A method, device, equipment and medium for wind-solar power consumption type water-fire power coordination scheduling of a high water power region power grid in dry season

Abstract

The present application relates to the field of power collaborative optimization scheduling, in particular to a kind of wind and light consumption type water and fire electricity collaborative scheduling method, device, equipment and medium of high water power area power grid in dry season;By obtaining load, new energy prediction, water and fire turbine parameters, standby demand, carbon quota and double market price as basic input;With predicted upper bound limiting new energy consumption, with unit parameter limiting output interval, construct the objective function with new energy consumption income and carbon quota surplus as positive term, with water power start-stop loss, fire power peak shaving compensation, fire power operation, standby procurement and carbon compliance cost as negative term;Establish the constraint set containing power balance, output limit, climbing, standby and carbon emission, set hierarchical check and dynamic penalty mechanism;Adopt hierarchical constraint self-adaptive correction particle swarm optimization algorithm to solve, obtain global optimal particle, analyze water turbine start-stop instruction, fire power and standby plan as scheduling scheme output, improve the comprehensive benefit of high water power grid operation in dry season.

Description

Technical Field

[0001] This invention relates to the field of power coordinated optimization dispatch, specifically to a wind-solar integrated hydropower coordinated dispatch method, device, equipment, and medium for power grids in high-hydropower areas during the dry season. Background Technology

[0002] In my country's southwest and other high-hydropower regions, natural runoff decreases significantly during the dry season, leading to a simultaneous decline in both hydropower peak-shaving and reserve capacity. Meanwhile, the scale of wind and solar power grid connection continues to grow, exacerbating intraday power fluctuations and highlighting the pressure on grid peak-shaving and reserve capacity. Under these conditions, it is necessary to rely on the deep peak-shaving and reserve support capabilities of existing thermal power units, while coordinating with frequent start-ups and shutdowns of hydropower units to avoid peak wind and solar power generation periods, in order to ensure full absorption of new energy and the safe and stable operation of the power grid.

[0003] Existing dispatching methods typically establish economic dispatching models based on thermal power, treating hydropower output as a fixed plan or allocated according to runoff proportion. Although some studies have introduced carbon emission constraints or reserve capacity constraints, they are often treated as independent boundaries, making it difficult to coordinate and unify the wind and solar power consumption targets during the dry season with the economic efficiency of grid operation and carbon emission reduction constraints, resulting in low overall efficiency of dispatching schemes. Summary of the Invention

[0004] The purpose of this invention is to provide a method, device, equipment, and medium for wind and solar power co-dispatch of hydropower and thermal power in power grids in high-hydropower areas during the dry season, which solves the problems in the prior art.

[0005] This invention is achieved through the following technical solution:

[0006] In a first aspect, embodiments of the present invention provide a wind-solar integrated hydropower-thermal power coordinated dispatching method for power grids in high-hydropower areas during the dry season, comprising:

[0007] The system load data, renewable energy output forecast data, hydropower and thermal power unit operating parameter data, system reserve demand data, carbon emission quota and carbon market price data, and electricity reserve market price data within the scheduling cycle are obtained as basic input parameters.

[0008] Based on the new energy output forecast data, the upper limit of the actual consumption of new energy is defined. Based on the operating parameters of the hydropower units, the output range of the hydropower units is defined. The new energy consumption revenue and carbon quota surplus revenue are positive terms, and the start-up and shutdown losses of hydropower units, peak-shaving compensation of thermal power, thermal power operation, standby procurement and carbon compliance costs are negative terms, and a comprehensive revenue objective function is constructed.

[0009] Based on the aforementioned basic input parameters, a set of coupled constraint conditions is established, including power balance constraints, unit output limit constraints, ramping constraints, reserve capacity constraints, and carbon emission constraints. A constraint processing mechanism with hierarchical verification and dynamic penalty based on safety priority is also set.

[0010] A hierarchical constraint adaptive correction particle swarm optimization algorithm is adopted to calculate the globally optimal particle based on the comprehensive benefit objective function and the constraint processing mechanism.

[0011] The start-stop status commands of hydropower units, the planned active power output of thermal power units, and the planned reserve capacity of thermal power units in each time period are extracted from the position vector of the globally optimal particle and used as the output of the collaborative optimization scheduling scheme.

[0012] Preferably, the operating parameter data of the hydropower and thermal power units includes the adjustable output upper and lower limits of each hydropower unit, the stable combustion lower limit and maximum output of each thermal power unit, and the ramp-up rate of each thermal power unit.

[0013] The new energy output forecast data includes upper limit data for wind power output forecast and upper limit data for photovoltaic output forecast;

[0014] The carbon emission allowance and carbon market price data include grid time-of-use free carbon emission allowance data, carbon market time-of-use transaction price data, and carbon excess penalty unit price data.

[0015] The electricity standby market price data refers to the time-of-use unit capacity electricity price data in the electricity standby market.

[0016] Preferably, the upper limit of actual renewable energy consumption is defined based on the renewable energy output forecast data, including:

[0017] The upper limit of the predicted wind power output is used as the upper limit of the actual active power absorbed by wind power, and the upper limit of the predicted photovoltaic output is used as the upper limit of the actual active power absorbed by photovoltaic power, thereby limiting the maximum grid-connected power of wind power and photovoltaic power in each time period.

[0018] Preferably, the revenue from renewable energy consumption is obtained by multiplying the sum of the average on-grid electricity price of wind and solar power, the revenue from green certificate transactions, and the unit price of the renewable energy excess consumption policy incentive by the actual active power consumed by wind and solar power.

[0019] The carbon allowance surplus revenue is obtained by multiplying the carbon market time-of-use transaction price by the time-of-use carbon allowance surplus amount, which is obtained by subtracting the actual carbon emissions of thermal power units from the grid time-of-use free carbon emission allowance.

[0020] The start-up and shutdown losses of the hydropower units are obtained by multiplying the unit price of the single start-up and shutdown loss by the absolute value of the change in the start-up and shutdown status of the unit, and then summing them over all hydropower units and over the entire time period.

[0021] The thermal power peak shaving compensation is obtained by multiplying the deep peak shaving unit compensation standard with the difference between the unit output and the peak shaving benchmark value, and then summing the results for all thermal power units and all time periods.

[0022] The standby procurement cost is obtained by multiplying the time-of-use unit capacity electricity price in the power standby market with the total standby capacity procured by the power grid and then summing them up over the entire time period.

[0023] The carbon compliance cost is calculated by multiplying the carbon market time-of-use transaction price by the carbon allowance gap, plus the carbon overrun penalty price by the excess amount of the carbon allowance gap, and then summing these products over the entire time period. The carbon allowance gap is obtained by subtracting the grid's time-of-use free carbon emission allowance from the actual carbon emissions of thermal power units.

[0024] Preferably, the reserve capacity constraint is obtained by subtracting the current active power output from the maximum output of the thermal power unit to obtain the maximum reserve capacity that a single thermal power unit can provide.

[0025] A rigid reserve capacity constraint for the power grid during the dry season is established, with each thermal power unit's actual reserved reserve capacity not exceeding its own maximum reserve capacity and the sum of all thermal power units' actual reserved reserve capacity not being lower than the system's minimum reserve requirement.

[0026] Preferably, the carbon emission constraint is obtained by multiplying the carbon emission intensity of the thermal power unit by the active power output of the thermal power unit to obtain the actual carbon emission of the thermal power unit.

[0027] A hard constraint is established on the total amount of time-of-use carbon emissions, based on the premise that the actual carbon emissions of the thermal power units do not exceed the time-of-use carbon emission control limit.

[0028] The time-of-use carbon quota gap is defined by the difference between the actual carbon emissions of the thermal power units and the time-of-use free carbon emission quota of the power grid, and a carbon quota compliance balance constraint is established.

[0029] Preferably, the step of employing a hierarchical constraint adaptive correction particle swarm optimization algorithm, which calculates the globally optimal particle based on the comprehensive benefit objective function and the constraint processing mechanism, includes:

[0030] Based on the comprehensive benefit objective function and the set of coupling constraints, a multidimensional particle population is generated. Each particle contains variables of the start-up and shutdown status of hydropower units, variables of the active power output of thermal power units, and variables of the reserve of thermal power units for each time period.

[0031] According to the power grid security priority from high to low, each constraint in the set of coupled constraints is checked in a step-by-step manner, and the dynamic penalty value is added to the particle fitness for particles that violate the constraints.

[0032] Particle velocity and position are updated using linearly decreasing inertial weights;

[0033] For the updated particles, the discrete variables of the start-stop state of hydropower units, the continuous variables of the active power output of thermal power units, and the continuous variables of the reserve capacity of thermal power units are over-limited. The correction is carried out by taking the start-stop state as close to 0 or 1 as the nearest value and the output value as the minimum stable combustion output or maximum output of the unit as the correction benchmark.

[0034] When the global optimal solution does not improve for several consecutive rounds, select some particles for perturbation and mutation with an adaptively adjusted mutation probability;

[0035] Repeat the iteration until the convergence condition is met to obtain the globally optimal particle.

[0036] Secondly, embodiments of the present invention provide a wind-solar integrated hydropower-thermal power coordinated dispatching device for power grids in high-hydropower areas during the dry season, comprising:

[0037] The acquisition module is used to acquire system load data, new energy output forecast data, hydropower and thermal power unit operating parameter data, system reserve demand data, carbon emission quota and carbon market price data, and electricity reserve market price data within the scheduling cycle, as basic input parameters.

[0038] The objective function module is used to limit the upper limit of actual new energy consumption based on the new energy output prediction data, limit the output range of hydropower units based on the operating parameters of hydropower units, and construct a comprehensive revenue objective function with new energy consumption revenue and carbon quota surplus revenue as positive terms, and hydropower unit start-up and shutdown losses, thermal power peak-shaving compensation, thermal power operation, standby procurement and carbon compliance costs as negative terms.

[0039] The constraint module is used to establish a set of coupled constraint conditions, including power balance constraints, unit output limit constraints, ramping constraints, reserve capacity constraints and carbon emission constraints, based on the basic input parameters, and to set a constraint processing mechanism of hierarchical verification and dynamic penalty according to safety priority.

[0040] The calculation module is used to calculate the globally optimal particle using a hierarchical constraint adaptive correction particle swarm algorithm based on the comprehensive benefit objective function and the constraint processing mechanism.

[0041] The parsing module is used to parse the start-stop status commands of hydropower units, the planned active power output of thermal power units, and the planned reserve capacity of thermal power units in each time period from the position vector of the globally optimal particle, and output them as a collaborative optimization scheduling scheme.

[0042] Thirdly, embodiments of the present invention provide an electronic device, including: at least one processor, at least one memory, and computer program instructions stored in the memory, which, when executed by the processor, implement the method of the first aspect described above.

[0043] Fourthly, embodiments of the present invention provide a storage medium storing computer program instructions, which, when executed by a processor, implement the method of the first aspect described above.

[0044] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0045] By acquiring multi-source basic data covering load, new energy forecasts, parameters of hydropower and thermal power units, reserve demand, carbon quotas, and dual-market prices, a comprehensive revenue objective function is constructed. This function has positive terms in wind and solar power consumption revenue and carbon quota surplus revenue, and negative terms in hydropower start-up and shutdown losses, thermal power peak-shaving compensation, thermal power operation, reserve procurement, and carbon compliance costs. A coupled constraint set, including power balance, unit output limits, ramp-up, reserve capacity, and carbon emissions, along with a hierarchical verification and dynamic penalty mechanism, is established. A hierarchical constraint adaptive correction particle swarm optimization algorithm is used for mixed-integer optimization. Finally, hydropower unit start-up and shutdown commands, thermal power output plans, and reserve requirements are extracted from the globally optimal particles. The reserved plan can simultaneously meet all boundary conditions—power balance, thermal power ramp-up safety, rigid reserve capacity not lower than the system's minimum demand, and total carbon emission control—under the condition of insufficient hydropower peak-shaving and reserve capacity during the dry season. It directly incorporates time-of-use electricity prices from the power reserve market and time-of-use carbon market prices into the objective function to quantify market-based procurement and carbon compliance costs on a time-by-time basis. Furthermore, it uses hydropower start-up and shutdown losses and thermal power peak-shaving compensation to simultaneously counterbalance the increased unit losses and peak-shaving expenditures brought about by renewable energy consumption. This ensures that the optimal solution converges within the feasible region where reserve safety boundaries, carbon emission caps, and various cost expenditures intersect, outputting a day-ahead hydro-thermal coordinated dispatch scheme. This improves the overall efficiency of the dispatch scheme. Attached Figure Description

[0046] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0047] Figure 1 A flowchart illustrating the wind-solar integrated hydropower-thermal power coordinated dispatch method for power grids in high-hydropower areas during the dry season, provided by this invention.

[0048] Figure 2 A schematic diagram of the wind-solar integrated hydropower and thermal power coordinated dispatching device for power grids in high-hydropower areas during dry seasons provided by the present invention;

[0049] Figure 3 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0051] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0052] It should be noted that all actions involving the acquisition of signals, information, or data in this invention are carried out in compliance with the relevant data protection laws and regulations of the locality and with authorization from the owner of the relevant device.

[0053] Example 1

[0054] Please see Figure 1 This invention provides a wind-solar integrated hydropower-thermal power coordinated dispatching method for power grids in high-hydropower areas during the dry season, comprising:

[0055] S1. Obtain system load data, new energy output forecast data, hydropower and thermal power unit operating parameter data, system reserve demand data, carbon emission quota and carbon market price data, and electricity reserve market price data within the scheduling cycle as basic input parameters;

[0056] Among them, the total dispatch period of the day The value is 24, with 1 hour as the smallest time granularity, and a total of 24 scheduling periods are included.

[0057] System load data refers to the total active power demand of the power grid in each time period. .

[0058] The renewable energy output forecast data includes the predicted maximum generating capacity of wind power for each time period. And the predicted maximum power output of photovoltaics in each time period. .

[0059] Operating parameter data for hydroelectric power units includes the total number of grid-connected hydroelectric power units. Lower limit of adjustable output of each unit in the cascade hydropower project and adjustable output limit Total number of grid-connected thermal power units Lower limit output for stable combustion of each thermal power unit Maximum output value of each thermal power unit Uphill ramp capacity of each thermal power unit and downhill capacity All output parameters mentioned above are in MW, and the climbing rate is in MW / h.

[0060] System reserve requirement data refers to the total spinning reserve capacity that the power grid is required to reserve for all operating thermal power units during the dry season to ensure frequency stability and power supply reliability. .

[0061] Carbon emission allowances and carbon market price data include the administratively allocated free carbon emission allowances to the power grid dispatch center in each time period. The transaction price of carbon emission rights in the national carbon emission trading market at different times. and the price of carbon excess units The unit for carbon emission allowances is tons. The carbon price is in yuan / ton. .

[0062] Electricity standby market price data refers to the unit capacity procurement price in the electricity ancillary services standby market for each time period. .

[0063] Specifically, a day-ahead planning model with a 24-hour scheduling cycle is adopted. The ultra-short-term load forecast results for the next day at 96 points are read from the power grid dispatch automation system, and the average value is taken every 15 minutes and merged into 24 whole hours to obtain the system load for each time period. The power output forecast results for each wind farm and photovoltaic power station for the following day were read from the new energy power forecast system and then merged into 24 hourly intervals to obtain the upper limit of wind power forecast output for each time period. And the upper limit of photovoltaic power generation forecast Extract the nameplate parameters and test data of all hydropower and thermal power units participating in grid connection and scheduling from the unit management system to obtain the number of hydropower units. Lower limit of output of each hydropower unit With upper limit of output Number of thermal power units Lower limit of stable combustion for each thermal power unit Maximum output of each thermal power unit and the uphill capacity of each thermal power unit Downhill capacity The minimum reserve requirements of the system during each period of the dry season are read from the power grid safety verification system. Obtain free carbon emission allowances from the power grid for each time period from the carbon emission management platform. The transaction prices of carbon allowances for each period were retrieved from publicly available data from the national carbon emission trading market. and the unit price of excess penalty Obtain the unit electricity price of reserve capacity for each time period of the following day from the power ancillary services reserve market trading platform. All the parameters mentioned above constitute the complete input boundary of the optimized scheduling model, providing data support for the subsequent construction of the objective function and the establishment of constraints.

[0064] In some embodiments, the operating parameter data of the hydropower and thermal power units include the adjustable output upper and lower limits of each hydropower unit, the stable combustion lower limit and maximum output of each thermal power unit, and the ramp-up rate of each thermal power unit.

[0065] The new energy output forecast data includes upper limit data for wind power output forecast and upper limit data for photovoltaic output forecast;

[0066] The carbon emission allowance and carbon market price data include grid time-of-use free carbon emission allowance data, carbon market time-of-use transaction price data, and carbon excess penalty unit price data.

[0067] The electricity standby market price data refers to the time-of-use unit capacity electricity price data in the electricity standby market.

[0068] Among them, the adjustable output upper and lower limits of each unit in the cascade hydropower system refer to the minimum and maximum active power that each turbine generator unit participating in the scheduling and grid connection in the cascade hydropower station group of the basin can stably operate under given head and flow conditions. The adjustable output lower limit of the Taiwan hydroelectric power unit is The value is expressed in MW, and its magnitude is determined by the boundary of the unit's vibration zone and cavitation characteristics. Operating below this lower limit will cause the unit to enter an unstable vibration zone, leading to increased mechanical wear and a sharp decline in efficiency. The adjustable output upper limit is... The value is expressed in MW, and its value is determined by the rated capacity of the unit and the flow capacity of the turbine. Exceeding this upper limit will result in generator overload or turbine over-rated output.

[0069] During the dry season, when the water flow is low, the maximum output of some generating units may be reduced due to the decrease in water head. The dispatching model needs to limit the output range according to the actual adjustable range of each generating unit.

[0070] The data on the lower limit of stable combustion and maximum output of each thermal power unit refer to the minimum active power that each thermal power unit can maintain stable combustion under the condition that the boiler does not inject oil for combustion assistance, and the maximum active power that the turbine generator unit can continuously and safely operate. The lower limit of stable combustion of Taiwan's thermal power units is This indicates that the unit is MW, and this value is determined by the boiler's minimum stable combustion load rate. When the output of the thermal power unit is below this lower limit, the furnace temperature is insufficient to maintain the self-sustaining combustion of pulverized coal, requiring the use of fuel oil for combustion assistance, which sharply increases operating costs and poses a risk of fire suppression and boiler shutdown; the maximum output is... This means that the value is determined by the minimum of the turbine's rated capacity, the boiler's maximum evaporation capacity, and the generator's rated power. Exceeding this upper limit will trigger the equipment's overload protection.

[0071] The ramp rate data for each thermal power unit refers to the upper limit of the allowable increase or decrease in active power per unit time for each thermal power unit. Upward ramp capacity is expressed as... This indicates that the downhill climbing capacity is... The figures are all in MW / h. This data is determined by a combination of turbine thermal stress control, boiler combustion response speed, and auxiliary equipment following capability. Too low a ramp rate will cause the unit to be unable to respond quickly to drastic fluctuations in load and new energy sources, while too high a ramp rate will accelerate rotor thermal fatigue and boiler tube wall damage.

[0072] The upper limit of wind power output forecast refers to the upper limit of the maximum active power that a wind farm can generate, predicted for each time period within the dispatch cycle, based on numerical weather forecasts and historical wind farm operation data. The value is expressed in MW. This data is generated by the power grid's new energy power prediction system and represents the theoretical upper limit of the actual dispatchable power of wind power.

[0073] The upper limit of photovoltaic output forecast refers to the upper limit of the maximum active power that a photovoltaic power station can generate in each time period within the dispatch cycle, based on solar radiation forecasts and the power generation characteristics of photovoltaic arrays. The value is expressed in MW. This data, also generated by the power grid's new energy power prediction system, represents the theoretical upper limit of the actual dispatchable power of photovoltaic power.

[0074] Time-of-use (TOU) free carbon emission allowances for the power grid refer to the total amount of carbon dioxide that the government's carbon reduction authorities allocate to the power grid free of charge during each time period within the dispatch cycle. It indicates that the unit is tons. The quota is determined based on the power supply tasks undertaken by the power grid and its benchmark carbon emission intensity. If it is still necessary to increase thermal power output after the carbon quota for a period is used up, additional quotas must be purchased from the national carbon emission trading market or penalties for exceeding the quota must be borne.

[0075] Carbon market time-of-day transaction price data refers to the real-time transaction price of carbon emission rights in the national carbon emission rights trading market during each period of the scheduling cycle. The unit is yuan / ton. The price fluctuates in real time with market supply and demand. During the dry season, when thermal power output is high, the demand for carbon allowances increases, carbon prices rise, and the carbon compliance costs of thermal power generation are directly increased.

