A low-carbon economic dispatching method for power system based on multi-type demand response and heat storage reconstruction

Abstract

The application aims to disclose a low-carbon economic dispatching method of power system considering multi-type demand response and heat storage reconstruction. First, different types of load characteristics are analyzed, and a multi-type demand response improvement model is established. According to the characteristics of different types of loads, appropriate demand response methods are adopted to fully tap the flexibility of load side resources. Second, the operation mechanism of multi-type demand response and heat storage reconstruction in promoting carbon emission reduction is analyzed, and a low-carbon economic layered model of power system considering multi-type demand response and heat storage reconstruction is established. Finally, the improved 30-machine system is taken as an example for calculation analysis, and the results show that the established model can effectively reduce the system carbon emission under the premise of controlling the comprehensive operation cost of the system.

Description

Technical Field

[0001] This invention belongs to the field of low-carbon economic dispatch of power systems, and specifically relates to a low-carbon economic dispatch method for power systems based on multi-type demand response and thermal storage retrofit. Background Technology

[0002] Renewable energy sources such as wind and solar power are gradually replacing high-carbon emission power sources due to their zero carbon emissions, mature technologies, and low marginal costs. However, the uncertainty of wind and solar power output not only limits their absorption capacity but also increases the peak-valley load difference, exacerbating the peak-shaving pressure on thermal power units and increasing their carbon emissions. Therefore, improving wind and solar power absorption while reducing carbon emissions from peak-shaving units is of great significance for achieving an economically low-carbon power system. Furthermore, in-depth exploration of demand response methods for load-side resources and flexible retrofitting technologies for thermal power units offer new pathways for carbon emission reduction.

[0003] Demand response can tap into adjustable resources on the demand side, promoting the consumption of renewable and clean energy and reducing system carbon emissions through reasonable load adjustment. Currently, scholars have studied the impact of demand response (DR) on the low-carbon and economic aspects of the power system. Many studies use the same demand response method for various loads without optimizing demand response pricing. There is a lack of literature utilizing multiple demand response methods to classify and implement policies for different types of loads to fully enhance load-side flexibility.

[0004] However, due to the uncertainties surrounding wind and solar power, the low-carbon principle of demand response—which involves shifting peak loads to off-peak periods to absorb wind curtailment—remains insufficient. Since current peak-shaving power sources are still primarily thermal power, it is necessary to fully utilize the flexibility of thermal power units to compensate for the inadequacy of the demand response low-carbon principle when off-peak wind curtailment levels are low. Currently, there is limited research considering the combination of deep peak shaving by thermal power units and demand response technology to reduce system carbon emissions. Summary of the Invention

[0005] To address the above problems, this invention provides a low-carbon economic dispatch method for power systems based on multi-type demand response and thermal storage retrofitting. First, the characteristics of different load types are analyzed, and an improved multi-type demand response model is established. Appropriate demand response methods are adopted according to the characteristics of different load types to fully exploit load-side flexibility resources. Second, the operational mechanism of multi-type demand response and thermal storage retrofitting in promoting carbon emission reduction is analyzed, and a low-carbon economic hierarchical model for power systems considering multi-type demand response and thermal storage retrofitting is established. Finally, a case study analysis using an improved 30-unit system demonstrates that this invention can effectively reduce system carbon emissions while controlling the overall operating cost of the system.

[0006] This invention proposes a low-carbon economic dispatch method for power systems based on multi-type demand response and thermal storage retrofitting. The specific design scheme is as follows:

[0007] Step 1: Based on the differences in price response and electricity consumption characteristics of different types of users, the load is divided into sensitive, relatively sensitive, and insensitive loads. Improved time-of-use pricing response, interruptible incentive response, and non-participation in response processing are adopted respectively, and response models are established.

[0008] Step 2: Based on the response model in Step 1, analyze the operation mechanism of multi-type demand response and the low-carbon economic operation of thermal storage retrofit, as well as the low-carbon economic operation mechanism of multi-type demand response combined with thermal storage retrofit.

[0009] Step 3: Based on the operating mechanism in Step 2, establish a low-carbon economic dispatch model and a robust optimization model for the power system that takes into account multiple types of demand response and thermal storage retrofitting.

[0010] Step 4: Based on the low-carbon economic dispatch model proposed in Step 3, establish the corresponding objective function and model constraints.

[0011] Furthermore, in step one, response models for different types of loads are established. The specific load response models are as follows:

[0012] The time-of-use pricing mechanism based on a consumer psychology model divides a scheduling cycle into three time periods: peak, flat, and valley. By setting electricity prices for these three time periods, it guides users to change their electricity consumption times, thereby optimizing the load curve. The calculation model for the load transfer rate during peak and valley periods and the peak-valley price difference is as follows:

[0013]

[0014] In the formula Represents the load transfer rate of type n. For the peak-valley electricity price difference, R n,t Let a be the slope of the linear region. n,t C represents the minimum electricity price difference during load transfer. n,t This represents the maximum load transfer rate between peak and valley periods.

[0015] The optimized load calculation model in the time-of-use electricity price response model, which optimizes and improves the electricity prices for peak, flat, and valley periods, is as follows:

[0016]

[0017] In the formula For user n in time period t after the time-of-use pricing response, and They represent The maximum and minimum values ​​during time period t. Peak, flat, and trough periods. and The calculation formula is as follows:

[0018]

[0019]

[0020]

[0021] In the formula The following are the formulas for calculating the time-of-use electricity price response cost, representing peak-to-flat, peak-to-valley, and flat-to-valley load transfer rates, respectively:

[0022]

[0023] In the formula: P n,t and These represent the load of user n in time period t before and after the time-of-use pricing response; w n,t and C represents the electricity price for user n during time period t before and after the time-of-use pricing response. Itou For time-of-use pricing response costs, N Itou The number of users participating in the time-of-use pricing response.

[0024] This invention aggregates dispersed demand response resources through a "load aggregator" (LA), making demand response implementation for interruptible loads more flexible. The load timing expression after interruptible load response is as follows:

[0025]

[0026] In the formula: and These represent the loads before and after the interruptible load excitation time period t; v n,t v represents the amount of charge reduction by user n during time period t. n,t A value of 0 or 1 indicates the normal state and the interrupted state, respectively.

[0027] Let n be the load that can be reduced for user n during time period t; n is the load aggregator that can provide interruptible load, and there are a total of N. IL One interruptible user aggregator. N IL =1, 2, 3, 4 represent the interruptible aggregators for industry, commerce, residential, and agriculture, respectively. During the scheduling process, the scheduling cost of interruptible demand response can be expressed as follows:

[0028]

[0029] In the formula, K t The unit power reduction compensation cost is the interruptible load capacity corresponding to the interruptible load capacity during time period t.

[0030] Furthermore, the specific steps of step two are as follows:

[0031] Price-based demand response (PDR) shifts reduced peak loads to off-peak periods through electricity pricing mechanisms, increasing wind power absorption during off-peak hours. Reduced peak loads also help decrease the frequency of thermal power unit start-ups and shutdowns. The low-carbon economic mechanism of PDR is equivalent to replacing expensive high-carbon unit output during peak periods with economical zero-carbon wind power or economical unit output.

[0032] IDR (Incentive-Driven Demand Response) can provide the system with spinning reserves by increasing or decreasing load shedding, effectively reducing the reserve capacity undertaken by thermal power units. IDR can also reduce the output of thermal power units by cutting peak loads, which is equivalent to replacing the output of high-carbon emission units with zero-carbon virtual power sources.

[0033] While demand response improves wind power absorption and reduces the number of unit start-ups and shutdowns by "peak shaving and valley filling," thus lowering system economic costs, its carbon reduction effect is still not ideal. This is mainly because the output of high-carbon emission units reduced by demand response is replaced by previously curtailed wind power and low-carbon emission units during off-peak periods. However, when the amount of curtailed wind power during off-peak periods is not high, the main source of power is still high-carbon emission units. When wind power absorption increases to a certain level, the degree of net load fluctuation may increase, thereby increasing the period of deep peak shaving for thermal power units and increasing system carbon emissions.

[0034] The low-carbon economic mechanism of thermal energy storage retrofit is as follows:

[0035] Energy transfer within a coal-fired power plant mainly involves three processes: 1) The boiler burns coal to generate heat, which heats water to produce high-temperature, high-pressure steam that flows into the high-pressure cylinder of the steam turbine to perform work; 2) The steam in the high-pressure cylinder flows into the boiler to form hot steam, which is then introduced into the low-pressure and intermediate-pressure cylinders of the turbine to perform work; 3) The condenser tower condenses the gas from the low-pressure cylinder into liquid water, which is then heated by a portion of the steam extracted from the cylinder to become high-temperature feedwater, and this process is repeated. The tight thermal coupling of the boiler, steam turbine, generator, and other components in a traditional coal-fired unit, along with boiler combustion stability conditions, limits the conventional peak-shaving depth of the unit. Traditional peak-shaving methods for coal-fired power units are mainly divided into "conventional peak-shaving," "peak-shaving without oil injection," and "peak-shaving with oil injection." As the peak-shaving depth increases, the unit's energy consumption per unit of electricity generated and carbon emissions increase significantly.

[0036] Thermal energy storage retrofit of power units involves adding a high- and low-temperature dual-cycle thermal energy storage device to the steam power cycle to store thermal energy. Under the condition of stable boiler combustion, this decouples the thermal energy between different parts of the unit, enabling a more flexible output modification for thermal power units. By increasing the thermal energy storage system to store heat, the output of thermal power units is limited to the conventional peak-shaving range. This is effectively equivalent to converting the output of high-carbon-emission units during periods of deep peak shaving into the output of low-carbon-emission units during conventional peak shaving, reducing the number of deep peak shaving operations and thus reducing the unit's carbon emissions per unit, while also reducing the cost of deep peak shaving. However, when the load peak-valley difference is large, due to the limitations of the thermal energy storage system capacity and the system's spinning reserve capacity, the output of high-carbon-emission units during peak hours still leads to an increase in system carbon emissions.

[0037] To verify the effectiveness of the proposed model, this paper compares and analyzes the low-carbon economic operation mechanism of multi-type demand response and thermal storage retrofitting in the following four scenarios:

[0038] 1) Scenario 1, without considering demand response and thermal power unit thermal storage retrofit;

[0039] 2) Scenario 2, considering the ordinary demand response type (TOU+IL), consider the thermal power unit heat storage retrofit;

[0040] 3) Scenario 3, considers the improvement model of multi-type demand response, but does not consider the thermal power unit thermal storage retrofit;

[0041] 4) Scenario 4, consider the improvement model of multi-type demand response, and consider the thermal storage retrofit of thermal power units (the model in this paper).

[0042] When the system only considers multiple types of demand response, it is equivalent to replacing the output of Scenario 4 during peak load periods with the output of Scenario 1. If the amount of wind curtailment is small, the proportion of units operating in Scenario 1 decreases, and the output of high-carbon emission units increases. After introducing thermal storage unit retrofits, the increased unit output through the thermal storage system is converted to Scenario 2 output, which is equivalent to converting the output of high-carbon emission units in deep peak shaving into the output of conventional low-carbon emission units. At the same time, it avoids the problem of increased carbon emissions from thermal power units during deep peak shaving periods caused by the increased fluctuation of the net load curve after considering demand response. When only the system operation mode after the unit thermal storage retrofit is considered, the large peak-valley difference means that high-carbon emission units still operate during peak periods, and the system is still insufficient in terms of low-carbon economy.

[0043] Furthermore, the specific steps for establishing a low-carbon economic dispatch model for the power system that takes into account multiple types of demand response and thermal storage retrofitting in step three are as follows:

[0044] The upper-level model optimizes the load curve with the goal of minimizing the load variance and demand response cost after demand response, based on constraints such as time-of-use pricing, interruptible load operation, wind and solar power output, and user satisfaction. Under the condition of minimizing demand response cost, the load curve with the minimum load fluctuation and peak-valley difference is obtained, and then the load curve is input into the lower-level model.

[0045] The lower-level model takes minimizing system operating cost and carbon emission cost as the objective function. Combining the load curve, it considers the operating constraints of thermal power unit thermal storage retrofit, unit start-up and shutdown constraints, robustness constraints, etc., to optimize the unit carbon emissions, thermal storage system charging and releasing power, thermal storage retrofit unit coal consumption and output, etc. Through optimization solution, the total system operating cost, including demand response cost, carbon emission cost, unit start-up and shutdown cost, and environmental protection cost, is minimized.

[0046] The operation model for thermal power unit thermal storage retrofit is as follows:

[0047] Thermal power unit thermal storage retrofit is based on the principle of thermal cycle, which can yield mathematical models of the heat storage at time t and the heat storage at the previous time, as well as the charging and discharging power at times t and t-1 in high and low temperature circulating thermal storage systems.

[0048]

[0049] In the formula, The stored heat of the high-temperature and low-temperature thermal storage systems at time t; The stored heat of the high-temperature and low-temperature systems at time t-1; μ h μ l These are the heat loss rates of high-temperature and low-temperature thermal storage systems, respectively. For high-temperature and low-temperature thermal storage systems, this refers to the stored and released thermal power; μ hc μ hd μ lc μ ld The thermal efficiency of high and low temperature systems.

[0050] The robust optimization model is as follows:

[0051] To address the uncertainty in wind and solar power output in the lower-level model, a robust optimization coefficient L is introduced, and an adjustable robust optimization scheduling model is established as follows:

[0052]

[0053] In the formula: P w,t With P pv,t The wind power output and solar power output during time period t are respectively, P wp,t θ is the power output error coefficient for wind and solar power. t As an auxiliary variable, its value is greater than or equal to the total wind power output, Xt This is the system net load (total demand load minus unit output).

