A multi-time scale optimization scheduling method for microgrid based on source-load flexibility

Abstract

The present invention relates to a microgrid multi-time scale optimization scheduling method based on source-load flexibility, comprising: constructing a microgrid multi-time scale optimization scheduling model; solving the microgrid multi-time scale optimization scheduling model to obtain a scheduling plan; calculating the scheduling priority of each real-time demand response load in the scheduling plan; and adjusting the scheduling order of each real-time demand response load in the scheduling plan according to the scheduling priority.

Description

Technical Field

[0001] The invention relates to a microgrid multi-time scale optimization scheduling method based on source-load flexibility, and belongs to the field of microgrid optimization scheduling. Background Art

[0002] Remote microgrids with distributed power sources as the main energy source have the characteristics of flexible power supply, short transmission distance, low energy consumption, and low pollution, which can solve the problem of insufficient energy supply in islands and remote areas. The capacity of microgrids themselves is limited, and the output of new energy such as wind power has inherent timing, seasonality and volatility, which will make remote microgrids face more severe problems of wind power consumption and peak shaving. Therefore, it is necessary to regulate microgrids on multiple time scales. Summary of the invention

[0003] In order to overcome the problems existing in the prior art, the present invention designs a microgrid multi-time scale optimization scheduling method based on source-load flexibility. A scheduling plan is obtained by establishing a microgrid multi-time scale optimization scheduling model. The day-ahead scheduling model is based on typical scenario values ​​of wind power and regional load, and optimizes the scheduling of the microgrid with the goal of minimizing the intraday operating cost. The intraday scheduling model is based on the day-ahead scheduling plan, comprehensively considers the characteristic indicators of the dispatchable resources, performs economic optimization scheduling, and obtains the scheduling plan of the load-side demand response resources within the day in two stages.

[0004] In order to achieve the above object, the present invention adopts the following technical solution:

[0005] A multi-time scale optimization scheduling method for a microgrid based on source-load flexibility comprises the following steps:

[0006] Construct a multi-time scale optimization dispatch model for microgrids;

[0007] Solving the multi-time scale optimization scheduling model of the microgrid to obtain a scheduling plan;

[0008] Calculating the dispatch priority of each real-time demand response load in the dispatch plan;

[0009] According to the scheduling priority, the scheduling order of each real-time demand response load in the scheduling plan is adjusted.

[0010] Furthermore, the scheduling priority is calculated according to the user's scheduling willingness and the schedulable capacity.

[0011] Furthermore, the scheduling priority is calculated and expressed as follows:

[0012]

[0013] Where: α i，t is the proportion of dispatchable capacity; βi，t is the user willingness; i，t is the scheduling priority; Q i，t is the curtailable capacity of the i-th real-time demand response load in time period t; m is the number of real-time demand response loads; ΔP LD (t) is the dispatch quantity at time t; α j，D is the unit dispatch cost of real-time demand response load; P LD (t) Electricity arrangement for real-time demand response load users before dispatch; i，f is the unit electricity price before type i dispatch.

[0014] Furthermore, it also includes: calculating user evaluation of the scheduling plan based on the user's electricity consumption comfort and electricity consumption economy.

[0015] Furthermore, the user evaluation is calculated and expressed as follows:

[0016]

[0017] Where: R i is the evaluation of the ith user; α1 is the user's economic preference coefficient; α2 is the user's comfort preference coefficient; t represents the scheduling time node; E cost E is the economic efficiency of electricity use; cosy For the comfort of electricity use; i，f is the unit electricity price that the i-th user needs to pay before dispatch; p i P is the unit electricity price that the i-th user needs to pay after dispatch; IDRD To optimize the scheduling arrangement of users after scheduling; LD To optimize electricity usage arrangements before dispatch.

[0018] Furthermore, the microgrid multi-time scale optimization scheduling model includes a day-ahead scheduling model and an intraday scheduling model.

[0019] Furthermore, the day-ahead scheduling model is expressed as follows:

[0020] F1=minC1=C PW +C loss +C EES1 +C A +C g

[0021]

[0022] Where: C1 is the day-ahead scheduling cost; C PW is the operating cost of wind turbines; C loss Penalty for load loss; C EES1 is the operating cost of the battery; C AThe dispatch cost of the demand response load with a response time of more than 1h; C g The cost of purchasing electricity from the large power grid; K loss P is the penalty coefficient for unit power of load loss; loss (t) is the load loss size of the microgrid at time t.

[0023] Furthermore, the intraday scheduling model is expressed as follows:

[0024] F2=minC2=C PW +C loss +C EES1 +C B +C g

[0025] F3=minC2=C PW +C loss +C EES2 +C DE +C C +C D +C g

[0026] Where: F2 is the cost of hourly scheduling within a day; F3 is the cost of 15-minute scheduling within a day; C PW is the operating cost of wind turbines; C loss Penalty for load loss; C EES1 is the operating cost of the battery; C EES2 is the operating cost of the supercapacitor; C DE is the operating cost of the diesel generator; C g Cost of purchasing electricity from the large power grid; C B is the dispatch cost of the demand response load with a response time of 15min-1h; C C is the dispatching cost of the demand response load with a response time of 5min-15min; C D The dispatch cost of load for real-time demand response.

[0027] Compared with the prior art, the present invention has the following characteristics and beneficial effects:

[0028] 1. The dispatch plan is obtained by establishing a multi-time scale optimization dispatch model for microgrids. The day-ahead dispatch model is based on the typical scenario values ​​of wind power and regional loads, and optimizes the dispatch of microgrids with the goal of minimizing the intraday operating cost. The intraday dispatch model is based on the day-ahead dispatch plan, comprehensively considers the characteristic indicators of dispatchable resources, and performs economic optimization dispatch. The dispatch plan of load-side demand response resources within the day is obtained in two stages.

[0029] 2. The user willingness is calculated based on the ratio of the predicted value of the user scheduling subsidy to the predicted value of the user load cost. The obtained user willingness can accurately reflect the user's willingness for the Class D IDR used by the user to be scheduled.

[0030] 3. Determine the dispatch priority according to the dispatchable capacity and user willingness, and adjust the dispatch order of the real-time demand response load in the dispatch plan according to the dispatch priority, which can effectively improve the user's electricity experience.

[0031] 4. