal cost per unit carbon limit, the sum of the minimized investment cost and maintenance cost, and the sum of the minimized operating cost and carbon trading cost, a low-carbon evolution path simulation model of the power system considering hydrogen long-term energy storage involving economic cost driving factors, clean energy electricity share driving factors, and carbon emission limit driving factors is constructed; According to the simulation model of the low-carbon evolution path of the power system, with the goal of minimizing the comprehensive benefit cost, a two-layer solution framework including the capacity layer and the operation layer is proposed. With the goal of minimizing the comprehensive benefit cost, the capacity layer adopts the PID search algorithm, and the operation layer adopts the improved differential evolution algorithm, so as to obtain the low-carbon evolution path of the power system considering hydrogen long-term energy storage.

2. The method for simulating the low-carbon evolution path of a power system according to claim 1, characterized in that: According to the relevant historical data and the future planning data, based on the two-factor learning curve model and taking into account the uncertainty of the technological innovation breakthrough rate, an investment cost evolution model that takes into account the uncertainty of the technological innovation breakthrough rate is constructed.

3. The method for simulating the low-carbon evolution path of a power system according to claim 1, characterized in that: The investment cost evolution model is: a y =α0A y B y ; Among them, α y is the unit capacity investment cost in the yth year; α0 is the initial unit capacity investment cost; A y B is the degree of decrease in unit capacity investment cost caused by the growth of installed capacity in year y; y N is the degree of decrease in unit capacity investment cost caused by the growth of R&D investment in year y; y is the cumulative installed capacity scale in the yth year; N0 is the initial installed capacity scale; a is the learning rate exponent of the cumulative installed capacity scale; u y It is a binary variable, which takes 1 when there is a breakthrough in technological innovation, and takes 0 when there is no breakthrough in technological innovation; C y is the cumulative R&D investment in the yth year; C0 is the initial R&D investment; b is the learning rate index of the cumulative R&D investment; B y-1 It is the degree of decrease in unit capacity investment cost caused by the increase in R&D investment in year y-1.

4. The method for simulating the low-carbon evolution path of a power system according to claim 1, characterized in that: The objective function of the capacity evolution sequence generation model is: minF1=min(F inv +F main ); Among them, F1 is the cost of the capacity evolution sequence generation model; F inv is the investment cost; F main For maintenance cost; kn For the newly added device collection; Ω y For planning year collection; is the unit capacity investment cost of each type of energy device; P kn,y is the installed capacity of the new devices added in year y; E is the unit energy investment cost of each type of energy device; kn,y F is the energy of the newly added device in year y; sa The residual value of the equipment that has not been retired within the planning period; Ω k is the collection of online energy devices; ρ k is the residual value recovery factor; is the unit capacity investment cost at the end of the planning period; P k,Y The installed capacity in the final year of the planning period; E is the unit energy investment cost at the end of the planning period; k,Y The energy for the last year of the planning period; H k,Y To plan the remaining life of various types of energy devices at the end of the project; k is the service life of each type of energy device; Y0 is the time when the energy device begins to be put into use; The unit capacity maintenance cost of each type of energy device; is the unit energy maintenance cost of each type of energy device; Y is the last year of the planning period.

5. The method for simulating the low-carbon evolution path of a power system according to claim 1, characterized in that: The objective function of the power system time-series production simulation model containing hydrogen long-term energy storage is: minF2=min(F om +F co ); Among them, F2 is the cost of the time series production simulation model; F om is the operating cost; F co is the carbon trading cost; k is the collection of online energy devices; Ω y For planning year collection;Ω s To run the scene collection; Ω t It is a collection of daily running times; is the unit coal consumption cost; is the coal consumption; is the unit natural gas consumption cost; is the natural gas consumption; is the unit hydrogen consumption cost; is the hydrogen consumption; Δt is the time interval; is the startup cost of the thermal power unit; For downtime costs; is the carbon trading price in year y; CO y is the total carbon trading volume in year y.

6. The method for simulating the low-carbon evolution path of a power system according to claim 1, characterized in that: According to the capacity evolution sequence generation model and the power system time series production simulation model containing hydrogen long-term energy storage, a data-model hybrid driving model capable of measuring system cleanliness and low carbon is constructed by using a convolutional neural network-bidirectional gated recurrent unit-attention mechanism, which specifically includes: Applying a random function to generate a capacity evolution random sequence based on the capacity evolution sequence generation model; Inputting the capacity evolution random sequence into the power system time-series production simulation model containing hydrogen long-term energy storage to obtain the minimized operating cost and carbon trading cost; According to the minimized operating costs and carbon trading costs, by adjusting the clean energy electricity share or carbon emission limit, the corresponding marginal cost per unit electricity share and the marginal cost per unit carbon limit are calculated; With the capacity evolution random sequence as input and the corresponding marginal cost per unit electricity share and the marginal cost per unit carbon limit as output, the convolutional neural network-bidirectional gated recurrent unit-attention mechanism is trained to obtain a data-model hybrid driven model capable of measuring the cleanliness and low-carbon nature of the system.

7. The method for simulating the low-carbon evolution path of a power system according to claim 1, characterized in that: The simulation model of the low-carbon evolution path of the power system considering long-term hydrogen energy storage is: V=ω F V F +ω new V new +ω co V co ; Among them, V is the comprehensive benefit cost value; ω F is the weight coefficient driven by economic cost; V F is the benefit cost value driven by economic cost; new is the weight coefficient driven by the proportion of clean energy electricity; V new is the benefit cost value driven by the proportion of clean energy electricity; ω co is the weight coefficient driven by carbon emission limit; V co is the benefit-cost value driven by carbon emission limits.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for simulating the low-carbon evolution path of a power system according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for simulating the low-carbon evolution path of a power system according to any one of claims 1 to 7 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for simulating the low-carbon evolution path of a power system according to any one of claims 1 to 7 is implemented.

Images