[0054] Furthermore, the specific steps for establishing the model objective function and constraints in step four are as follows:

[0055] The objective function of the upper-level model is to minimize the load variance of the power grid and the cost of responding to multiple types of demand, i.e.:

[0056]

[0057] In the formula: f1 represents the variance of the power grid load. For the load of the power grid in time period t after demand response, This represents the average load of the power grid after demand response.

[0058] min f2=C Itou +C IL

[0059] In the formula: f2 represents the demand response cost.

[0060] By using the linear weighting method, multiple objectives are transformed into a single objective solution, thus obtaining the objective function that simultaneously considers the net load variance and demand response cost after considering multiple types of demand response:

[0061]

[0062] In the formula: and These represent the maximum and minimum values ​​of the load variance, respectively. and These represent the maximum and minimum values ​​of demand response cost, respectively. R1 and R2 represent the weighting factors corresponding to load variance and demand response cost, respectively. This invention considers reducing load curve fluctuations and reducing demand response costs to be equally important, therefore, the weighting factors are both set to 0.5.

[0063] The upper-level model needs to satisfy the following constraints: wind and solar power output forecast, upper and lower limits of load after time-of-use pricing optimization, electricity price constraints for various types of loads, satisfaction constraints for improved time-of-use pricing, and constraints for interruptible loads. The specific constraints are as follows:

[0064]

[0065]

[0066]

[0067]

[0068]

[0069] In the formula: The electricity price for the nth type of load during time period t. Electricity price during normal periods for load type n. and These represent the ratios of the highest electricity price during peak hours to the lowest electricity price during off-peak hours. T represents 24 time periods. Let W be the initial electricity consumption of user type n during time period t. n This represents the grid concession cost coefficient for the nth type of user. The initial electricity price for user type n during time period t. This represents the maximum number of interruptions for user n within the time period t. This indicates the maximum load reduction that can be achieved for user type n.

[0070] The objective function of the lower-level model mainly optimizes the system's carbon emissions and economic operating costs. The objective function mainly includes the fuel cost of thermal power units, start-up and shutdown costs, system reserve costs, demand response costs, carbon emission costs, and environmental protection taxes.

[0071]

[0072] In the formula: C coal This indicates the price per unit of coal. This represents the coal consumption of thermal power unit j during time period t. This represents the start-up and shutdown status of unit j during time period t. A value of 1 indicates the running state, and a value of 0 indicates the shutdown state. C represents the start-up and shutdown cost of unit j; carbon C B C P These are the system carbon emission costs, spinning reserve costs, and environmental protection taxes, ω T ω represents the total number of typical daily operating times. con This represents the total number of operating units. Carbon emission cost C carbon :

[0073]

[0074] E c This represents the total carbon trading quota of the system. This represents the output power of unit j during time period t. ε is the initial carbon allowance coefficient per unit of electricity generated by the unit. t,j Let j be the actual carbon emission coefficient per unit of electricity generated by a thermal power unit. Cost per unit of carbon emissions.

[0075] The costs of spinning reserve and environmental protection tax are as follows:

[0076]

[0077]

[0078] Where: β res e represents the spinning reserve cost factor; d e w e pv These are the error coefficients for load, wind power, and photovoltaic output, respectively; P d,t , This represents the load power, wind power absorption, and photovoltaic power absorption during time period t, where H is the tax payable per unit of pollution, and q is the total load power, wind power absorption, and photovoltaic power absorption. s q N This indicates the SO2 and NO produced per unit of coal consumption. x mass; η s η N This indicates that the environmental protection equipment desulfurizes SO2 and NO. x efficiency; J s With J N SO2 and NO x The pollution equivalent number, P coal To provide power to the generator unit.

[0079] The lower-level model needs to satisfy power balance constraints, generator set operation constraints, unit ramp-up constraints, unit minimum start-up and shutdown time constraints, power system spinning reserve constraints, and thermal storage system constraints. The specific constraints are as follows:

[0080]

[0081]

[0082]

[0083]

[0084]

[0085]

[0086] In the formula: P w,t With P pv,t These represent wind power output and photovoltaic power output during time period t, respectively. This represents the state variable of the thermal power unit; a value of 1 indicates the unit is powered on, and a value of 0 indicates it is powered off. These represent the upper and lower limits of the output of unit j, respectively. and These represent the maximum and minimum climbing speeds of unit j at its output, respectively. on With T off These represent the maximum start-up and maximum shutdown durations of unit j, respectively. For the maximum and minimum output of unit j; These represent positive and negative rotating reserve capacities, respectively. The positive and negative rotation reserve capacity indicates the actual reserve capacity of the system during positive and negative rotation. These represent the maximum and minimum thermal storage capacities of the high and low temperature thermal storage system during time period t. t represents the thermal storage capacity of the high and low temperature thermal storage system at time t.

[0087] Compared with the prior art, the beneficial effects of the present invention are:

[0088] 1) Compared with the traditional demand response model, the improved multi-type demand response model established in this invention can fully optimize the load curve, effectively reduce the demand response cost, improve the wind and solar absorption rate during off-peak periods, and effectively reduce the total carbon emissions during the scheduling cycle.

[0089] 2) By combining demand response with thermal energy storage retrofit technology, carbon emissions during system scheduling can be further reduced, the unit carbon emissions of thermal power units can be reduced, and the shortcomings of demand response in reducing carbon emissions can only be made up for during off-peak periods.

[0090] 3) By establishing a robust optimization model that considers the uncertainty of wind power and discussing the impact of the robust optimization coefficient on the model of the present invention, the effectiveness of the model of the present invention is verified. Attached Figure Description

[0091] Figure 1 This is the hierarchical optimization scheduling model of the present invention;

[0092] Figure 2 A schematic diagram of energy flow during the thermal storage retrofit of a thermal power unit;

[0093] Figure 3 A diagram illustrating the low-carbon economic principle of coordinating multi-type demand response with thermal storage retrofitting.

