Source-load interaction incentive decision-making method based on clean heating of heat accumulating type electric boiler

Abstract

The invention belongs to the technical field of source load interaction of clean heating, and particularly provides a source load interaction excitation decision-making method based on heat accumulating type electric boiler clean heating, which comprises the following steps: constructing an interaction excitation framework of a clean heating load aggregator, and modeling house heat demand, heat accumulating type electric heating load and'photovoltaic + heat accumulating type electric heating 'load; on the basis of a photovoltaic and heat storage type electric heating model, load electricity consumption behaviors are divided into a valley electricity heat storage process, a flat electricity tracking process and a peak electricity heat storage and supply process; and providing a heat accumulating type electric boiler cluster internal regulation and control strategy based on a margin index. And establishing a target function by taking the maximum operation income of the clean heating load provider as an optimization target, constraining the interactive excitation architecture of the clean heating load aggregator in the optimization scheduling process, and solving by adopting an improved particle swarm algorithm. The method promotes the user to participate in the source-load interaction and reduces the system operation cost on the basis of considering the stepped demand response of the user on the load side.

Description

Technical Field

[0001] The present invention belongs to the technical field of source-load interaction for clean heating, and in particular relates to a source-load interaction incentive decision method for clean heating based on a heat storage electric boiler. Background Art

[0002] With the continuous advancement of the marketization of power trading, higher requirements are placed on the flexibility and regulation capabilities of the power system. In some areas, the resources on the load side are dispersed and huge in number, and the massive distributed dispatchable resources are subject to communication restrictions. Therefore, clean heating load aggregators with a large amount of renewable energy output on the power supply side and heat storage electric boilers on the load side have emerged. As an important link between the power supply end and the source-load interaction between various flexible loads, clean heating load aggregators can not only directly conduct market transactions with the power grid as an intermediary, improve the efficiency of demand response, save time and cost, but also fully tap the dispatching potential on the load side. There are some existing electricity price incentive policies to encourage users to participate in source-load interaction. However, these policies are insufficient to motivate users, and users are not willing to actively participate in source-load interaction. Consider proposing a clean heating source-load interaction incentive decision method based on heat storage electric boilers to alleviate the contradiction between the rising supply-side costs and the reduction of energy costs on the demand side. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a source-load interaction incentive decision-making method based on clean heating with a heat storage electric boiler, which aims to improve the existing time-of-use electricity price system, consider the uncertainty of wind and solar power on the source side, and promote user participation in source-load interaction and reduce system operating costs based on the user's step-by-step demand response on the load side.

[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is: a source-load interactive incentive decision-making method based on clean heating of a heat storage electric boiler, comprising the following steps: Step 1: Build an interactive incentive framework for clean heating load aggregators, and model house heat demand, thermal storage electric heating load, and “PV + thermal storage electric heating” load in the built framework; Step 2: Based on the "PV + thermal storage electric heating" model constructed in step 1, the load electricity consumption behavior is subdivided into three processes, including valley power thermal storage process, flat power tracking process and peak power thermal storage heating process, and then the three processes are modeled separately; Step 3: Based on the interactive incentive framework of clean heating load aggregators constructed in step 1 and the model built in step 2, a thermal storage electric boiler cluster internal control strategy based on margin indicators is proposed; Step 4: Based on the internal control strategy of the thermal storage electric boiler cluster based on the margin index proposed in step 3, a reasonable peak-valley time-of-use electricity price plan is formulated by using the peak-flat-valley time division strategy, and the objective function is established with the maximum operating profit of the clean heating load merchant as the optimization goal, and the interactive incentive framework of the clean heating load aggregator is constrained in the optimization scheduling process; Step 5: Based on the objective function and constraints established in step 4, the improved particle swarm algorithm with nonlinear variable inertia weight algorithm is used to solve the problem.

[0005] The present invention provides a source-load interaction incentive decision-making method based on clean heating with a heat storage electric boiler, which has the following beneficial effects: 1. The present invention proposes a "photovoltaic + thermal storage electric heating" load model based on source-load interaction of clean heating with thermal storage electric boilers. On the basis of the constructed clean heating load aggregator interactive incentive framework, an internal control strategy for the thermal storage electric boiler cluster based on margin indicators is proposed, aiming to extract indicators that can reflect various controllable resources, and based on this, select the load with the highest dispatchable potential for priority control, providing ideas for the coordinated control among multiple types of flexible load resources.