[0076] Carbon excess penalty unit price data refers to the administrative penalty fee levied on the excess portion based on the unit carbon emission when the actual carbon emissions of the power grid exceed the administratively exempted allowance at any given time. The unit is yuan / ton. The penalty price is set by government carbon emission reduction regulations and is usually significantly higher than the carbon market transaction price. It is used to enforce that the total carbon emissions during a given period must not exceed the control limit.

[0077] The time-of-use unit capacity price data in the power reserve market refers to the unit price at which spinning reserve capacity is purchased in each time period within the dispatch cycle in the power ancillary services reserve market. The price is expressed in yuan / MW. This price is determined by centralized bidding in the reserve market. The larger the reserve gap in the power grid, the higher the price per unit capacity, which directly determines the economic cost that the power grid needs to pay when purchasing reserve capacity from thermal power units.

[0078] S2. Based on the new energy output forecast data, limit the upper limit of the actual consumption of new energy, and based on the operating parameters of the hydropower units, limit the output range of the hydropower units. With the new energy consumption revenue and carbon quota surplus revenue as positive terms, and the start-up and shutdown losses of hydropower units, peak-shaving compensation of thermal power, thermal power operation, standby procurement and carbon compliance costs as negative terms, construct a comprehensive revenue objective function.

[0079] Among them, the actual active power absorbed by wind power Indicates the time period The actual grid-connected wind power capacity received by the internal power grid.

[0080] Actual active power absorbed by photovoltaic power Indicates the time period The actual grid-connected photovoltaic power received by the internal power grid.

[0081] Renewable energy consumption revenue refers to the comprehensive economic revenue generated per unit of wind and solar power generation, and its conversion weight. Average feed-in tariff of wind and solar power Green certificate unit transaction income and the unit price of the policy incentive for excess consumption of new energy The sum of the three is obtained, that is .

[0082] Carbon allowance surplus revenue refers to the economic value generated when the actual carbon emissions of the power grid in each period are lower than the administratively granted allowances, and the surplus allowances can be sold in the carbon market. Its weighting is as follows: Take the average price of intraday transactions in the national carbon market .

[0083] Start-up and shutdown losses of hydropower units refer to the total mechanical wear, no-load water wastage, and additional maintenance costs incurred during the start-up and shutdown of hydropower units, and their weighting. The average cost of a single start-up and shutdown of all hydropower units in the basin is taken.

[0084] Thermal power peak-shaving compensation refers to the special economic compensation paid by the power grid to thermal power units for deeply reducing their output and adopting new energy sources, and its conversion weighting The compensation standards for deep peak shaving units are taken from the regional ancillary services market's public announcement.

[0085] Comprehensive operating cost of thermal power The cost of coal consumption per unit of generating unit, cost of ramp-up loss per unit, and average carbon allowance procurement cost converted to unit power generation are integrated.

[0086] Weighting of standby procurement costs The value is the day-ahead time-of-use average electricity price in the electricity standby market.

[0087] Weighting of carbon compliance penalty costs The value is the fixed penalty unit price for excess carbon emissions in the carbon market.

[0088] Specifically, the sum of the revenue from renewable energy consumption and the revenue from carbon quota surplus is taken as the positive term, and the start-up and shutdown losses of hydropower units are taken as the negative term. Cost of deep peak shaving compensation for thermal power Comprehensive operating cost of thermal power Spinning standby capacity procurement costs in the standby market and the total cost of carbon trading compliance and penalties The sum of these terms is negative. Therefore, we construct the overall return objective function, which is to maximize:

[0089] .

[0090] in, For time period The total actual carbon emissions of all thermal power units in the network. This objective function integrates wind and solar power consumption revenue, carbon emission reduction incentives, hydropower start-up and shutdown costs, thermal power peak-shaving costs, thermal power operating costs, reserve procurement costs, and carbon compliance costs into a unified quantitative framework. This allows dispatch decisions to automatically weigh the impact of various cost expenditures on the overall revenue while pursuing maximum renewable energy consumption, avoiding the problem of unilaterally pursuing consumption while neglecting operational economics.

[0091] In some implementations, the upper limit of actual renewable energy consumption is defined based on the renewable energy output forecast data, including:

[0092] The upper limit of the predicted wind power output is used as the upper limit of the actual active power absorbed by wind power, and the upper limit of the predicted photovoltaic output is used as the upper limit of the actual active power absorbed by photovoltaic power, thereby limiting the maximum grid-connected power of wind power and photovoltaic power in each time period.

[0093] Among them, the actual active power absorbed by wind power refers to the actual wind power absorbed and integrated into the grid during each period of the dispatch cycle. The value is expressed in MW. This variable is one of the decision variables in the optimization model, representing the actual grid-connected power output of wind power in the scheduling scheme. Its value is determined by the optimization algorithm by finding the optimal value between zero and the upper limit of the predicted wind power output. This variable directly participates in the calculation of the wind and solar power consumption revenue term in the system active power balance constraint and the comprehensive revenue objective function.

[0094] The actual active power absorbed by the photovoltaic grid refers to the actual photovoltaic grid-connected power absorbed by the grid in each time period during the dispatch cycle. The value is expressed in MW. This variable is also a decision variable in the optimization model. Its value is determined by the optimization algorithm between zero and the upper limit of the photovoltaic output prediction data, and it directly participates in the calculation of the system active power balance constraint and the comprehensive benefit objective function.

[0095] In this step, the actual active power absorbed by wind power is... The upper bound is taken as the upper limit of wind power output prediction. Actual active power absorbed by photovoltaic power The upper bound is taken as the upper limit of photovoltaic power output prediction. The necessity of setting an upper limit lies in the fact that the output of new energy sources is constrained by natural conditions, and there is a physical limit to the actual power generation. The amount of new energy consumption arranged in the dispatch plan cannot exceed this limit value; otherwise, the plan cannot be implemented.

[0096] Maximum grid-connected power refers to the upper limit of active power that renewable energy power plants are allowed to inject into the grid during each time period. For wind power... For photovoltaics This upper limit serves as a boundary constraint for the compilation of dispatch instructions, and the renewable energy grid connection plan for all time periods must not exceed this value. When the grid's peak-shaving capacity is insufficient or the reserve capacity is tight, the optimization algorithm can select partial consumption between zero and the maximum grid-connected power. At this time, the unutilized wind and solar power is the amount of wind and solar curtailment, which is included in the renewable energy consumption rate assessment indicator.

[0097] Specifically, the upper limit of wind power output forecast. and the upper limit of photovoltaic output forecast As fixed inputs to the constraints, they are explicitly written into the variable domain of the optimization model. and Two sets of inequality constraints. When the optimization algorithm provides... and When determining the value of , first check whether the two sets of inequalities hold true. Greater than Then the scheduling scheme is physically unfeasible because the wind farm cannot generate more power than predicted at the current wind speed; if Exceeding Similarly, it lacks physical feasibility. Two types of inequality constraints forcibly limit the actual absorption of new energy within a closed interval from zero to the predicted upper limit. The lower limit of zero indicates that the dispatch scheme allows for full wind or solar curtailment, while the upper limit is the predicted value, representing complete wind and solar absorption. When this constraint works in conjunction with system reserve capacity constraints and thermal power output limits, if thermal power has no room for further reduction and hydropower has been completely shut down, the absorption of new energy is automatically limited to a value below the predicted upper limit. This value represents the maximum absorbable wind and solar power under the current peak-shaving capacity of hydropower and thermal power units. The amount of wind and solar curtailment is determined by... and The calculation is as follows. After introducing this upper bound, the wind-solar energy consumption benefit term in the objective function... We will not blindly pursue wind and solar power outputs that exceed the limits allowed by natural conditions, thus ensuring that the optimization results match the actual renewable energy generation capacity.

[0098] In some implementations, the revenue from renewable energy consumption is obtained by multiplying the sum of the average on-grid electricity price of wind and solar power, the revenue from green certificate transactions, and the unit price of the renewable energy excess consumption policy incentive by the actual active power consumed by wind and solar power.

[0099] The carbon allowance surplus revenue is obtained by multiplying the carbon market time-of-use transaction price by the time-of-use carbon allowance surplus amount, which is obtained by subtracting the actual carbon emissions of thermal power units from the grid time-of-use free carbon emission allowance.

[0100] The start-up and shutdown losses of the hydropower units are obtained by multiplying the unit price of the single start-up and shutdown loss by the absolute value of the change in the start-up and shutdown status of the unit, and then summing them over all hydropower units and over the entire time period.

[0101] The thermal power peak shaving compensation is obtained by multiplying the deep peak shaving unit compensation standard with the difference between the unit output and the peak shaving benchmark value, and then summing the results for all thermal power units and all time periods.

[0102] The standby procurement cost is obtained by multiplying the time-of-use unit capacity electricity price in the power standby market with the total standby capacity procured by the power grid and then summing them up over the entire time period.

[0103] The carbon compliance cost is calculated by multiplying the carbon market time-of-use transaction price by the carbon allowance gap, plus the carbon overrun penalty price by the excess amount of the carbon allowance gap, and then summing these products over the entire time period. The carbon allowance gap is obtained by subtracting the grid's time-of-use free carbon emission allowance from the actual carbon emissions of thermal power units.

[0104] Among them, the revenue from renewable energy consumption refers to the comprehensive economic return generated by the power grid's actual consumption of wind and solar active power during the dispatch cycle. This revenue consists of three parts: revenue from wind and solar grid connection fees, revenue from green certificate trading, and revenue from excess grid connection rewards, collectively reflecting the comprehensive economic value of each unit of wind and solar power generation to the grid. This revenue appears as a positive term in the objective function, and its value monotonically increases with the actual amount of new energy absorbed, making it the core incentive term driving the optimization algorithm to improve the level of wind and solar grid connection.

[0105] The average on-grid tariff for wind and solar power refers to the revenue generated per unit of electricity sold after wind and solar power are connected to the grid, in accordance with the on-grid tariff policy. This price is the weighted average of the feed-in tariffs for wind and solar power, with the weights determined by the respective projected generation volume. The feed-in tariffs for wind and solar power are typically higher than the benchmark tariff for conventional thermal power, reflecting policy support for clean energy generation.

[0106] Green certificate unit trading revenue refers to the average income that can be obtained by selling green electricity certificates corresponding to each unit of wind and solar power generated in a voluntary green certificate trading market. This means that each megawatt-hour of wind and solar power generation can generate one green certificate. The grid can sell these green certificates to electricity sales companies and industrial and commercial users who have renewable energy consumption responsibility weight assessment obligations. This revenue is independent of the grid connection fee and represents the added economic value of new energy power generation.