[0094] Figure 4 The unit output status for Scenario 1 at different times;

[0095] Figure 5 The unit output status for each time period in Scenario 4;

[0096] Figure 6 For comparison of load curves before and after demand response;

[0097] Figure 7 The process of charging and releasing heat in a thermal storage system;

[0098] Figure 8 The system operating cost and wind-solar absorption rate under different wind fluctuations;

[0099] Figure 9 For wind and solar forecasts and load curves;

[0100] Figure 10 The solution process for the hierarchical scheduling model. Detailed Implementation

[0101] The invention is further illustrated below with reference to embodiments. To better illustrate the invention, the proposed mathematical model is verified using Matlab numerical simulation, and the results are as follows. Figure 4-8 As shown. The specific steps are as follows:

[0102] Step 1: Based on the differences in price response and electricity consumption characteristics of different user types, loads are divided into sensitive, relatively sensitive, and insensitive loads. Improved time-of-use pricing response, interruptible incentive response, and non-participation in response processing are then applied to establish response models for each type of load. Specific steps are as follows:

[0103] The time-of-use pricing mechanism based on a consumer psychology model divides a scheduling cycle into three time periods: peak, flat, and valley. By setting electricity prices for these three time periods, it guides users to change their electricity consumption times, thereby optimizing the load curve. The calculation model for the load transfer rate during peak and valley periods and the peak-valley price difference is as follows:

[0104]

[0105] In the formula Represents the load transfer rate of type n. For the peak-valley electricity price difference, R n,t Let a be the slope of the linear region. n,t C represents the minimum electricity price difference during load transfer. n,t This represents the maximum load transfer rate between peak and valley periods.

[0106] The optimized load calculation model in the time-of-use electricity price response model, which optimizes and improves the electricity prices for peak, flat, and valley periods, is as follows:

[0107]

[0108] In the formula For user n in time period t after the time-of-use pricing response, and They represent Maximum and minimum values ​​during time period t. Peak, flat, and trough periods. and The calculation formula is as follows:

[0109]

[0110]

[0111]

[0112] In the formula The following are the formulas for calculating the time-of-use electricity price response cost, representing peak-to-flat, peak-to-valley, and flat-to-valley load transfer rates, respectively:

[0113]

[0114] In the formula: P n,t and These represent the load of user n in time period t before and after the time-of-use pricing response; w n,t and C represents the electricity price for user n during time period t before and after the time-of-use pricing response. Itou For time-of-use pricing response costs, N Itou The number of users participating in the time-of-use pricing.

[0115] This invention aggregates dispersed demand response resources through a "load aggregator" (LA), making demand response implementation for interruptible loads more flexible. The load timing expression after interruptible load response is as follows:

[0116]

[0117] In the formula: and These represent the loads before and after the interruptible load excitation time period t; v n,t v represents the amount of charge reduction by user n during time period t. n,t A value of 0 or 1 indicates the normal state and the interrupted state, respectively.

[0118] Let n be the load that can be reduced for user n during time period t; n is the load aggregator that can provide interruptible load, and there are a total of N. IL One interruptible user aggregator. N IL =1, 2, 3, 4 represent the interruptible aggregators for industry, commerce, residential, and agriculture, respectively. During the scheduling process, the scheduling cost of interruptible demand response can be expressed as follows:

[0119]

[0120] In the formula, K t The unit power reduction compensation cost is the interruptible load capacity corresponding to the interruptible load capacity during time period t.

[0121] Step 2, based on the response model in Step 1, analyzes the operational mechanism of multiple types of demand response and the low-carbon economic operation of thermal storage retrofit, as well as the low-carbon economic operation mechanism of the combination of multiple types of demand response and thermal storage retrofit. The specific steps are as follows:

[0122] Price-based demand response (PDR) shifts reduced peak loads to off-peak periods through electricity pricing mechanisms, increasing wind power absorption during off-peak hours. Reduced peak loads also help decrease the frequency of thermal power unit start-ups and shutdowns. The low-carbon economic mechanism of PDR is equivalent to replacing expensive high-carbon unit output during peak periods with economical zero-carbon wind power or economical unit output.

[0123] IDR (Incentive-Driven Demand Response) can provide the system with spinning reserves by increasing or decreasing load shedding, effectively reducing the reserve capacity undertaken by thermal power units. IDR can also reduce the output of thermal power units by cutting peak loads, which is equivalent to replacing the output of high-carbon emission units with zero-carbon virtual power sources.

[0124] Although demand response improves wind power absorption and reduces the number of unit start-ups and shutdowns by "peak shaving and valley filling" of loads, thus lowering system economic costs, its system carbon reduction effect is still not ideal. This is mainly because the output of high-carbon emission units reduced by demand response is replaced by the original wind curtailment and low-carbon emission units during off-peak periods. However, when the amount of wind curtailment during off-peak periods is not high, high-carbon emission units still provide power. When wind power absorption increases to a certain level, the net load fluctuation may increase, thus increasing the deep peak shaving period of thermal power units and increasing carbon emissions.

[0125] The low-carbon economic mechanism of thermal energy storage retrofit is as follows:

[0126] Energy transfer within a coal-fired power plant mainly involves three processes: 1) The boiler burns coal to generate heat, which heats water to produce high-temperature, high-pressure steam that flows into the high-pressure cylinder of the steam turbine to perform work; 2) The steam in the high-pressure cylinder flows into the boiler to form hot steam, which is then introduced into the low-pressure and intermediate-pressure cylinders of the turbine to perform work; 3) The condenser tower condenses the gas from the low-pressure cylinder into liquid water, which is then heated by a portion of the steam extracted from the cylinder to become high-temperature feedwater, and this process is repeated. The tight thermal coupling of the boiler, steam turbine, generator, and other components in a traditional coal-fired unit, along with boiler combustion stability conditions, limits the conventional peak-shaving depth of the unit. Traditional peak-shaving methods for coal-fired power units are mainly divided into "conventional peak-shaving," "peak-shaving without oil injection," and "peak-shaving with oil injection." As the peak-shaving depth increases, the unit's energy consumption per unit of electricity generated and carbon emissions increase significantly.

[0127] Thermal energy storage retrofit of power units involves adding a high- and low-temperature dual-cycle thermal energy storage device to the steam power cycle to store thermal energy. Under the condition of stable boiler combustion, this decouples the thermal energy between different parts of the unit, enabling a more flexible output modification for thermal power units. By increasing the thermal energy storage system to store heat, the output of thermal power units is limited to the conventional peak-shaving range. This is effectively equivalent to converting the output of high-carbon-emission units during periods of deep peak shaving into the output of low-carbon-emission units during conventional peak shaving, reducing the number of deep peak shaving operations and thus reducing the unit's carbon emissions per unit, while also reducing the cost of deep peak shaving. However, when the load peak-valley difference is large, due to the limitations of the thermal energy storage system capacity and the system's spinning reserve capacity, the output of high-carbon-emission units during peak hours still leads to an increase in system carbon emissions.

[0128] To verify the effectiveness of the proposed model, this paper compares and analyzes the low-carbon economic operation mechanism of multi-type demand response and thermal storage retrofitting in the following four scenarios:

[0129] 1) Scenario 1, without considering demand response and thermal power unit thermal storage retrofit;

[0130] 2) Scenario 2, considering the ordinary demand response type (TOU+IL), consider the thermal power unit heat storage retrofit;

[0131] 3) Scenario 3, considers the improvement model of multi-type demand response, but does not consider the thermal power unit thermal storage retrofit;

[0132] 4) Scenario 4, consider the improvement model of multi-type demand response, and consider the thermal storage retrofit of thermal power units (the model in this paper).