[0006] 2. In step 1 of the present invention, the constructed interactive incentive framework for clean heating load aggregators enables clean heating load aggregators to not only act as intermediaries to directly conduct market transactions with the power grid, thereby improving the efficiency of demand response, saving time and costs, but also fully tapping the scheduling potential on the load side.

[0007] 3. In step 2 of the present invention, the load power consumption behavior is subdivided into three processes, and the three processes are modeled and analyzed respectively, so that the user's power demand response in the "photovoltaic + thermal storage electric heating" mode is clearer, providing a more accurate and realistic basis for formulating decision-making plans.

[0008] 4. In step three of the present invention, a thermal storage electric boiler cluster internal control strategy based on margin indicators is proposed. This strategy can be used to extract indicators that can reflect various types of controllable resources, and based on this, the loads with the highest dispatchable potential can be screened out for priority control, providing ideas for coordinated control among multiple types of flexible load resources.

[0009] 5. In step 4 of the present invention, an objective function is established with the maximum operating profit of clean heating load merchants as the optimization goal. The peak-valley time-of-use electricity price scheme formulated by constraining the interactive incentive framework of clean heating load aggregators in the optimization scheduling process has an obvious effect of peak shaving and valley filling, reduces the volatility of loads, and improves the stability of the power system. It can not only effectively encourage electric heating loads to participate in grid peak regulation, but also divide more valley price periods into normal periods, which has huge economic benefits.

[0010] 6. In step 5 of the present invention, an improved particle swarm algorithm of nonlinear variable inertia weight algorithm is proposed for solving, and the solution has good convergence. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The present invention will be further described below in conjunction with the accompanying drawings and embodiments: Figure 1 an interactive incentive framework for clean heating load aggregators; Figure 2 It is a control strategy diagram of thermal storage electric boiler cluster based on margin index; Figure 3 It is a diagram of the dynamic time-of-use electricity price incentive strategy based on load aggregators; Figure 4 This is the standard particle swarm algorithm flow chart; Figure 5 The outdoor temperature at different time periods under four typical weather conditions; Figure 6 The light intensity at different time periods under four typical weather conditions; Figure 7 Storing heat changes for thermal storage clusters; Figure 8 Changes in the operating status of the thermal storage cluster; Fig. 9 The electricity load of a certain county before and after participating in interactive incentives; Fig.10 This is the total electricity load of the other five counties before and after participating in the interactive incentive. DETAILED DESCRIPTION

[0012] Combination Figure 1 to Figure 10 The specific embodiments of the present invention are described in further detail.

[0013] A source-load interactive incentive decision-making method based on clean heating with thermal storage electric boilers includes: constructing an interactive incentive framework for clean heating load aggregators, modeling house heat demand, thermal storage electric heating load, and "photovoltaic + thermal storage electric heating" load in the constructed framework; on the basis of the "photovoltaic + thermal storage electric heating" model, subdividing the load electricity consumption behavior into three processes, including valley electricity thermal storage process, flat electricity tracking process, and peak electricity thermal storage heating process, and modeling the three processes respectively; then, based on the constructed clean heating load aggregator interactive incentive framework, a thermal storage electric boiler cluster internal control strategy based on margin indicators is proposed. Considering improving the existing time-of-use electricity price system to promote user participation in source-load interaction, an objective function is established with the maximum operating profit of clean heating load merchants as the optimization goal, the interactive incentive framework of clean heating load aggregators is constrained in the optimization scheduling process, and an improved particle swarm algorithm is used for solution.

[0014] The specific steps include: Step 1: Build an interactive incentive framework for clean heating load aggregators, and model house heat demand, thermal storage electric heating load, and "photovoltaic + thermal storage electric heating" load in the constructed framework.

[0015] The interactive incentive structure of clean heating load aggregators is as follows: Figure 1 As shown in the figure, the architecture is mainly composed of a coordination layer, a local response layer and a control center; in the constructed architecture, the house heat demand, thermal storage electric heating load and "photovoltaic + thermal storage electric heating" load are modeled.

[0016] 1) Housing heat demand model The relationship between the heat transfer of building temperature and unit time change is shown in formula (1): (1); Where: is the indoor temperature change of the building; The amount of heat output from heating equipment required to maintain a specific indoor temperature in a building; The heat from the sun hitting the building; and They are heat conducted through the building walls and heat exchanged with outdoor air; is the total heat capacity of air.