[0107] The incentive price per unit of renewable energy excess consumption policy refers to the administrative incentive compensation given per unit of electricity for the portion of wind and solar power actually consumed by the power grid that exceeds the consumption responsibility target set by the provincial energy authority. The unit is yuan / MWh. This incentive policy is designed to encourage the power grid to minimize wind and solar curtailment during periods of high renewable energy generation. Any amount exceeding the consumption target will receive additional economic compensation, further strengthening the positive incentive of the objective function for high consumption levels.

[0108] The actual active power absorbed by wind power is It indicates that the actual active power absorbed by photovoltaic power is... This indicates that both are optimization models for the time period. The decision variables, in step S2, have been limited to a range from 0 to their respective upper prediction limits. The sum of the two... Representative time period The total active power of renewable energy absorbed by the power grid is directly proportional to the revenue from renewable energy absorption.

[0109] Carbon allowance surplus revenue refers to the economic value of surplus allowances when the actual carbon emissions of the power grid at any given time are lower than the administratively waived carbon emission allowances. Indicates. When When the carbon allowance surplus is positive, it indicates that the power grid's carbon emissions are lower than the administratively allocated quota during that period, and the surplus allowance can be sold for profit in the national carbon emissions trading market; when When this term is negative, it actually represents an increase in subsequent carbon compliance costs. This benefit term is treated as a positive term in the objective function, incentivizing the optimization algorithm to prioritize power generation from thermal power units with lower carbon emission intensity, proactively reducing total carbon emissions throughout the entire period, while ensuring power supply demand is met.

[0110] Time-of-use carbon allowance surplus refers to the carbon allowance surplus during a given period. Internal power grid administrative free carbon emission quota Subtract the actual carbon emissions of all thermal power units in the network The difference is expressed as:

[0111] ;

[0112] A positive difference indicates a surplus in the carbon allowance, while a negative difference indicates a deficit. This surplus serves as the basis for calculating the carbon allowance surplus revenue item, and its magnitude directly determines the sign and absolute value of the carbon allowance surplus revenue item.

[0113] Actual carbon emissions from thermal power units refer to emissions during a specific time period. The total amount of carbon dioxide emitted by all operating thermal power units due to coal-fired power generation is expressed as:

[0114] ;

[0115] in For the first Carbon emission intensity of Taiwan's thermal power units, in tons / MWh is determined by the unit's coal consumption characteristics and the carbon content of the coal.

[0116] The start-up and shutdown losses of hydropower units refer to the total variable dynamic costs incurred by all hydropower units during the scheduling cycle due to start-up and shutdown operations, expressed as:

[0117] ,

[0118] in For the first The unit price of losses during a single start-up and shutdown of a hydroelectric generating unit. During startup, the turbine runner accelerates from rest to rated speed, subjecting the thrust bearing and guide bearing to alternating stress impacts. During shutdown, the water flow is cut off instantaneously, generating water hammer pressure pulsations that cause fatigue damage to the volute and draft tube. Furthermore, during start-up and shutdown, to maintain unobstructed flow, the guide vanes need to be opened for no-load operation. This water flow through the turbine is discharged directly without generating electricity, constituting no-load water wastage losses. Unit price of losses during a single start-up and shutdown. This is the average of the sum of the aforementioned mechanical wear costs, additional maintenance costs, and water wastage losses due to no-load operation. (Absolute value term) Used to accurately identify changes in the unit's status between adjacent time periods: when and When it is started, and The time indicates a shutdown. Both starting and stopping incur losses. Absolute value calculations treat state changes in both directions as a single loss event.

[0119] The unit price of single start-up and shutdown loss refers to the total cost of comprehensive mechanical wear, additional maintenance, and no-load water wastage losses incurred by a hydropower unit during a single start-up or shutdown operation. The unit price varies depending on the unit capacity, model, and head conditions. This model uses the average cost of a single start-up and shutdown of all hydropower units in the basin as a uniform weight. This unit price is multiplied by the absolute value of the state change and then accumulated for each unit and each time period to obtain the total start-up and shutdown loss of the hydropower units throughout the entire cycle.

[0120] The absolute value of the change in the start-up and shutdown status of the unit refers to the first The absolute value of the difference between the start-stop 0-1 state values ​​of the hydroelectric generating units between adjacent time periods, in order to Indicates that when the unit status changes from shutdown to start-up (…). , ) or from start to stop ( , When the unit's state remains unchanged, the absolute value is 1, indicating that a valid start-stop action has been included in the loss statistics; when the unit's state remains unchanged, the absolute value is 0, and no start-stop loss is generated. Introducing the absolute value to replace the square operation ensures that the start-stop loss is linearly proportional to the frequency of state changes and does not generate nonlinear amplification due to directional differences.

[0121] Thermal power peak-shaving compensation refers to the total special economic compensation paid by the power grid to thermal power units under deep peak-shaving conditions (output below the peak-shaving benchmark value) due to increased coal consumption, equipment wear and tear, and power generation loss, expressed as:

[0122] ,

[0123] in For the first The unit compensation standard for deep peak shaving of thermal power units in Taiwan is expressed in yuan / MW. The deep peak-shaving benchmark output of the unit is measured in MW. Typically, 50% to 60% of the unit's rated output is taken as the starting point for peak-shaving compensation. For time period The unit has actual active power output; For this unit's time period The 0-1 power-on state. When When the generating unit enters the deep peak-shaving range, the power grid needs to pay the power plant compensation for the difference in electricity volume compared to the benchmark value; when At this time, the unit is in its normal peak-shaving range and does not generate additional compensation. This compensation cost is a negative term in the objective function, reflecting the economic cost of the deep reduction in thermal power output caused by the adoption of new energy sources.

[0124] The unit compensation standard for deep peak shaving refers to the compensation unit price paid by the power auxiliary service market to thermal power units for the unit reduction in power output during the deep peak shaving period. This standard is formulated by the ancillary service markets in each region based on the unit capacity level and peak-shaving depth, and is published and implemented in the market rules. Larger capacity thermal power units have poorer boiler combustion stability when operating at low loads, and the corresponding unit compensation standard is usually higher.

[0125] The difference between the unit output and the peak-shaving benchmark value refers to the difference when the actual active power output of the thermal power unit is lower than the peak-shaving benchmark value. Output below deep peak shaving benchmark At that time, the difference between the two The difference is multiplied by the power-on status. Compensation is only calculated when the unit is connected to the grid; it is not included in peak-shaving compensation calculations when the unit is out of service. The larger this difference is, the more power generation capacity the unit has given up for renewable energy, and the higher the corresponding compensation expenditure.

[0126] Reserve procurement cost refers to the total cost paid by the power grid to purchase spinning reserve capacity from the power ancillary services reserve market during the dispatch cycle, expressed as:

[0127] ,

[0128] in For time period The unit capacity electricity price in the standby market is determined by competitive bidding between supply and demand in the standby market; For time period The total amount of spinning reserve capacity purchased by the power grid from the reserve market is equal to the sum of the actual reserved reserve capacity of all thermal power units in the entire grid. During the dry season, hydropower reserve capacity is insufficient, and all reserve capacity must be purchased from thermal power units through the reserve market. This cost increases with the increase in reserve demand and the rise in reserve market electricity prices.

[0129] Total power grid procurement reserve capacity refers to the total amount of spinning reserve capacity procured by the power grid from the reserve market during each time period of the dispatch cycle. This means that, in the optimization model, this variable is equal to the sum of the reserved reserve capacity of all thermal power units in the entire network. This amount is allocated and reserved by each operating thermal power unit. This figure is subject to the system's minimum reserve requirement. The constraint is that the value must not be lower than the required value.

[0130] Carbon compliance costs refer to all carbon-related economic expenditures borne by the power grid to meet carbon emission control requirements, including the cost of purchasing carbon allowances through the market and the administrative penalties for exceeding carbon allowances, expressed as:

[0131] ,

[0132] in The time-sharing price in the carbon market. The unit price for carbon excess penalties This represents a carbon allowance shortfall. When actual carbon emissions fall below the free allowance, The first term is negative. (Negative values) generate negative compliance costs, which are actually reflected as allowance surplus revenue, consistent with the carbon allowance surplus revenue term in the objective function; when actual carbon emissions exceed the free allowances, For a positive value, the first term represents the cost of purchasing the shortage quota at market price, and the second term... This represents administrative penalties for exceeding emission limits. This cost item incorporates both market-based carbon price signals and administrative carbon control limits into the dispatch target, forcing the optimization algorithm to proactively weigh carbon emission costs in thermal power output scheduling.

[0133] Carbon quota gap refers to the period Actual carbon emissions of internal thermal power units Exceeding the grid's free carbon emission allowance The part is represented as:

[0134] ,

[0135] when When carbon allowances are scarce, the power grid needs to purchase allowances from the carbon market or accept penalties; When there is a carbon allowance surplus, the surplus allowances can be sold for profit. This gap value serves as the calculation benchmark for both the carbon allowance surplus revenue item and the carbon compliance cost item, achieving a unified measurement of carbon-related income and expenditure.

[0136] The excess value of the carbon quota gap refers to the carbon quota gap. The positive part in the expression is represented as:

[0137] ,

[0138] This positive operation ensures the unit price of carbon excess penalty. The penalty is levied only on the positive shortfall from carbon emission allowances; there is no negative penalty for a negative shortfall (i.e., a surplus). This excess amount is subject to the administrative carbon emission quota's hard constraint. This linkage means that when actual carbon emissions approach or reach the control limit, this value increases sharply, and the cost of punishment rises rapidly.

[0139] Specifically, the calculation results of the above six sub-items are ultimately combined into all non-thermal power operating cost items in the objective function, along with the comprehensive operating cost of thermal power. Together, they constitute the objective function. In each iteration of the HC-APSO algorithm, the particle fitness calculation module sequentially calls the above six sub-functions, summarizing the results of each sub-function by sign into the overall fitness. Used for updating and After iterative convergence, the globally optimal particle... The corresponding six sub-items form a complete economic accounting list, which can serve as the basis for a comprehensive economic evaluation of the scheduling plan.

[0140] S3. Based on the aforementioned basic input parameters, establish a set of coupled constraint conditions including power balance constraints, unit output limit constraints, ramping constraints, reserve capacity constraints, and carbon emission constraints, and set a constraint processing mechanism with hierarchical verification and dynamic penalty according to safety priority.