[0133] Figure 3 To consider the low-carbon economic principles of system operation when considering both scenarios, when the system only considers multiple types of demand response, it is equivalent to replacing the output of Scenario 4 during peak load periods with the output of Scenario 1. If the amount of wind curtailment is small, the proportion of units operating in Scenario 1 decreases, and the output of high-carbon emission units increases. After introducing thermal storage unit retrofits, the increased unit output through the thermal storage system is converted to Scenario 2 output, which is equivalent to converting the output of high-carbon emission units in deep peak shaving into the output of conventional low-carbon emission units. At the same time, it avoids the problem of increased carbon emissions from thermal power units during deep peak shaving periods caused by the increased fluctuation of the net load curve after considering demand response. When only considering the system operation mode after the unit thermal storage retrofit, the large peak-valley difference means that high-carbon emission units still operate during peak periods, and the system is still insufficient in terms of low-carbon economy.

[0134] Step 3: Establish a low-carbon economic dispatch model and a robust optimization model for the power system that considers multiple types of demand response and thermal storage retrofitting. The specific steps are as follows:

[0135] A low-carbon economic dispatch model for the power system, considering multiple types of demand response and thermal storage retrofitting, is established. The process of the established hierarchical optimization dispatch model is as follows: Figure 1 As shown.

[0136] The upper-level model optimizes the load curve with the goal of minimizing the load variance and demand response cost after demand response, based on constraints such as time-of-use pricing, interruptible load operation, wind and solar power output, and user satisfaction. Under the condition of minimizing demand response cost, the load curve with the minimum load fluctuation and peak-valley difference is obtained, and then the load curve is input into the lower-level model.

[0137] The lower-level model takes minimizing system operating cost and carbon emission cost as the objective function. Combining the load curve, it considers the operating constraints of thermal power unit thermal storage retrofit, unit start-up and shutdown constraints, robustness constraints, etc., to optimize the unit carbon emissions, thermal storage system charging and releasing power, thermal storage retrofit unit coal consumption and output, etc. Through optimization solution, the total system operating cost, including demand response cost, carbon emission cost, unit start-up and shutdown cost, and environmental protection cost, is minimized.

[0138] The operation model for thermal power unit thermal storage retrofit is as follows:

[0139] Thermal power unit thermal storage retrofit is based on the principle of thermal cycle, combined with Figure 2 From the energy diagram, we can obtain mathematical models of the heat storage at time t and the heat storage at the previous time, as well as the heat charge and release power at times t and t-1 in the high and low temperature circulating thermal storage system.

[0140]

[0141] In the formula, The stored heat of the high-temperature and low-temperature thermal storage systems at time t; The stored heat of the high-temperature and low-temperature systems at time t-1; μ h μ l These are the heat loss rates of high-temperature and low-temperature thermal storage systems, respectively. For high-temperature and low-temperature thermal storage systems, this refers to the stored and released thermal power; μ hc μ hd μ lc μ lc The thermal efficiency of high and low temperature systems.

[0142] The robust optimization model is as follows:

[0143] To address the uncertainty in wind and solar power output in the lower-level model, a robust optimization coefficient L is introduced, and an adjustable robust optimization scheduling model is established as follows:

[0144]

[0145] In the formula: P w,t With P pv,t The wind power output and solar power output during time period t are respectively, Pwp,t θ is the power output error coefficient for wind and solar power. t As an auxiliary variable, its value is greater than or equal to the total wind power output, X t This is the system net load (total demand load minus unit output).

[0146] Step 4: Based on the mathematical model considered in the above steps, set the objective function and constraints. The specific steps are as follows:

[0147] The objective function of the upper-level model is to minimize the load variance of the power grid and the cost of responding to multiple types of demand, i.e.:

[0148]

[0149] In the formula: f1 represents the variance of the power grid load. For the load of the power grid in time period t after demand response, This represents the average load of the power grid after demand response.

[0150] min f2=C Itou +C IL

[0151] In the formula: f2 represents the demand response cost.

[0152] By using the linear weighting method, multiple objectives are transformed into a single objective solution, thus obtaining the objective function that simultaneously considers the net load variance and demand response cost after considering multiple types of demand response:

[0153]

[0154] In the formula: and These represent the maximum and minimum values ​​of the load variance, respectively. and These represent the maximum and minimum values ​​of demand response cost, respectively. R1 and R2 represent the weighting factors corresponding to load variance and demand response cost, respectively. This invention considers reducing load curve fluctuations and reducing demand response costs to be equally important, therefore, the weighting factors are both set to 0.5.

[0155] The upper-level model needs to satisfy the following constraints: wind and solar power output forecast, upper and lower limits of load after time-of-use pricing optimization, electricity price constraints for various types of loads, satisfaction constraints for improved time-of-use pricing, and constraints for interruptible loads. The specific constraints are as follows:

[0156]

[0157]

[0158]

[0159]

[0160]

[0161] In the formula: The electricity price for the nth type of load during time period t. Electricity price during normal periods for load type n. and These represent the ratios of the highest electricity price during peak hours to the lowest electricity price during off-peak hours. T represents 24 time periods. This represents the initial electricity consumption of user type n during time period t. W n This represents the grid concession cost coefficient for the nth type of user. This represents the initial electricity price for user type n during time period t. This represents the maximum number of interruptions for user n within the time period t. This indicates the maximum load reduction that can be achieved for user type n.

[0162] The objective function of the lower-level model mainly optimizes the system's carbon emissions and economic operating costs. The objective function mainly includes the fuel cost of thermal power units, start-up and shutdown costs, system reserve costs, demand response costs, carbon emission costs, and environmental protection taxes.

[0163]

[0164] In the formula: C coal This indicates the price per unit of coal. This represents the coal consumption of thermal power unit j during time period t. This represents the start-up and shutdown status of unit j during time period t. A value of 1 indicates the running state, and a value of 0 indicates the shutdown state. C represents the start-up and shutdown cost of unit j; carbon C B C P These are the system's carbon emission costs, spinning reserve costs, and environmental protection taxes, respectively; ω T ω represents the total number of typical daily operating times. con This represents the total number of operating units. Carbon emission cost C carbon :

[0165]

[0166] In the formula: E c This represents the total carbon trading quota of the system. This represents the output power of unit j during time period t. ε is the initial carbon allowance coefficient per unit of electricity generated by the unit. t,j Let j be the actual carbon emission coefficient per unit of electricity generated by a thermal power unit. Cost per unit of carbon emissions.

[0167] The costs of spinning reserve and environmental protection tax are as follows:

[0168]

[0169]

[0170] Where: β res e represents the spinning reserve cost factor; d e w e pv These are the error coefficients for load, wind power, and photovoltaic output, respectively; P d,t , Let t represent the load power, wind power absorption, and photovoltaic power absorption during the time period, H represent the tax payable per unit of pollution, and q represent the total load power, wind power absorption, and photovoltaic power absorption during the time period. s q N This indicates the SO2 and NO produced per unit of coal consumption. x mass; η s η N For environmental protection equipment to desulfurize SO2 and NO x Efficiency; J s With J N SO2 and NO x Pollution equivalent number; P coal To provide power to the generator unit.

[0171] The lower-level model needs to satisfy power balance constraints, generator set operation constraints, unit ramp-up constraints, unit minimum start-up and shutdown time constraints, power system spinning reserve constraints, and thermal storage system constraints. The specific constraints are as follows:

[0172]

[0173]

[0174]

[0175]

[0176]

[0177]

[0178] In the formula: P w,t With P pv,t These represent wind power output and photovoltaic power output during time period t, respectively. This represents the state variable of the thermal power unit; a value of 1 indicates the unit is powered on, and a value of 0 indicates it is powered off. These represent the upper and lower limits of the output of unit j, respectively. and These represent the maximum and minimum climbing speeds of unit j at its output, respectively. on With T offThese represent the maximum start-up and maximum shutdown durations of unit j, respectively. For the maximum and minimum output of unit j; These represent positive and negative rotating reserve capacities, respectively. The positive and negative rotation reserve capacity indicates the actual reserve capacity of the system during positive and negative rotation. These represent the maximum and minimum thermal storage capacities of the high and low temperature thermal storage system during time period t. t represents the thermal storage capacity of the high and low temperature thermal storage system at time t.

[0179] To verify the effectiveness of the proposed model, this invention utilizes an improved 30-unit system in the MATLAB 2021b environment for example analysis. The system comprises six thermal power units retrofitted with thermal storage, one wind farm, and one photovoltaic power station, with total capacities of 2300MW, 1800MW, and 650MW, respectively. The wind and solar power output versus load forecast curves are shown below. Figure 9 As shown.

[0180] The simulation parameters are as follows:

[0181] Upper-level model parameters: Time-of-use electricity pricing coefficient W for various user types n The initial electricity prices for various user groups are 5%, 0%, 5%, and 6%. That is, the electricity price during normal periods and the unit cost of interruption K. t The cost is 127.4 yuan / MW.h, the number of interruptions, and the maximum and minimum interruption times. The maximum proportionality coefficient γ for commercial and residential load interruptions is 4 hours and 2 hours respectively. n The values ​​are 0.25 and 0.15 respectively, and the remaining demand response parameters are as follows:

[0182] User category and percentage parameters:

[0183]

[0184] Improve the upper and lower limits of peak, valley, and normal electricity prices for each user under the time-of-use pricing model:

[0185]

[0186] Improved time-of-use pricing peak-valley load transfer factor:

[0187]

[0188] Improved time-of-use pricing for each user during peak, valley, and normal periods:

[0189]

[0190] Key parameters of the lower-level model:

[0191]

[0192] The hierarchical optimization scheduling model established in this invention consists of two layers. The upper-layer planning model is a mixed-integer nonlinear programming model, which cannot be directly solved using solvers such as Gurobi. Therefore, the upper-layer model is solved using the particle swarm optimization algorithm. The lower-layer model is a mixed-integer linear programming model, which can be directly solved using solvers such as Gurobi. The specific solution process is detailed in [link to solution details]. Figure 10 .

[0193] Comparative analysis of scheduling results:

[0194] To verify the effectiveness of the proposed model, this invention conducts a comparative analysis on the following four scenarios:

[0195] Scenario 1, without considering demand response and thermal power unit thermal storage retrofit;

[0196] Scenario 2, considering the general demand response type (TOU+IL), consider the thermal power unit thermal storage retrofit;

[0197] Scenario 3 considers a multi-type demand response improvement model, but does not consider thermal power unit thermal storage retrofitting;

[0198] Scenario 4 considers the improvement model of multi-type demand response and the thermal power unit thermal storage retrofit (the model of this invention).

[0199] Comparing Scenario 1, Scenario 2, and Scenario 3, Scenario 3's system carbon emissions are 21,028 tons, a reduction of 2,976 tons and 2,523 tons compared to Scenario 1 and Scenario 2, respectively. The operating cost of Scenario 3 is 8.3806 million yuan, a reduction of 703,800 yuan and 604,500 yuan compared to Scenario 1 and Scenario 2, respectively. This is mainly because Scenario 3 considers improved demand response, thereby fully optimizing the load curve and reducing load fluctuations. This allows the system's thermal power peak-shaving units to reduce peak-shaving depth and increase wind and solar power absorption rates, correspondingly reducing the amount of coal burned by thermal power units. Therefore, Scenario 3 reduces both the total system operating cost and total carbon emissions compared to Scenario 1 and Scenario 2.

[0200] Scenario 4, based on Scenario 3, incorporates thermal energy storage retrofitting of thermal power units, further considering the coordinated operation of various demand response mechanisms and the thermal energy storage retrofit. As shown in the table below, the total operating cost of Scenario 4 is 8.1149 million yuan, lower than Scenario 2 and Scenario 3; the carbon emissions of Scenario 4 are reduced by 3275 tons and 752 tons compared to Scenario 2 and Scenario 3, respectively; and the wind and solar curtailment rates are reduced by 4.646% and 6.246% compared to Scenario 2 and Scenario 3, respectively. This indicates that, considering various demand response mechanisms, the thermal energy storage retrofitting technology of thermal power units can further reduce system operating costs and carbon emissions, thus verifying the effectiveness of the model of this invention.

[0201]

[0202] Comparative Analysis of Unit Scheduling

[0203] To analyze the low-carbon economic processes in the scheduling of the improved multi-type demand response model and the thermal energy storage retrofit of generating units, the system scheduling conditions under scenario 1 and scenario 4 were compared. (See...) Figure 4 and Figure 5 As follows, by Figure 4 It can be seen that the total unit output in Scenario 1 is greater than that in Scenario 4, and the unit output is greater in the off-peak period from 1:00 to 5:00 and the peak period from 13:00 to 15:00. The wind and solar power absorption rate is also lower in Scenario 1. This is because Scenario 1 does not consider demand response, which leads to a greater load peak-valley difference than Scenario 4. At the same time, it does not consider the increased peak shaving depth of thermal power units due to the modification of thermal storage units. In this scenario, the carbon emissions and the total system operating cost are very high due to the large output of thermal power units.

[0204] Building upon Scenario 1, Scenario 4 considers not only multiple demand response mechanisms but also thermal power unit thermal storage retrofit technology. Compared to Scenario 1, it optimizes the original load and reduces the depth of peak shaving for thermal power units, thereby lowering unit carbon emissions. Figure 5 It can be seen that demand response shifts a portion of the load from 9:00-20:00 to 1:00-5:00 and 21:00-24:00, improving the wind and solar power absorption rate during these periods and reducing system carbon emissions through wind-solar balancing. During periods such as 9:00-13:00, the thermal power unit's thermal storage system reduces the unit's output by releasing heat absorbed during other periods, further tapping into the power system's low-carbon potential. From Figure 4 and Figure 8 It can also be seen that the wind and solar power output in scenario 4 is significantly higher than that in scenario 1. This is mainly due to the coordinated cooperation between the improved multi-type demand response and the thermal energy storage retrofit of the unit.

[0205] Analysis of the advantages of the optimization model

[0206] In summary, considering the coordinated use of improved multi-type demand response and thermal storage retrofit, the carbon emissions and total operating cost of the system can be reduced. Its main advantage lies in the fact that the multi-type demand response mechanism can fully optimize the load curve, while the thermal storage system can reduce the unit carbon emissions. Figure 6 and Figure 7 These are a comparison chart of loads after optimization of various demand responses and a diagram of the heat storage system's charging and discharging process.

[0207] from Figure 6As can be seen from the table, Scenario 4 uses an improved multi-type demand response mechanism to "smooth out and fill in" the load curve, achieving a slightly higher level of optimization than Scenario 2. However, according to Table 2, the demand response cost efficiency of Scenario 4 is higher than that of Scenario 2. In Scenario 4, the improved time-of-use pricing mechanism transfers the peak and normal load periods of 8:00-20:00 to other loads, and further optimizes the load curve by interrupting commercial and residential loads during periods such as 8:00-11:00, 13:00-16:00, and 18:00-20:00.

[0208] Depend on Figure 7 It is easy to see that during periods such as 0:00-8:00, 14:00, and 21:00, the thermal storage system stores the heat generated by the boiler, reducing the steam intake of the turbine and thus reducing the output of the thermal power unit. This is equivalent to limiting the boiler output of the unit during normal peak shaving and reducing the peak shaving depth of the thermal power unit. However, during periods such as 9:00-13:00 and 16:00, the thermal storage system releases the previously stored heat, increasing the steam intake of the turbine and increasing the output of the unit. This is equivalent to reducing the carbon emissions per unit of unit output.

[0209] Impact analysis of robust optimization coefficient

[0210] Wind power uncertainty is a significant factor affecting system operation. This invention modifies the robust optimization coefficient under scenario 4 to analyze system operating costs and wind-solar integration rates under different wind power fluctuations. The results are as follows: Figure 8 As shown.

[0211] Depend on Figure 8 It can be seen that the greater the fluctuation in wind and solar power output, the slightly higher the total system operating cost, while the lower the wind and solar power absorption rate. Within the total system operating cost, fuel costs, carbon trading costs, and environmental protection taxes increase slightly, while other costs remain unchanged. It can be seen that the output shortfall caused by wind and solar power fluctuations is mainly supplemented by thermal power units, while thermal storage retrofitting can fully utilize the heat from boiler combustion to reduce some of the unit's output, effectively reducing system carbon emissions. However, as the fluctuation in wind and solar power output increases, the system's ability to absorb wind and solar power output also decreases. This is mainly because when wind and solar power output is uncertain, the system chooses more stable and safer thermal power unit output, resulting in a low wind and solar power absorption rate.

[0212] This invention proposes a low-carbon optimized dispatching method for achieving low-carbon economic dispatching of power systems, considering multiple types of demand response and thermal power unit thermal storage retrofitting. Simulation optimization results show that:

[0213] 1) Compared with the traditional demand response model, the improved multi-type demand response model established in this invention can fully optimize the load curve, effectively reduce the demand response cost, improve the wind and solar absorption rate during off-peak periods, and effectively reduce the total carbon emissions during the scheduling cycle.

[0214] 2) By combining demand response with thermal energy storage retrofit technology, carbon emissions during system scheduling can be further reduced, the unit carbon emissions of thermal power units can be reduced, and the shortcomings of demand response in reducing carbon emissions can only be made up for during off-peak periods.

[0215] 3) By establishing a robust optimization model that considers the uncertainty of wind power and discussing the impact of the robust optimization coefficient on the model of the present invention, the effectiveness of the model of the present invention is verified.

[0216] The above specific implementation examples are only for the purpose of helping those skilled in the art to understand the present invention. However, the present invention is not limited to the situations in the examples. For those skilled in the art, as long as the various changes are within the spirit and scope of the present invention as defined and determined by the appended claims, these changes are obvious. All inventions utilizing the concept of the present invention are protected.

Claims

1. A low-carbon economic dispatch method for power systems based on multi-type demand response and thermal storage retrofitting, characterized in that, Includes the following steps: Step 1: Based on the differences in price response and electricity consumption characteristics of different types of users, the load is divided into sensitive, relatively sensitive, and insensitive loads. Improved time-of-use pricing response, interruptible incentive response, and non-participation in response processing are adopted respectively, and response models are established. Step 2: Based on the response model in Step 1, analyze the operation mechanism of multi-type demand response and the low-carbon economic operation of thermal storage retrofit, as well as the low-carbon economic operation mechanism of multi-type demand response combined with thermal storage retrofit. Step 3: Based on the operating mechanism in Step 2, establish a low-carbon economic dispatch model and a robust optimization model for the power system that takes into account multiple types of demand response and thermal storage retrofitting. Step 4: Based on the low-carbon economic dispatch model proposed in Step 3, establish the corresponding objective function and model constraints; In step one, response models for different types of loads are established. The specific load response models are as follows: The time-of-use pricing mechanism based on a consumer psychology model divides a scheduling cycle into three time periods: peak, flat, and valley. By setting electricity prices for these three time periods, it guides users to change their electricity consumption times, thereby optimizing the load curve. The calculation model for the load transfer rate during peak and valley periods and the peak-valley price difference is as follows: ， In the formula express Class of load transfer rate, For the peak-valley electricity price difference, The slope of the linear region. This represents the minimum electricity price difference during load transfer. This represents the maximum peak-valley load transfer rate. The optimized load calculation model in the time-of-use electricity price response model, which optimizes and improves the electricity prices for peak, flat, and valley periods, is as follows: ， In the formula For users responding to time-of-use pricing exist Load during a specific time period and They represent exist The maximum and minimum values ​​for the time period; Peak, flat, and valley periods and The calculation formula is as follows: ， ， ， In the formula , The formula for calculating the response cost of time-of-use electricity pricing, which represents the load peak-to-flat and peak-to-valley shift rates, is as follows: ， In the formula: and Users before and after the time-of-use pricing response exist Load during a specific time period; and Indicates the user's response before and after the time-of-use pricing. During the period Electricity price For time-of-use electricity pricing response costs, The number of users participating in the time-of-use pricing response; By aggregating dispersed demand response resources through load aggregators (LAs), demand response implementation for interruptible loads becomes more flexible; the load timing expression after interruptible load response is as follows: ， In the formula: and Before and after interruptible load excitation Load during a specific time period; Indicates load user During the period The amount of charge reduction, A value of 0 or 1 indicates the normal state and the interrupted state, respectively. For users During the period The load can be reduced; For load aggregators that can provide interruptible loads, there are a total of A user aggregator that can be interrupted; When = 1, 2, 3, and 4, they represent the interruptible aggregators for industry, commerce, residential, and agriculture, respectively. During the scheduling process, the scheduling cost of interruptible demand response can be expressed as follows: ， In the formula, for The unit power reduction compensation fee corresponding to the interruptible load capacity during the time period.