[0017] In formula (1) , , The calculation method of is shown in formula (2)-(4): (2); (3); (4); Where: , are the heat transfer and heat exchange coefficients of the building respectively; , are the indoor and outdoor temperatures of the building, respectively; is the average daily outdoor wind speed; and are the solar radiation intensity and the lighting area of ​​the building respectively.

[0018] According to the principle of conservation of energy, when the room temperature of the building remains unchanged, we can get: (5); 2) Thermal storage electric heating load model The heat storage characteristics of heat storage equipment can be expressed as the relationship between heat storage, heat storage / release power and heat loss: (6); (7); Where: for t The amount of heat stored in period electric heating; is the thermal storage power during period t; for t Heat release power in each period; is the heat storage loss rate; is the heating power of the thermal storage electric boiler during period t; is the heat loss rate; For the time interval.

[0019] 3) “Photovoltaic + Thermal Storage Electric Heating” Load Model Ignore the thermal inertia of the building and assume t Heating power of period heat storage electric boiler Heating load demand In real-time balance, the following relationship exists: (8); (9); Where: for t Electric power of period heat storage electric boiler; For energy efficiency ratio.

[0020] Step 2: Based on the "photovoltaic + thermal storage electric heating" model constructed in step 1, the load electricity consumption behavior is subdivided into three processes, including valley electricity thermal storage process, flat electricity tracking process and peak electricity thermal storage heating process, and then the three processes are modeled separately.

[0021] Under the "photovoltaic + thermal storage electric heating" application mode, the basic principles of load electricity use are as follows: give priority to using photovoltaic power for heating and energy storage; if the photovoltaic output power cannot meet the user's thermal demand, try to use the off-peak period for energy storage; minimize the use of peak power and use , , They represent the valley period, normal period and peak period of daily heating load respectively. Based on this, the load electricity consumption behavior of "photovoltaic + thermal storage electric heating" is described as follows: 1) Valley power heat storage process: photovoltaic heat storage is used first and heat is provided in real time. The power consumption characteristic model of this process is: (10); Where: for t Photovoltaic power generation during the period; for t Thermal storage power in each period; for t Heat storage during the period.

[0022] 2) Flat power tracking process: Electric heating gives priority to using photovoltaic power generation for electric heating. If the heat storage body is not fully charged during the off-peak period, it can be used to further store energy. The model is: (11); 3) Peak electricity heat storage heating process: Electric heating gives priority to using heat storage equipment for heating during this period. When the energy storage body cannot meet the heating demand, photovoltaic power is used for electric heating. The model is: (12); Step 3: Based on the interactive incentive framework of clean heating load aggregators constructed in step 1 and the model built in step 2, a thermal storage electric boiler cluster internal control strategy based on margin indicators is proposed, aiming to extract indicators that can reflect various controllable resources, and based on this, select the loads with the highest dispatchable potential for priority control, providing ideas for the coordinated control among multiple types of flexible load resources.

[0023] The specific steps are as follows: 1) Through the thermal storage electric boiler terminal, real-time data such as the operating status, heat load demand, and thermal energy storage of the thermal storage electric boiler are collected.

[0024] 2) Define margin indicators based on heat load demand and thermal energy reserves of thermal storage electric boilers, and mainly set up energy margin to increase cluster control efficiency and power margin , the larger the margin index is, the greater the controllability of the load is, and the load should be prioritized for response. The energy margin is the margin of the remaining capacity of the heat storage tank at time t, as expressed in formula (13). The value is a percentage, with a maximum of 100% and a minimum of 0%. The power margin is expressed in formula (14), with a maximum of 100% and a minimum of 0%.

[0025] (13); Where: express t The capacity of the heat storage tank that has stored heat at any given moment.

[0026] (14); Where: express t The operating power of the thermal storage electric boiler at all times; Indicates the rated power of the thermal storage electric boiler.

[0027] Combining the energy margin index and the power margin index can get the comprehensive margin index : (15); Where: α + β =1, α and β Represents the weight coefficient.

[0028] 3) According to the state of the heat storage equipment, the possible states at each moment can be divided into idle group, heat charging group and heat releasing group, which are represented by a set, as shown in formula (16): (16); Where: , , They are the equipment groups in the heat storage group, idle group and heat release group at time t respectively; n 1 、n 2 、n 3 are the number of devices included in each group, and the total number of thermal storage devices is n = n 1+ n 2+ n 3.