[0141] Among them, the power balance constraint requires that the sum of the active power output of all power sources in the power grid be equal to the system load and the power transmitted across regions in real time, and its mathematical expression is: ,in For the total output of cascade hydropower, For the total output of thermal power, This refers to power transmitted across regions.

[0142] The power output limit constraint requires the active power output of each hydropower unit. In start / stop state When it is 1, it is between and During the shutdown state, the output is zero, which is expressed as ,in Active power output of each thermal power unit In start / stop state When it is 1, it is between and During the shutdown state, the output is zero, which is expressed as ,in .

[0143] The ramping constraint requires that the change in power output of a thermal power unit between adjacent time periods must not exceed the unit's ramping capacity, expressed as: .

[0144] Reserve capacity constraints require that all reserve capacity be reserved in the operating thermal power units. The sum shall not be less than the system's minimum standby requirement. Furthermore, the reserve capacity of a single generating unit must not exceed the difference between its maximum output and current output, expressed as... ,as well as .

[0145] Carbon emission constraints require actual carbon emissions from thermal power units Not exceeding the time-of-use carbon emission control limit ,in For the first The carbon emission intensity of Taiwan's thermal power units is defined by the difference between actual carbon emissions and free allowances for a given period. This provides a unified benchmark for calculating carbon compliance costs.

[0146] A constraint handling mechanism based on safety priority and dynamic penalties is established. The constraints are divided into five verification levels according to the power grid safety level, from highest to lowest: the first level verifies the start-up and shutdown matching logic of hydropower units to prevent active power output during shutdown; the second level verifies the system's active power balance; the third level verifies the upper and lower limits of thermal power unit output and ramp-up rates; the fourth level verifies the system's rigid reserve capacity constraints during the dry season; and the fifth level verifies the time-of-use carbon emission total and carbon quota compliance balance constraints. Based on this hierarchical sequence, constraint violations are determined layer by layer for each particle's decision variable combination in the particle swarm optimization algorithm. For particles that violate constraints at each level, a penalty term is added to the fitness value according to a preset dynamic penalty coefficient gradient. The initial penalty coefficient is 10, and it increases linearly by a factor of 1.5 every 10 iterations. This allows the algorithm to tolerate boundary violations moderately in the early stages of iteration to maintain population diversity, and tightens the constraint boundaries in the later stages of iteration to converge to a feasible solution. This mechanism, which combines hierarchical verification with dynamic penalties, explicitly embeds grid security priorities into the optimization process, ensuring that hard constraints such as high-security power balance and reserve capacity are given priority in the optimization process, and preventing the algorithm from sacrificing grid operation safety in pursuit of absorption benefits.

[0147] Furthermore, the set of coupling constraints includes:

[0148] 1. System active power balance constraints

[0149] ,

[0150] In the formula: For the total output of cascade hydropower; Total output of thermal power; Total wind power output; Contribute to the overall photovoltaic power generation; This represents the system load. This refers to power transmitted across regions.

[0151] 2. Start-up and shutdown logic constraints for hydropower units

[0152] , ,

[0153] In the formula: The power output of the kth hydropower station during time period t; These are the lower and upper limits of the output of the hydropower unit.

[0154] 3. Start-up and shutdown 0-1 state constraints of thermal power units + upper and lower output limit constraints

[0155] , ,

[0156] In the formula: This represents the lower limit of stable combustion and the maximum output of a single thermal power plant.

[0157] 4. Unit ramp-up rate constraints

[0158] ,

[0159] In the formula: This represents the unit's maximum downhill ramp capacity; This represents the unit's maximum uphill climbing capacity.

[0160] 5. Upper limit constraints on wind and solar power output

[0161] , ,

[0162] In the formula: This represents the upper limit of the predicted wind power output. This represents the upper limit of photovoltaic power output forecast.

[0163] 6. Rigid reserve capacity constraints of the power grid during the dry season

[0164] ,

[0165] In the formula: Reserved capacity for the i-th thermal power plant; This represents the minimum backup requirement for the system.

[0166] 7. Hard constraints on total carbon emissions at different times of day

[0167] ,

[0168] In the formula: The carbon emission intensity of the i-th thermal power plant; This is the upper limit for time-of-use carbon emission control.

[0169] 8. Supply and demand balance constraints in the power reserve market

[0170] ,

[0171] In the formula: This represents the total reserve capacity of thermal power plants across the entire grid.

[0172] 9. Carbon quota compliance balance constraints

[0173] .

[0174] In some implementations, the reserve capacity constraint is obtained by subtracting the current active power output from the maximum output of the thermal power unit to obtain the maximum reserve capacity that a single thermal power unit can provide.

[0175] A rigid reserve capacity constraint for the power grid during the dry season is established, with each thermal power unit's actual reserved reserve capacity not exceeding its own maximum reserve capacity and the sum of all thermal power units' actual reserved reserve capacity not being lower than the system's minimum reserve requirement.

[0176] The reserve capacity constraint refers to the inequality restriction imposed on the spinning reserve capacity of all operating thermal power units to ensure the grid can quickly restore power balance in the event of unplanned unit outages or sudden drops in wind and solar power output. This constraint includes two sub-conditions: first, the actual reserved reserve capacity of a single thermal power unit must not exceed the maximum reserve capacity that the unit can provide at its current output level; second, the sum of the actual reserved reserve capacities of all operating thermal power units must not be less than the system's minimum reserve requirement. This constraint is rigid and insurmountable during the dry season because hydropower reserve capacity is almost non-existent during this period, and the entire spinning reserve of the grid must be borne by thermal power units. If the reserved reserve capacity is insufficient, the grid will face the risk of frequency instability in the event of a power disturbance.

[0177] The maximum standby capacity that a single thermal power unit can provide refers to the... Taiwan thermal power units during the period Currently contributing efforts Based on this, it can increase the maximum amount of output in a short period of time, in order to This maximum reserve capacity is indicated by the unit's maximum output. Subtract current active effort The mathematical expression is: The lower the current output of a generating unit, the greater the difference between its maximum and current output, and the more reserve capacity it can provide. Conversely, when a unit is nearing full capacity, the available reserve capacity approaches zero. This difference constitutes the physical upper bound of the reserve capacity of a single generating unit, representing the actual reserve capacity reserved for that unit in any scheduling plan. None of them may exceed this upper limit.

[0178] The maximum output index of thermal power units The maximum active power that a thermal power unit can continuously and safely operate under rated operating conditions, in order to This value is determined by the minimum of the turbine's rated capacity, the boiler's maximum evaporation capacity, and the generator's rated power, and is a fixed input value in the operating parameter data of thermal power units.

[0179] Current merit contribution index Taiwan thermal power units during the period The actual power generation determined by the optimization model, in order to This indicates that the variable is one of the core decision variables in the optimization model, and its value is within the lower limit of stable combustion of thermal power units. With maximum output The value is determined by an optimization algorithm. The lower the value, the more the unit reduces its output during the deep peak-shaving range to absorb renewable energy, and the more upward adjustment space it has, enabling it to provide more spinning reserve capacity.

[0180] The actual reserved reserve capacity of thermal power units refers to the capacity allocated to the first unit in the dispatching plan. Taiwan thermal power units during the period The amount of rotating reserve tasks undertaken, in order to This indicates that the variable is one of the core decision variables in the optimization model, and is related to the active power output of the thermal power unit. and the unit's maximum output Together, they form the inequality relationship of the backup constraint. This variable satisfies... This means that the actual reserved reserve capacity ranges from zero to the maximum reserve capacity that a single unit can provide. By introducing the reserve capacity into the model as a constraint rather than a fixed value, the optimization algorithm can flexibly adjust the reserve reserve of each unit based on the reserve market electricity price signal while meeting the minimum reserve requirements of the system, thus avoiding cost redundancy caused by excessive reservation.

[0181] The minimum reserve requirement of the system refers to the lower limit of the total spinning reserve capacity that the power grid is required to reserve for all operating generating units during the dry season to withstand random power disturbances. This demand value is determined by the power grid operation department based on a comprehensive calculation of the next day's load forecast level, wind and solar power forecast accuracy, maximum single-unit capacity, and historical disturbance statistics. During the dry season when hydropower reserve capacity is insufficient, this demand is entirely borne by thermal power units, constituting a rigid and insurmountable safety baseline for the power grid. This data originates from system reserve demand data.

[0182] The rigid reserve capacity constraint of the power grid during the dry season refers to the mandatory restriction that, under the special condition that the adjustable reserve capacity of hydropower is insufficient during the dry season, the sum of the actual reserved reserve capacity of all operating thermal power units shall not be less than the minimum reserve requirement of the system. The mathematical expression is:

[0183] ,

[0184] The rigidity of this constraint is reflected in two aspects: First, hydropower lacks the capacity to provide backup power during the dry season and cannot replace thermal power in fulfilling backup tasks; second, insufficient backup capacity will directly lead to the grid losing its safety margin in dealing with power disturbances, and there is no way to exempt this constraint through economic compensation or market procurement. This constraint is related to the constraint on the maximum standby capacity of a single unit. Together, they constitute a complete set of standby capacity constraints.

[0185] Specifically, in the S3 step of constructing the set of coupling constraints, the spare capacity constraint is written into the optimization model in the form of two sets of inequalities. The first set of inequalities... Imposing restrictions on each operating thermal power unit individually: the right side of the equation The variable represents the unit's startup status (0-1). When the unit is shut down... The right end is zero, at this time Forced shutdown of the unit means the unit cannot provide backup capacity; when the unit is restarted... The right side shows the maximum reserve capacity that the unit can currently provide. This inequality directly couples the output level of thermal power with reserve capacity: when wind and solar power generation is high, the output of thermal power is suppressed. As the power output of wind and solar power increases, the reserve capacity provided by a single unit also increases; when the output of wind and solar power is insufficient, thermal power plants need to increase their output, narrowing the gap and correspondingly reducing the reserve capacity provided by a single unit. This coupling relationship accurately describes the mutual constraint between peak shaving and reserve capacity of thermal power plants—the lower the power output, the more wind and solar power can be accommodated, but at the same time, more reserves can be provided. The fact that both directions are aligned actually alleviates the crowding-out contradiction.

[0186] The second set of inequalities A lower limit constraint is imposed on the total reserve capacity of all operating units: the actual reserve capacity reserved by each unit. The sum must be greater than or equal to the system's minimum standby requirement. When the total reserve capacity of each unit in the particle scheme is less than When this occurs, the particle will be deemed to have violated the backup safety constraint and will be captured by the fourth layer of verification in the hierarchical verification mechanism set in step S3. A dynamic penalty value will be added to its fitness, guiding the algorithm to search for feasible regions that meet the backup requirements. When the two layers of inequalities work together, the maximum backup constraint for a single unit limits the upper limit that each unit can share, while the total backup constraint requires that the total backup capacity of all units meet the system requirements. Together, they ensure that the thermal power backup capacity is neither excessively concentrated in a single unit (limited by the upper limit of a single unit) nor lower than the minimum safety standard of the entire network (limited by the lower limit of the total capacity).

[0187] This reserve capacity constraint and the reserve procurement cost item in the objective function Forming a two-way linkage: the reserve procurement cost item follows (equal The increase in ) drives the algorithm to satisfy The goal is to minimize the total amount of reserve capacity while ensuring grid security. The lower bound of the reserve capacity constraint prevents the algorithm from reducing reserve capacity to an unsafe level in an effort to save on procurement costs. This two-way balance mechanism ensures that the optimal solution falls precisely near the reserve demand constraint boundary, guaranteeing grid security while avoiding cost waste caused by excessive reserve procurement.

[0188] In some embodiments, the carbon emission constraint is obtained by multiplying the carbon emission intensity of the thermal power unit by the active power output of the thermal power unit to obtain the actual carbon emission of the thermal power unit.

[0189] A hard constraint is established on the total amount of time-of-use carbon emissions, based on the premise that the actual carbon emissions of the thermal power units do not exceed the time-of-use carbon emission control limit.

[0190] The time-of-use carbon quota gap is defined by the difference between the actual carbon emissions of the thermal power units and the time-of-use free carbon emission quota of the power grid, and a carbon quota compliance balance constraint is established.

[0191] Among them, the carbon emission intensity of thermal power units refers to the first The carbon dioxide emissions corresponding to one megawatt-hour of active power generated by Taiwan's thermal power units, in order to It indicates that the unit is tons. / MWh. This intensity is determined by the unit's boiler efficiency, the carbon content of the coal, and the degree of combustion completeness. Significant differences exist between units of different capacity levels and years of commissioning: ultra-supercritical large units have lower carbon emission intensity, while subcritical and lower-level small units have higher carbon emission intensity. The carbon emission intensity of each unit is comprehensively calibrated based on unit performance tests and online monitoring data, and used as a fixed input parameter in the calculation of actual carbon emissions.

[0192] The active power output index of thermal power units in the optimization model Taiwan thermal power units during the period The determined actual power generation capacity, in order to express.

[0193] Actual carbon emissions from thermal power units refer to the time period The total amount of carbon dioxide directly emitted by all operating thermal power units due to coal-fired power generation is expressed as:

[0194] ,

[0195] This calculation formula will determine the carbon emission intensity of each unit. Rather than contributing effort Multiplying these values ​​and then summing them up one by one reflects a linear proportional relationship between thermal power output and total carbon emissions—the higher the total thermal power output, the greater the proportion of high-carbon-emission units. The larger the value, the more important it is to determine whether the carbon emission constraint limit has been exceeded. This value is the source of all subsequent judgments and calculations in the carbon emission constraint system. It is used to determine whether the total carbon emission limit has been exceeded, and it also serves as the input benchmark for calculating the carbon quota gap and the breakdown of carbon compliance costs.

[0196] Time-of-use carbon emission control limits refer to the limits set by the government's carbon emission reduction authorities during specific time periods. The maximum allowable total carbon dioxide emissions from grid-connected thermal power units are as follows: This upper limit is determined by the power supply tasks undertaken by the power grid and its benchmark carbon emission intensity, and is a rigid and inviolable carbon emission red line. and The difference is: For free quotas, lower than Exceeding The excess can be made up by purchasing quotas on the market, but the excess... Some of these emissions are administrative violations and are not permitted under any circumstances.

[0197] The time-of-use carbon emission quota refers to the requirement that the actual carbon emissions of thermal power units be capped for each time period. Not exceeding the time-of-use carbon emission control limit The inequality restrictions are expressed mathematically as follows:

[0198] .

[0199] This constraint is a rigid, unbreakable safety constraint, located at the fifth layer in the hierarchical verification mechanism set in step S3. When the thermal power output scheme given by the optimization algorithm leads to... At that point, the particle was deemed to have exceeded the carbon emission limit and was subject to dynamic penalties, forcing the algorithm to adjust the thermal power output plan to below the total carbon emission limit. This hard constraint corresponds to carbon administrative management measures.

[0200] Time-of-use (TOU) free carbon emission allowances for the power grid refer to the allowances granted by the government's carbon reduction authorities during specific time periods. Carbon emission allowances allocated to the power grid free of charge, This quota is typically below the carbon emission control ceiling. lower than The emissions portion constitutes a quota surplus, which can be sold for profit in the carbon market. This parameter is derived from the carbon emission quota and carbon market price data read in step S1, serving as a common calculation benchmark for both the carbon quota surplus revenue item and the carbon compliance cost item.

[0201] Time-of-use carbon quota gap refers to the period during which the carbon quota gap is filled. Actual carbon emissions of internal thermal power units Exceeding the grid's time-of-use free carbon emission quota Part of, It means that when When the carbon allowance is insufficient, the excess must be purchased from the carbon market or subject to penalties; when This indicates a surplus in carbon allowances, which can be sold for profit. This deficit value serves as a common calculation benchmark for both the carbon allowance surplus revenue item and the carbon compliance cost item.

[0202] Carbon quota compliance balance constraints refer to the actual carbon emissions of thermal power units Time-of-use free carbon emission allowances with the power grid The difference is defined as the time-of-use carbon quota gap. The equation relationship is expressed mathematically as follows: This constraint is a numerical definition rather than an inequality restriction. Its function is to establish a quantitative link between actual carbon emissions and free allowances, providing a unified gap calculation benchmark for carbon allowance surplus revenue and carbon compliance costs. When the carbon quota surplus revenue term in the objective function is a negative contribution and the carbon compliance cost term is a positive expenditure, the double penalty prompts the algorithm to reduce thermal power output; when At that time, the carbon allowance surplus revenue is a positive contribution and the first carbon compliance cost is a negative expenditure. The dual reward recognizes the low-carbon output scheme, thus making the carbon dimension constraint consistently quantified in the objective function.

[0203] Specifically, in the S3 step of constructing the set of coupled constraints, carbon emission constraints are written into the optimization model as two types of sub-constraints with different properties. The first type is rigid inequality constraints:

[0204] :

[0205] When a particle's thermal power output scheme causes its actual carbon emissions to exceed the time-of-use control limit during a certain period, that particle violates the hard carbon emission constraint and is subject to dynamic penalties. This constraint is a bottom-line constraint for the safe operation of the power grid and cannot be exempted through market means. Therefore, it is located at the fifth layer in the S3 step layered verification and, together with the reserve capacity constraint, constitutes an insurmountable hard boundary.

[0206] The second type is numerically defined equality constraints:

[0207] :

[0208] This constraint does not directly limit the feasibility of particles, but rather provides a unified framework for carbon allowance surplus revenue and carbon compliance costs. Calculation baseline. In step S2, the carbon allowance surplus item. Directly referencing this gap value, when (i.e., when there is a surplus) positive revenue is generated; carbon compliance cost item:

[0209] ,

[0210] The market-based procurement costs and penalty costs are also calculated based on this gap value. The relationship between the two types of constraints is clear: the rigid constraint defines the absolute upper limit of total carbon emissions. The carbon quota gap is defined by... and The difference range between the two values ​​maps the actual carbon emissions to the benefit-cost coordinate of the objective function, so that the economic impact of carbon emissions in both administrative assessment and market transactions is perceived by the optimization algorithm.

[0211] The two types of carbon emission constraints, together with the carbon-related terms in the objective function, form a three-tiered progressive balance: hard carbon emission constraints Adhering to the administrative emission red line, carbon quota surplus revenue item Positive incentives are given to low-carbon solutions, and carbon compliance costs are included. The system imposes market penalties on excessive emission schemes. A three-tiered coordination mechanism ensures that optimized schemes do not violate carbon emission control red lines at any time. Simultaneously, it proactively reduces carbon emissions when carbon market prices are high to save on compliance costs, and moderately increases carbon emissions when carbon market prices are low to obtain allowance surpluses. This achieves coordinated optimization of carbon administrative red lines and carbon market price signals in scheduling decisions.

[0212] S4. Using a hierarchical constraint adaptive correction particle swarm algorithm, the global optimal particle is obtained by calculating based on the comprehensive benefit objective function and the constraint processing mechanism.

[0213] Specifically, the hierarchical constraint adaptive modified particle swarm optimization algorithm is a heuristic global optimization method for mixed-integer nonlinear programming problems. In this model, the decision variables include the start-up and shutdown states of the hydropower units. 0-1 discrete variables, active power output of thermal power units Continuous variables and reserve capacity of thermal power units The problem involves continuous variables and belongs to the mixed-integer optimization problem. Traditional gradient descent methods cannot handle the coupling between discrete variables and nonlinear constraints. The particle swarm optimization (PSO) algorithm, by simulating swarm intelligence search behavior, evaluates a large number of candidate solutions in parallel within the feasible region, effectively avoiding local extremum traps in non-convex feasible regions. Each particle in the swarm represents a complete 24-hour combined water and fire scheduling scheme, and its position vector... Includes all decision variables , and velocity vector This includes the iterative adjustment step size corresponding to each decision variable. Based on the comprehensive benefit objective function constructed in step S2 and the constraint handling mechanism set in step S3, the fitness of each particle is calculated. The fitness value directly reflects the comprehensive economic benefit of the scheduling scheme. The individual historical best position of each particle during the iteration process is recorded. And the global optimal position of the population among all particles. This guides the population to gather in areas with high fitness.

[0214] In some implementations, the use of a hierarchical constraint adaptive correction particle swarm optimization algorithm, which calculates the globally optimal particle based on the comprehensive benefit objective function and the constraint processing mechanism, includes:

[0215] Based on the comprehensive benefit objective function and the set of coupling constraints, a multidimensional particle population is generated. Each particle contains variables of the start-up and shutdown status of hydropower units, variables of the active power output of thermal power units, and variables of the reserve of thermal power units for each time period.

[0216] According to the power grid security priority from high to low, each constraint in the set of coupled constraints is checked in a step-by-step manner, and the dynamic penalty value is added to the particle fitness for particles that violate the constraints.

[0217] Particle velocity and position are updated using linearly decreasing inertial weights;

[0218] For the updated particles, the discrete variables of the start-stop state of hydropower units, the continuous variables of the active power output of thermal power units, and the continuous variables of the reserve capacity of thermal power units are over-limited. The correction is carried out by taking the start-stop state as close to 0 or 1 as the nearest value and the output value as the minimum stable combustion output or maximum output of the unit as the correction benchmark.

[0219] When the global optimal solution does not improve for several consecutive rounds, select some particles for perturbation and mutation with an adaptively adjusted mutation probability;

[0220] Repeat the iteration until the convergence condition is met to obtain the globally optimal particle.

[0221] Specifically, a linearly decreasing inertia weight is adopted. Update particle velocity:

[0222] ,

[0223] in For the current iteration round, The maximum number of iterations, Take 0.9, A weight of 0.4 is used to maintain a larger weight in the early stages of iteration to expand the global search scope, and then a smaller weight is used in the later stages of iteration to enhance the local fine-tuning search. The particle velocity update formula is:

[0224] ,

[0225] in and The learning factor is set to 2.0. and The values ​​are uniformly random numbers in the range of 0 to 1. The particle position update formula is:

[0226] .

[0227] Regarding the start-up and shutdown status of hydropower units Characterized by its 0-1 discrete variable nature, it doesn't directly superimpose continuous velocity values ​​onto position. Instead, it uses velocity magnitude as the threshold for state reversal. That is, the start / stop state is reversed when the absolute velocity value exceeds a set threshold; otherwise, the original state is maintained. This allows for synchronous updates of discrete and continuous variables within a unified iterative framework. For instances of hydropower start / stop state exceeding limits, thermal power output exceeding limits, and thermal power reserve exceeding limits in the updated particles, targeted corrections are implemented: when the start / stop state deviates from the 0 or 1 range, it is reset to the nearest 0 or 1; when the output is higher than the upper limit, it falls back to the maximum output value; when it is lower than the lower limit, it rises to the stable combustion lower limit; when the reserve capacity exceeds the limit, it adaptively adjusts to match the thermal power unit's operating scale. During the iteration process, the global optimal fitness is continuously monitored. If five consecutive iterations occur... If no update is performed, a subset of particles are randomly selected and perturbed using an adaptively increased mutation probability (from a base value of 0.05 to 0.12). This randomly adjusts the internal hydropower start-stop combination and thermal power output allocation scheme, breaking the homogeneous distribution of the population and helping the algorithm escape local optima. The algorithm continues until the preset maximum number of iterations is reached. Or the change in the global optimal fitness value is less than the preset convergence threshold after 8 consecutive iterations. When the iteration terminates, output the current iteration. As the globally optimal particle.

[0228] The following example illustrates the hierarchical constraint adaptive correction particle swarm algorithm.

[0229] Step 1: Initialization of global parameters for the algorithm

[0230] All hyperparameters of the algorithm were assigned fixed values. All parameter values ​​are consistent with the actual day-ahead dispatching of the high-hydropower grid, with no subjective random assignment. The specific parameter settings are as follows:

[0231] Particle population size M: takes a value of 100, covering the feasible domain of all operating conditions of hydro-thermal units, to avoid premature convergence in optimization;

[0232] Maximum number of iterations The value is 120, which is suitable for the optimization time of the model under high-dimensional coupling constraints;

[0233] Linearly decreasing inertia weight: Upper limit of inertia weight Lower bound of inertia weight In the early stages of iteration, large weights expand the global search scope, while in the later stages of iteration, small weights strengthen local fine-grained optimization.

[0234] Learning factors: Individual learning factors Global learning factor Balancing the individual particle's historical optimal experience with the population's global optimal experience;

[0235] Constraint dynamic penalty coefficient gradient: The initial penalty coefficient is 10. Every 10 iterations, the penalty coefficient increases linearly by 1.5 times the gradient, gradually tightening the constraint boundary.

[0236] Particle adaptive mutation probability: The basic mutation probability is 0.05. When the global optimal solution has no update for 5 consecutive iterations, the mutation probability is adaptively increased to 0.12 to help the particle escape the local optimal solution.

[0237] Step 2: Generation of multi-dimensional particle encoding including hydropower start-stop states

[0238] Clarify the correspondence between particles, generator sets, and scheduling schemes: a single particle does not correspond to a single generator, and a single particle represents a complete 24-hour day-ahead water-thermal joint scheduling scheme.

[0239] Constructing multidimensional particle position vectors: The vector dimension contains all three types of decision variables: hydropower unit start-stop 0-1 variables, thermal power unit active power output variables, and thermal power unit standby and reserve variables.

[0240] Synchronous definition of particle velocity vector: Speed ​​is the iterative correction step size of the corresponding dimension decision variable. Positive or negative speed indicates the tendency to increase / decrease output and switch between start and stop states. The speed value represents the magnitude of variable adjustment.

[0241] An initial population is randomly generated, while invalid particles that directly violate the upper and lower limits of unit output and start-up / shutdown logic in the initial stage are removed, reducing the redundancy of subsequent iteration calculations.

[0242] Step 3: Multi-constraint ladder verification and particle fitness calculation

[0243] A tiered constraint verification mechanism, progressing from loose to tight, is adopted. All constraints are verified sequentially according to the power grid safety priority, and gradient dynamic penalties are applied to violating particles.

[0244] First layer of verification: matching constraints between the start-up and shutdown of hydropower units and their output, to prevent logical errors that still result in active power output when the unit is shut down;

[0245] The second layer of verification: system active power balance constraints, verifying the matching relationship between the entire network's power generation and load, and the power transmitted to other regions;

[0246] The third layer of verification: lower limit on the number of thermal power units started and unit ramp-up rate constraints, to ensure the safety of thermal power unit operation;

[0247] Fourth layer of verification: Rigid reserve capacity constraints of the system during the dry season to safeguard the bottom line of power grid operation safety;

[0248] The fifth layer of verification includes time-of-use carbon emission limits and carbon quota compliance balance constraints, ensuring the system meets low-carbon operation requirements.

[0249] The step constraint penalty value is superimposed on the objective function, the comprehensive fitness value for each particle is calculated, and the optimal position of each particle during the iteration process is recorded. and the global optimal position of the population .

[0250] Step 4: Iterative Update of Particle Velocity and Position

[0251] The particle velocity and position are updated using a linearly decreasing inertia weight, and the update formula is as follows:

[0252] 1) Formula for dynamic update of inertia weight:

[0253] ;

[0254] In the formula: For the current iteration round, This represents the maximum number of iterations.

[0255] 2) Particle velocity update formula:

[0256] ;

[0257] 3) Particle position update formula:

[0258] ;

[0259] In the formula: Uniform random numbers in the range of 0 to 1 are used to increase the randomness of the population search; for the 0-1 discrete variables of hydropower start-up and shutdown, continuous speed values ​​are not directly superimposed, but only the speed magnitude is used as the threshold for the start-up and shutdown state reversal, so as to realize the synchronous iterative update of discrete variables and continuous variables.

[0260] Step 5: Water and electricity start-up and shutdown and directional correction of operating variables exceeding limits

[0261] To address the common issue of variable exceeding limits during iteration, scenario-specific targeted corrections are implemented to avoid ineffective iterations:

[0262] Hydropower start / stop status over-limit correction: If the start / stop status calculated by the particle deviates from the 0 / 1 integer range, it will be directly forced to return to 0 or 1 as the nearest value to ensure that the start / stop logic is compliant;

[0263] Unit output and reserve capacity over-limit correction: If the output is higher than the upper limit, it will be forced to drop back to the maximum output; if it is lower than the lower limit, it will be raised to the minimum stable combustion output of the unit; if the reserve capacity exceeds the limit, it will be synchronously matched with the thermal power plant start-up scale for adaptive adjustment.

[0264] Global constraint over-limit correction: When reserve capacity is insufficient or carbon emissions exceed the standard, priority will be given to increasing the start-up scale of thermal power and reserve capacity, and then slightly reducing the grid-connected output of wind and solar power, so as to prioritize the protection of grid security constraints.

[0265] Step 6: Particle Adaptive Mutation Operation

[0266] The algorithm monitors changes in the global optimal fitness value of the population. If the global optimal solution does not improve significantly after 5 consecutive iterations, it is determined that the algorithm is trapped in a local optimum. Based on the adaptively adjusted mutation probability, some particles in the population are randomly selected, and their internal hydropower unit start-stop combination and thermal power output allocation scheme are randomly perturbed and mutated to break the existing population distribution structure, broaden the algorithm's search space, and improve the global optimization capability.

[0267] Step 7: Iterative Convergence Determination and Initial Scheduling Scheme Output

[0268] A dual convergence criterion is set, and the iterative calculation will terminate when either condition is met:

[0269] Rule 1: The number of iterations reaches the preset maximum number of iterations. Force the iteration to stop;

[0270] Criterion 2: After 8 consecutive iterations, the change in the global optimal fitness of the population is less than the preset minimum convergence threshold. The algorithm is considered to have converged, and the optimization process is complete.

[0271] After iterative convergence, the position vector corresponding to the globally optimal particle in the population is extracted, and the optimal scheduling results are directly output: the daily start-up and shutdown plan of hydropower units, the optimal active power output scheme of thermal power units, and the optimal reserve scheme of thermal power units.

[0272] S5. Extract the start-stop status commands of hydropower units, the planned active power output of thermal power units, and the planned reserve capacity of thermal power units from the position vector of the globally optimal particle, and output them as a collaborative optimization scheduling scheme.

[0273] Specifically, the position vector of the globally optimal particle. Includes the optimal combination of all decision variables over all time periods, including the start-up and shutdown state variables of hydropower units. The dimension is Variable active power output of thermal power units The dimension is Variable in standby reserve capacity of thermal power units The dimension is also The values ​​of each dimension are extracted from the location vector according to the variable type and unit number, and the start-up and shutdown status commands of the hydropower unit at each time period are obtained respectively. Planned active power output of thermal power units in each time period and the planned reserve capacity of thermal power units in each time period. These output instructions serve as the output of a collaborative optimization scheduling scheme. The time granularity of all three types of output instructions is hourly, covering the 24 scheduling periods per day, forming a complete sequence of scheduling instructions that can be directly issued to each power plant for execution. Specifically, the start-stop status instructions for hydropower units are used to control the start-stop operations of cascade hydropower units; the planned active power output values ​​for thermal power units are used to control the power generation of thermal power units in each time period; and the planned reserve capacity values ​​for thermal power units are used to ensure that the power grid maintains necessary emergency response capabilities at any time. Thus, under the premise of satisfying all constraints regarding power balance, unit limits, ramp-up, reserve, and carbon emissions, the scheduling system has obtained the optimal daytime operation plan that balances wind and solar power consumption with various cost expenditures.