2. The low-carbon economic dispatch method for power systems based on multi-type demand response and thermal storage retrofitting as described in claim 1, characterized in that, The specific steps of step three are as follows: The following is a low-carbon economic dispatch model for the power system that takes into account multiple types of demand response and thermal storage retrofitting: The upper-level model optimizes the load curve with the goal of minimizing the load variance and demand response cost after demand response, based on the constraints of time-of-use pricing and interruptible load operation, wind and solar power output, and user satisfaction. Under the condition of minimizing demand response cost, the load curve with the minimum load fluctuation and peak-valley difference is obtained, and then the load curve is input into the lower-level model. The lower-level model takes minimizing system operating cost and carbon emission cost as the objective function. Combining the load curve, it considers the operating constraints of thermal power unit thermal storage retrofit, unit start-up and shutdown constraints, and robust constraints. It optimizes the unit carbon emissions, thermal storage system heat charging and releasing power, coal consumption and output of the thermal storage retrofitted unit. Through optimization, it minimizes the total system operating cost, including demand response cost, carbon emission cost, unit start-up and shutdown cost, and environmental protection cost. The operation model for thermal power unit thermal storage retrofit is as follows: Thermal power unit thermal storage retrofit is based on the principle of thermal cycle, which can produce high and low temperature circulating thermal storage systems. The amount of heat stored at any given time is the same as the amount of heat stored at the previous time. and Mathematical model of constant charge and discharge heat power: ， In the formula, , for The heat stored in high-temperature and low-temperature thermal storage systems at all times; , for The heat stored in high-temperature and low-temperature systems at all times; , These are the heat loss rates of high-temperature and low-temperature thermal storage systems, respectively. , , , The stored and released heat power of high-temperature and low-temperature thermal storage systems; , , , For the thermal efficiency of high and low temperature systems; The robust optimization model is as follows: For the lower-level model, robust optimization coefficients need to be introduced to consider the uncertainty of wind and solar power output. The following adjustable robust optimization scheduling model is established: ， In the formula: and They are respectively Wind power output and solar power output during different time periods This is the wind and solar power output error coefficient. As an auxiliary variable, its value is greater than or equal to the total wind power output. This represents the net load of the system.

3. The low-carbon economic dispatch method for power systems based on multi-type demand response and thermal storage retrofitting as described in claim 2, characterized in that, The specific steps of step four are as follows: The objective function of the upper-level model is to minimize the load variance of the power grid and the cost of responding to multiple types of demand, i.e.: , In the formula: Indicates the variance of the power grid load. For demand response power grid Time-of-use load, This represents the average load of the power grid after demand response. ， In the formula: Indicates demand response cost; By using the linear weighting method, multiple objectives are transformed into a single objective solution, thus obtaining the objective function that simultaneously considers the net load variance and demand response cost after considering multiple types of demand response: ， In the formula: and These represent the maximum and minimum values ​​of the load variance, respectively. and These represent the maximum and minimum values ​​of the demand response cost, respectively. and These represent the weighting factors corresponding to load variance and demand response cost, respectively. The upper-level model needs to satisfy the following constraints: wind and solar power output forecast, upper and lower limits of load after time-of-use pricing optimization, electricity price constraints for various types of loads, satisfaction constraints for improved time-of-use pricing, and constraints for interruptible loads. The specific constraints are as follows: ， In the formula: For the first Load Electricity price during specific time periods No. Electricity price during normal periods for similar loads and These represent the ratio coefficients of the highest electricity price during peak hours and the lowest electricity price during off-peak hours, respectively. It is represented as 24 time periods. Indicates the first User Initial electricity consumption for the period; Indicates the first User-type power grid concession cost coefficient Indicates the first User Initial electricity price for the period Indicates user exist Maximum value of interruption during time period Represents the type of user Maximum load can be reduced; The objective function of the lower-level model mainly optimizes the system's carbon emissions and economic operating costs. The objective function includes the fuel cost of thermal power units, start-up and shutdown costs, system reserve costs, demand response costs, carbon emission costs, and environmental protection taxes. ， In the formula: This indicates the price per unit of coal. Indicates thermal power unit exist Coal consumption during a given period For the unit exist The start / stop status of a time period is 1, indicating the running state, and 0, indicating the stopped state. Indicates the unit Start-up and shutdown costs; , , These are the system's carbon emission costs, spinning reserve costs, and environmental protection taxes. This represents the total number of operating hours on a typical day. This represents the total number of operating units; Carbon emission costs : ， This represents the total carbon trading quota of the system. Indicates the unit exist Power output during a given time period This is the initial carbon allowance coefficient per unit of electricity generated by the generating unit. For thermal power units Actual carbon emission coefficient per unit of electricity Cost per unit of carbon emissions; The costs of spinning reserve and environmental protection tax are as follows: ， ， In the formula: Indicates the spinning reserve cost factor; , , These are the error coefficients for load, wind power, and photovoltaic output, respectively. , , express Time-of-use load power, wind power absorption, and photovoltaic power absorption. The amount of tax that an entity must pay for pollution. , Indicates the output per unit of coal consumed and quality; , Indicates desulfurization of environmental protection equipment , Efficiency; and for and The pollution equivalent number, To provide power to the generator unit; The lower-level model needs to satisfy power balance constraints, generator set operation constraints, unit ramp-up constraints, unit minimum start-up and shutdown time constraints, power system spinning reserve constraints, and thermal storage system constraints. The specific constraints are as follows: ， ， ， ， ， ， In the formula: and They are respectively Wind power output and solar power output during different time periods; This represents the state variable of the thermal power unit; a value of 1 indicates the unit is powered on, and a value of 0 indicates it is powered off. , The units The upper and lower limits of output; and The units Maximum and minimum climbing speeds when exerting force; and They represent the generating units. Maximum power-on and maximum power-off duration; , For the unit Maximum and minimum output; , respectively; , The positive and negative rotation reserve capacity indicates the actual reserve capacity of the system during positive and negative rotation. , , , They are respectively Maximum and minimum thermal storage capacity of high and low temperature thermal storage systems during specific time periods; , t represents the thermal storage capacity of the high and low temperature thermal storage system at time t.

Images