[0029] 4) Based on the real-time heat load situation and margin index, multiple thermal storage electric boilers are intelligently regulated by load aggregators, and cluster coordinated control of thermal storage equipment is implemented to meet the overall heat load requirements. The operating power of the entire equipment group at time t is , expressed by formula (17) (17); Where: , Respectively t Moment i The operating power of a thermal storage electric boiler is , the entire heat storage cluster is in a heat storage state, otherwise it is in a heat release state.

[0030] Figure 2 This is a diagram of the control strategy for the cluster of thermal storage equipment based on the margin index. As can be seen from the figure, the strategy process includes equipment grouping, response group determination, indicator sequence generation and determination of the number of participating response equipment.

[0031] Step 4: Based on the internal control strategy of the thermal storage electric boiler cluster based on margin indicators proposed in step 3, consider improving the existing time-of-use electricity price system to promote users to participate in source-load interaction, and use the peak-flat-valley time division strategy to formulate a reasonable peak-valley time-of-use electricity price plan; establish an objective function with the maximum operating profit of clean heating load merchants as the optimization goal, and constrain the interactive incentive framework of clean heating load aggregators in the optimization scheduling process.

[0032] Establishing a complete load aggregator dynamic time-of-use electricity price incentive strategy is an effective way to promote source-load interaction. At present, many places have a complete time-of-use electricity price system. The electricity price and duration are set by the provincial company and cannot be easily changed. Therefore, the present invention considers reducing the cost of the power grid by changing the peak, valley and flat periods of electricity prices. Traditional user-side peak shaving takes the user load curve as the research object. After considering the source-load interaction, the research object of user-side peak shaving needs to consider both the output of new energy and the load characteristics. Optimize the equivalent load curve by transferring user load. Therefore, it is proposed to use the peak-flat-valley period division strategy to formulate a reasonable peak-valley time-of-use electricity price scheme, such as Figure 3 As shown, the specific steps are as follows: 1) Based on historical data, the outdoor temperature at each time period of a typical winter day, the indoor heating temperature of electric heating users, and the original grid electricity price are input as initial data to obtain the heat load demand curve of electric heating users at each time period; 2) Initialize the electric heating load and equipment parameters of each household in the aggregator, calculate the power consumption of various electric heating loads in each period within the scheduling cycle T according to the electric heating user demand response model and the load characteristics of thermal storage electric heating, and construct the LA original load curve; 3) According to the peak-shaving power of LA in each period and the original power consumption curve of the load, the net load power curve of LA under period T is obtained, the peak and valley period membership of the user load is calculated, and the peak, valley and flat power periods are divided according to the calculation results and actual requirements.

[0033] Based on the internal control strategy of the thermal storage electric boiler cluster based on the margin index proposed in step 3, the existing time-of-use electricity price system is considered to be improved to promote users to participate in source-load interaction. The objective function is established with the maximum operating profit of the clean heating load merchant as the optimization goal. The interactive incentive framework of the clean heating load aggregator is constrained in the optimization scheduling process. The objective function and constraint conditions are: (1) Objective function The purpose of establishing a time-of-use electricity price user incentive model based on load aggregators is to guide users to participate in clean heating source-load interaction by formulating reasonable dynamic time-of-use electricity prices. The objective function of the model is to maximize the operating benefits of clean heating load merchants.

[0034] (18); Where: Peaking revenue for clean heating load aggregators; heating revenue for clean heating load aggregators; Aggregate electricity purchase costs for clean heating loads.

[0035] (19); Where: , They are clean heating load aggregators in t The peak-shaving price and peak-shaving capacity when participating in peak-shaving auxiliary services at all times.

[0036] (20); Where: , , They are time-of-use peak, average and valley electricity prices respectively; , , They are the number of peak, flat and valley periods in a day; Expressed as t The power consumption of users' electric heating load during the period.

[0037] (twenty one); Where: , They are t The day-ahead spot market electricity price during the period and the power purchased from the power grid by clean heating load aggregators.

[0038] (2) Constraints The house heat balance constraint is shown in equation (5); the power constraint is shown in equations (22)-(23); the heat demand constraint of electric heating users is shown in equation (24); the time-of-use electricity price constraint is shown in equation (25); and the peak load constraint is shown in equation (26): (twenty two); (twenty three); (twenty four); Where: for t Indoor temperature of users during the time period; and They are the indoor temperatures under the highest and lowest heat demands acceptable to users respectively.

[0039] (25); Where: is the constant of the peak-valley electricity price ratio. According to the notice of the National Development and Reform Commission, is 5; The maximum allowed electricity price.