[0274] Example 2

[0275] Please see Figure 2 This invention provides a wind-solar integrated hydropower-thermal power coordinated dispatching device for power grids in high-hydropower areas during the dry season, comprising:

[0276] The acquisition module 201 is used to acquire system load data, new energy output forecast data, hydropower and thermal power unit operating parameter data, system reserve demand data, carbon emission quota and carbon market price data, and electricity reserve market price data within the scheduling cycle as basic input parameters.

[0277] The objective function module 202 is used to limit the upper limit of actual new energy consumption based on the new energy output prediction data, limit the output range of hydropower units based on the operating parameters of hydropower units, and construct a comprehensive revenue objective function with new energy consumption revenue and carbon quota surplus revenue as positive terms, and hydropower unit start-up and shutdown losses, thermal power peak shaving compensation, thermal power operation, standby procurement and carbon compliance costs as negative terms.

[0278] The constraint module 203 is used to establish a set of coupled constraint conditions, including power balance constraints, unit output limit constraints, ramp constraints, reserve capacity constraints and carbon emission constraints, based on the basic input parameters, and to set a constraint processing mechanism of hierarchical verification and dynamic penalty according to safety priority.

[0279] The calculation module 204 is used to perform calculations based on the comprehensive benefit objective function and the constraint processing mechanism using a hierarchical constraint adaptive correction particle swarm algorithm to obtain the globally optimal particle.

[0280] The parsing module 205 is used to parse the start-stop status commands of hydropower units, the planned active power output of thermal power units, and the planned reserve capacity of thermal power units in each time period from the position vector of the globally optimal particle, and output them as a collaborative optimization scheduling scheme.

[0281] It should be noted that each module and unit in the wind-solar-hydropower co-dispatch device for high-hydropower areas during the dry season in this embodiment corresponds one-to-one with each step in the wind-solar-hydropower co-dispatch method for high-hydropower areas during the dry season in the aforementioned embodiment. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned wind-solar-hydropower co-dispatch method for high-hydropower areas during the dry season, and will not be repeated here.

[0282] Example 3

[0283] Please see Figure 3 This embodiment provides an electronic device, including at least one processor 301 and a memory 302. Optionally, the device further includes a communication component 303. The processor 301, memory 302, and communication component 303 are connected via a bus 304.

[0284] In a specific implementation, at least one processor 301 executes computer execution instructions stored in memory 302, causing at least one processor 301 to perform the above-described method.

[0285] The specific implementation process of processor 301 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0286] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0287] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0288] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0289] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0290] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0291] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0292] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0293] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0294] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0295] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0296] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0297] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0298] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A wind-solar integrated hydropower-thermal power coordinated dispatch method for power grids in high-hydropower areas during the dry season, characterized in that, include: The system load data, renewable energy output forecast data, hydropower and thermal power unit operating parameter data, system reserve demand data, carbon emission quota and carbon market price data, and electricity reserve market price data within the scheduling cycle are obtained as basic input parameters. Based on the new energy output forecast data, the upper limit of the actual consumption of new energy is defined. Based on the operating parameters of the hydropower units, the output range of the hydropower units is defined. The new energy consumption revenue and carbon quota surplus revenue are positive terms, and the start-up and shutdown losses of hydropower units, peak-shaving compensation of thermal power, thermal power operation, standby procurement and carbon compliance costs are negative terms, and a comprehensive revenue objective function is constructed. Based on the aforementioned basic input parameters, a set of coupled constraint conditions is established, including power balance constraints, unit output limit constraints, ramping constraints, reserve capacity constraints, and carbon emission constraints. A constraint processing mechanism with hierarchical verification and dynamic penalty based on safety priority is also set. A hierarchical constraint adaptive correction particle swarm optimization algorithm is adopted to calculate the globally optimal particle based on the comprehensive benefit objective function and the constraint processing mechanism. The start-stop status commands of hydropower units, the planned active power output of thermal power units, and the planned reserve capacity of thermal power units in each time period are extracted from the position vector of the globally optimal particle and used as the output of the collaborative optimization scheduling scheme.

2. The method according to claim 1, characterized in that, The operating parameter data of the hydropower and thermal power units include the adjustable output upper and lower limits of each hydropower unit, the stable combustion lower limit and maximum output of each thermal power unit, and the ramp rate data of each thermal power unit. The new energy output forecast data includes upper limit data for wind power output forecast and upper limit data for photovoltaic output forecast; The carbon emission allowance and carbon market price data include grid time-of-use free carbon emission allowance data, carbon market time-of-use transaction price data, and carbon excess penalty unit price data. The electricity standby market price data refers to the time-of-use unit capacity electricity price data in the electricity standby market.

3. The method according to claim 2, characterized in that, The upper limit of actual renewable energy consumption is defined based on the renewable energy output forecast data, including: The upper limit of the predicted wind power output is used as the upper limit of the actual active power absorbed by wind power, and the upper limit of the predicted photovoltaic output is used as the upper limit of the actual active power absorbed by photovoltaic power, thereby limiting the maximum grid-connected power of wind power and photovoltaic power in each time period.

4. The method according to claim 1, characterized in that, The revenue from renewable energy consumption is calculated by multiplying the sum of the average on-grid electricity price of wind and solar power, the revenue from green certificate transactions, and the unit price of the renewable energy excess consumption policy incentive by the actual active power consumed by wind and solar power. The carbon allowance surplus revenue is obtained by multiplying the carbon market time-of-use transaction price by the time-of-use carbon allowance surplus amount, which is obtained by subtracting the actual carbon emissions of thermal power units from the grid time-of-use free carbon emission allowance. The start-up and shutdown losses of the hydropower units are obtained by multiplying the unit price of the single start-up and shutdown loss by the absolute value of the change in the start-up and shutdown status of the unit, and then summing them over all hydropower units and over the entire time period. The thermal power peak shaving compensation is obtained by multiplying the deep peak shaving unit compensation standard with the difference between the unit output and the peak shaving benchmark value, and then summing the results for all thermal power units and all time periods. The standby procurement cost is obtained by multiplying the time-of-use unit capacity electricity price in the power standby market with the total standby capacity procured by the power grid and then summing them up over the entire time period. The carbon compliance cost is calculated by multiplying the carbon market time-of-use transaction price by the carbon allowance gap, plus the carbon overrun penalty price by the excess amount of the carbon allowance gap, and then summing these products over the entire time period. The carbon allowance gap is obtained by subtracting the grid's time-of-use free carbon emission allowance from the actual carbon emissions of thermal power units.

5. The method according to claim 1, characterized in that, The reserve capacity constraint is obtained by subtracting the current active power output from the maximum output of the thermal power unit to obtain the maximum reserve capacity that a single thermal power unit can provide. A rigid reserve capacity constraint for the power grid during the dry season is established, with each thermal power unit's actual reserved reserve capacity not exceeding its own maximum reserve capacity and the sum of all thermal power units' actual reserved reserve capacity not being lower than the system's minimum reserve requirement.

6. The method according to claim 5, characterized in that, The carbon emission constraint is obtained by multiplying the carbon emission intensity of the thermal power unit by the active power output of the thermal power unit to obtain the actual carbon emission of the thermal power unit. A hard constraint is established on the total amount of time-of-use carbon emissions, based on the premise that the actual carbon emissions of the thermal power units do not exceed the time-of-use carbon emission control limit. The time-of-use carbon quota gap is defined by the difference between the actual carbon emissions of the thermal power units and the time-of-use free carbon emission quota of the power grid, and a carbon quota compliance balance constraint is established.

7. The method according to claim 1, characterized in that, The layered constraint adaptive correction particle swarm optimization algorithm calculates the globally optimal particle based on the comprehensive benefit objective function and the constraint processing mechanism, including: Based on the comprehensive benefit objective function and the set of coupling constraints, a multidimensional particle population is generated. Each particle contains variables of the start-up and shutdown status of hydropower units, variables of the active power output of thermal power units, and variables of the reserve of thermal power units for each time period. According to the power grid security priority from high to low, each constraint in the set of coupled constraints is checked in a step-by-step manner, and the dynamic penalty value is added to the particle fitness for particles that violate the constraints. Particle velocity and position are updated using linearly decreasing inertial weights; For the updated particles, the discrete variables of the start-stop state of hydropower units, the continuous variables of the active power output of thermal power units, and the continuous variables of the reserve capacity of thermal power units are over-limited. The correction is carried out by taking the start-stop state as close to 0 or 1 as the nearest value and the output value as the minimum stable combustion output or maximum output of the unit as the correction benchmark. When the global optimal solution does not improve for several consecutive rounds, select some particles for perturbation and mutation with an adaptively adjusted mutation probability; Repeat the iteration until the convergence condition is met to obtain the globally optimal particle.

8. A wind-solar integrated hydropower-thermal power coordinated dispatching device for power grids in high-hydropower areas during the dry season, characterized in that, include: The acquisition module is used to acquire system load data, new energy output forecast data, hydropower and thermal power unit operating parameter data, system reserve demand data, carbon emission quota and carbon market price data, and electricity reserve market price data within the scheduling cycle, as basic input parameters. The objective function module is used to limit the upper limit of actual new energy consumption based on the new energy output prediction data, limit the output range of hydropower units based on the operating parameters of hydropower units, and construct a comprehensive revenue objective function with new energy consumption revenue and carbon quota surplus revenue as positive terms, and hydropower unit start-up and shutdown losses, thermal power peak-shaving compensation, thermal power operation, standby procurement and carbon compliance costs as negative terms. The constraint module is used to establish a set of coupled constraint conditions, including power balance constraints, unit output limit constraints, ramping constraints, reserve capacity constraints and carbon emission constraints, based on the basic input parameters, and to set a constraint processing mechanism of hierarchical verification and dynamic penalty according to safety priority. The calculation module is used to calculate the globally optimal particle using a hierarchical constraint adaptive correction particle swarm algorithm based on the comprehensive benefit objective function and the constraint processing mechanism. The parsing module is used to parse the start-stop status commands of hydropower units, the planned active power output of thermal power units, and the planned reserve capacity of thermal power units in each time period from the position vector of the globally optimal particle, and output them as a collaborative optimization scheduling scheme.

9. An electronic device, characterized in that, include: At least one processor, at least one memory, and computer program instructions stored in the memory, which, when executed by the processor, implement the method as described in any one of claims 1-7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, The method as described in any one of claims 1-7 is implemented when the computer program instructions are executed by the processor.

Images