[0040] (26); Where: Maximum control capacity for clean heating load aggregators.

[0041] Step 5: Based on the objective function and constraints established in step 4, the improved particle swarm algorithm with nonlinear variable inertia weight algorithm is used to solve the problem.

[0042] The expressions of the standard particle swarm algorithm are shown in equations (27)-(28). The algorithm flow Figure 4 As shown in the figure, the advantage of the standard particle swarm algorithm is that it has a good convergence speed, but it is easy to fall into the local optimal solution. Therefore, the improved particle swarm algorithm considering the nonlinear variation inertia weight algorithm is used to solve the problem.

[0043] (27) (28) Where: Indicated in k Moment i The velocity of a particle in dimension d, is the inertia weight, , is the learning factor, , is a random number in the range (0,1). represents individual extreme values, Represents the global extreme value, i=1,2,…x; d=1,2,…D.

[0044] From formula (27), we can see that Indicates the speed of the particle in the last cycle, which is very important for the initial movement of the particle. When the value is too small, the particle swarm optimization algorithm has poor global optimization ability and good local optimization ability; When the value is too large, the particle swarm optimization algorithm has a strong global optimization ability, but a poor local optimization ability. Therefore, two variable operators Y and U are proposed to respectively Optimization is performed to enhance the algorithm's early global search capability and later local search capability. The algorithms for the operator and inertia weight are shown in Equations (29), (30), and (31).

[0045] (29); (30); (31); Where: is the current iteration number, is the maximum number of iterations, and the Y and U operators can be adaptively adjusted through the learning factor. =0, ≠0, the particle swarm converges quickly and is prone to premature convergence; ≠0, = 0, particles focus on their own information, and there is less communication between particles, which greatly reduces the convergence speed and makes particles drift within a certain range. In order to avoid falling into the local optimum and obtain the global optimum, it is necessary to and The value of is dynamically adjusted using the formulas shown in equations (32) and (33). and Perform adaptive optimization.

[0046] (32); (33); Where: = =1 The steps for solving the problem using the improved algorithm are: (1) Input the outdoor temperature at different time periods on typical winter days, indoor heating temperature of clean heating users, electricity price and other data as initial data; (2) Initialize system parameters and algorithm parameters, and assign random initial values ​​to each particle; (3) The particle iterates, updates the position, speed, learning factor, and inertia weight after iteration, compares them with the fitness of the particle in the previous cycle, and memorizes and updates the individual extreme value; (4) Save the optimal individual extreme value as the global extreme value and update and save the individual extreme value; (5) The iteration ends when the global state is stable for five or more times in a row or the maximum number of iterations is reached.

[0047] Example: The present invention uses actual data from a region in northwest my country as an example for example analysis. The region contains large-scale thermal storage electric boilers, which can form a clean heating load aggregator organizational structure and a thermal storage electric boiler cluster. It is known that the current local time-of-use electricity price policy is that the valley electricity price is 0.16 yuan / kWh, from 9 am to 5 pm; the peak electricity price is 0.54 yuan / kWh, from 7 pm to 1 am; the flat electricity price is 0.35 yuan / kWh, and the implementation time is outside the valley and peak periods. The outdoor temperatures in four typical winter weather conditions in the region are as follows: Figure 5 As shown, the light intensity is Figure 6 shown.

[0048] In order to verify the effectiveness of the above-mentioned clean heating thermal storage electric boiler cluster control strategy, a corresponding mathematical model is established, and the internal control results of the thermal storage cluster are as follows: Figure 7 and Figure 8The changes in the stored energy and the operating status of some heat loads are given respectively. The stored energy of heat load refers to the storage of heat energy in the load of the heat storage electric boiler, which can also be called the energy reserve of the heat load. The operating status of the heat storage electric boiler at any time refers to the specific working state and operating status of the heat storage electric boiler, such as heat storage, heat release and idle state.

[0049] Depend on Figure 8 As shown in the figure, the operating state of the heat storage equipment has changed significantly after cluster regulation, reducing the overlapping operating time. The heat storage device is concentrated on storing heat during the low load period. For heat storage devices with a relatively large initial storage energy state, they will be quickly filled during this period and enter the idle group in advance, and then enter the heat release group in advance; the heat storage group is concentrated on releasing heat during the peak load period, and its overall heat storage and release time period is consistent with the overall power adjustment of the load aggregator. Figure 7 It can be seen that the energy stored in the heat storage equipment has changed significantly after cluster regulation. Although some equipment is in the same charging or discharging state, the heat storage rate and the time of heat storage are different. This shows that some heat storage electric boilers have a small adjustable amount or the heat storage equipment is facing the danger of overcharging or over-discharging. The cluster internal regulation strategy can give priority to controlling some equipment, which is conducive to the safe and stable operation of the system.

[0050] In order to verify the effectiveness of the dynamic time-of-use electricity price incentive strategy based on load aggregators mentioned above, a corresponding mathematical model is established to compare the electricity load of a county in the region before and after the other five counties participate in the incentive on a typical day. Fig. 9 and Fig.10 .

[0051] By simulating the electricity load data of the above-mentioned county and the other five counties before and after participating in the interactive incentive, and dividing the peak, valley and normal periods according to the membership method, the following results can be obtained:

[0052] According to Table 1, the electricity price of a certain county and the other five counties are divided into time periods. Through calculation, the various benefits can be obtained as shown in Table 2.

[0053]

[0054] from Fig. 9 and Fig.10It can be seen that after the dynamic time-of-use electricity price incentive, there is an obvious effect of peak shaving and valley filling. The lower load peak and higher valley load indicate that users have increased electricity consumption during the cheap electricity price period. At the same time, the smoothness of the two curves is relatively smooth after the electricity price incentive, indicating that the volatility of the load has been successfully reduced and the stability of the power system has been improved. From Table 1, it can be seen that the use of dynamic time-of-use electricity prices by load aggregators can not only effectively encourage electric heating loads to participate in the peak load regulation of the power grid, but also more valley price periods are divided into normal periods. It can be calculated from Table 2 that after the residents of a certain county actively respond to the load aggregator incentive strategy, the daily peak load regulation income can reach 5327.68 yuan, and the all-day electricity sales income can reach 21372.87 yuan; after the residents of the other five counties actively respond to the load aggregator incentive strategy, the daily peak load regulation income can reach 4544.1 yuan, and the all-day electricity sales income can reach 10491.3 yuan, which has huge economic benefits.

[0055] In summary, considering a source-load interaction incentive decision-making method based on clean heating with thermal storage electric boilers can improve the operating efficiency of each subject. The improved particle swarm algorithm with nonlinear variable inertia weight algorithm is used for solution, and the solution has good convergence. The effect verifies that the proposed strategy can promote user participation in source-load interaction and reduce system operating costs.

Claims

1. A source-load interactive incentive decision-making method based on clean heating with a thermal storage electric boiler, characterized in that: The following steps are involved: Step 1: Build an interactive incentive framework for clean heating load aggregators, and model house heat demand, thermal storage electric heating load, and "photovoltaic + thermal storage electric heating" load in the built framework; Step 2: Based on the "PV + thermal storage electric heating" model constructed in step 1, the load electricity consumption behavior is subdivided into three processes, including valley power thermal storage process, flat power tracking process and peak power thermal storage heating process, and then the three processes are modeled separately; Step 3: Based on the interactive incentive framework of clean heating load aggregators constructed in step 1 and the model built in step 2, a thermal storage electric boiler cluster internal control strategy based on margin indicators is proposed; Step 4: Based on the internal control strategy of the thermal storage electric boiler cluster based on the margin index proposed in step 3, a reasonable peak-valley time-of-use electricity price plan is formulated by using the peak-flat-valley time division strategy, and the objective function is established with the maximum operating profit of the clean heating load merchant as the optimization goal, and the interactive incentive framework of the clean heating load aggregator is constrained in the optimization scheduling process; Step 5: Based on the objective function and constraints established in step 4, the improved particle swarm algorithm with nonlinear variable inertia weight algorithm is used to solve the problem.

2. According to claim 1, a source-load interactive incentive decision-making method based on clean heating with a thermal storage electric boiler is characterized in that: In step 1, the interactive incentive architecture of the clean heating load aggregator consists of a coordination layer, a local response layer, and a control center; in the constructed architecture, the house heat demand, the thermal storage electric heating load, and the "photovoltaic + thermal storage electric heating" load are modeled, which are the house heat demand model, the thermal storage electric heating load model, and the "photovoltaic + thermal storage electric heating" load model, respectively, as follows: 1) Housing heat demand model The expression of the relationship between the heat transfer of building temperature and the change per unit time is: (1)； Where: is the indoor temperature change of the building; The amount of heat output from heating equipment required to maintain a specific indoor temperature in a building; The heat from the sun hitting the building; and They are heat conducted through the building walls and heat exchanged with outdoor air; is the total heat capacity of air; in, , , The calculation expression is: (2)； (3)； (4)； Where: , are the heat transfer and heat exchange coefficients of the building respectively; , are the indoor and outdoor temperatures of the building, respectively; is the average daily outdoor wind speed; and are the solar radiation intensity and the lighting area of ​​the building respectively; According to the principle of conservation of energy, when the room temperature of the building remains unchanged, we can get: (5)； 2) Thermal storage electric heating load model The heat storage characteristics of the heat storage equipment are expressed as the relationship between the heat storage, heat storage / release power and heat loss, and the expression is: (6)； (7)； Where: for t The amount of heat stored in period electric heating; for t Thermal storage power in each period; for t Heat release power in each period; is the heat storage loss rate; is the heating power of the thermal storage electric boiler during period t; is the heat loss rate; is the time interval; 3) "Photovoltaic + thermal storage electric heating" load model Ignoring the thermal inertia of the house, assume that the heating power of the thermal storage electric boiler during period t is Heating load demand In real-time balance, the following relationship exists: (8)； (9)； Where: for t Electric power of period heat storage electric boiler; For energy efficiency ratio.

3. According to claim 1, a source-load interactive incentive decision-making method based on clean heating with a thermal storage electric boiler is characterized in that: In step 2, under the "photovoltaic + thermal storage electric heating" application mode, the basic principles of load electricity use are as follows: give priority to using photovoltaic power for heating and energy storage. If the photovoltaic output power cannot meet the user's thermal demand, try to use the valley power period for energy storage; minimize the use of peak power, and use , , They represent the valley period, normal period and peak period of daily heating load respectively. The load electricity consumption behavior of "photovoltaic + thermal storage electric heating" includes: 1) Valley power heat storage process: photovoltaic heat storage is used first and heat is provided in real time. The power consumption characteristic model of this process is: (10)； Where: for t Photovoltaic power generation during the period; for t Thermal storage power in each period; for t Heat storage during the period, It is the maximum value of heat storage during valley power thermal storage; 2) Flat power tracking process: Electric heating gives priority to using photovoltaic power generation for electric heating. If the heat storage body is not fully charged during the off-peak period, it can be used to further store energy. The model is: (11)； 3) Peak electricity heat storage heating process: Electric heating gives priority to using heat storage equipment for heating during this period. When the energy storage body cannot meet the heating demand, photovoltaic power is used for electric heating. The model is: (12)。 4. According to claim 1, a source-load interactive incentive decision-making method based on clean heating with a thermal storage electric boiler is characterized by: In step 3, the internal control strategy of the thermal storage electric boiler cluster based on the margin index includes the following steps: 1) Through the thermal storage electric boiler terminal, real-time data such as the operating status, heat load demand, and thermal energy storage of the thermal storage electric boiler are collected; 2) Define margin indicators and energy margin according to heat load demand and thermal energy reserves of thermal storage electric boilers and power margin ,The larger the margin index is, the greater the load controllability is, and the response should be prioritized; The energy margin is the margin of the remaining capacity of the heat storage tank at time t. The value is a percentage, with a maximum of 100% and a minimum of 0%. The expression is: (13)； Where: express t The capacity of the heat storage tank that has been stored at any moment; The maximum power margin is 100% and the minimum is 0%. The expression is: (14)； Where: express t The operating power of the thermal storage electric boiler at all times; Indicates the rated power of the thermal storage electric boiler; Combining the energy margin index and the power margin index can get the comprehensive margin index : (15)； Where: α and β represents the weight coefficient, α + β =1; 3) The heat storage equipment is divided into groups according to its status. The possible status at each moment is divided into an idle group, a heat charging group and a heat releasing group, which are represented by a set. The expression is: (16)； Where: , , They are t Equipment groups that are always in heat storage group, idle group and heat release group; It is the heat storage equipment in the heat storage group. For the heat storage equipment in the idle group, It is the heat storage device in the heat release group. n 1 、n 2 、n 3 are the number of devices included in each group, and the total number of thermal storage devices is n = n 1+ n 2+ n 3; 4) Based on the real-time heat load situation and margin index, multiple thermal storage electric boilers are intelligently regulated by load aggregators, and cluster coordinated control of thermal storage equipment is implemented to meet the overall heat load requirements. t The operating power at all times is , the expression is: (17)； Where: , Respectively t Heat storage group and heat release group at all times i The operating power of a thermal storage electric boiler is , the entire heat storage cluster is in a heat storage state, otherwise it is in a heat release state.

5. According to claim 1, a source-load interaction incentive decision-making method based on clean heating with a thermal storage electric boiler is characterized in that: In step 4, a reasonable peak-valley time-of-use electricity price scheme is formulated by using the peak-flat-valley time division strategy, including the following steps: 1) Based on historical data, the outdoor temperature at each time period of a typical winter day, the indoor heating temperature of electric heating users, and the original grid electricity price are input as initial data to obtain the heat load demand curve of electric heating users at each time period; 2) Initialize the electric heating load and equipment parameters of each household in the aggregator, calculate the power consumption of various electric heating loads in each period within the scheduling cycle T according to the electric heating user demand response model and the load characteristics of thermal storage electric heating, and construct the LA original load curve; 3) According to the peak-shaving power of LA in each period and the original power consumption curve of the load, the net load power curve of LA under period T is obtained, the peak and valley period membership of the user load is calculated, and the peak, valley and flat power periods are divided according to the calculation results and actual requirements.

6. According to claim 1, a source-load interactive incentive decision-making method based on clean heating with a thermal storage electric boiler is characterized in that: In step 4, the objective function and constraints are: 1) Objective function The objective function of the model is to maximize the operating profit of the clean heating load provider, and the expression is: (18)； Where: Peaking revenue for clean heating load aggregators; heating revenue for clean heating load aggregators; Aggregate electricity purchase costs for clean heating loads; (19)； Where: , They are clean heating load aggregators in t The peak-shaving price and peak-shaving capacity when participating in the peak-shaving auxiliary service at all times; (20)； Where: , , They are time-of-use peak, average and valley electricity prices respectively; , , They are the number of peak, flat and valley periods in a day; Expressed as t Electricity power consumption of electric heating load of users during the period; (21)； Where: , They are t The day-ahead spot market electricity price and the power purchased from the grid by clean heating load aggregators; 2) Constraints The house heat balance constraint is shown in equation (5); the power constraint is shown in equations (22)-(23); the heat demand constraint of electric heating users is shown in equation (24); the time-of-use electricity price constraint is shown in equation (25); and the peak load constraint is shown in equation (26): (22)； (23)； (24)； Where: for t Indoor temperature of users during the time period, It is the heat storage equipment in the heat storage group. It is the heat storage device in the heat release group. is the indoor temperature of the building at time t; and They are the indoor temperatures under the highest and lowest heat demands acceptable to users respectively; (25)； Where: is the constant of the peak-to-valley price ratio, is the maximum allowed electricity price; (26) Where: Maximum control capacity for clean heating load aggregators.

7. According to claim 1, a source-load interactive incentive decision-making method based on clean heating with a thermal storage electric boiler is characterized in that: In step 5, an improved particle swarm algorithm based on a nonlinear variable inertia weight algorithm is used for solving the problem; the expression of the standard particle swarm algorithm is shown in equations (27)-(28): (27)； (28)； Where: Indicated in k Moment i The velocity of a particle in dimension d, is the inertia weight, , is the learning factor, , is a random number in the range (0,1). represents individual extreme values, represents the global extreme value, i=1,2,…x; d=1,2,…D; Two variable operators Y and U are proposed, which are used to Optimize and enhance the algorithm's early global search capability and late local search capability. The algorithms for operators and inertia weights are shown in equations (29), (30), and (31); (29)； (30)； (31)； Where: is the current iteration number, is the maximum number of iterations; Using the formulas shown in equations (32) and (33) and Perform adaptive optimization: (32)； (33)； Where: = =1.

8. According to claim 7, a source-load interactive incentive decision-making method based on clean heating with a thermal storage electric boiler is characterized in that: The solution steps using the improved particle swarm algorithm using the nonlinear variable inertia weight algorithm are as follows: 1) Input the outdoor temperature at each time period of a typical winter day, the indoor heating temperature of clean heating users, and the electricity price as initial data; 2) Initialize system parameters and algorithm parameters, and assign random initial values ​​to each particle; 3) The particle iterates, updates the position, speed, learning factor and inertia weight after iteration, compares them with the fitness of the particle in the previous cycle, and memorizes and updates the individual extreme value; 4) Save the optimal individual extreme value as the global extreme value and update and save the individual extreme value; 5) The iteration ends when the global state is stable for five or more times in a row or the maximum number of iterations is reached.

Images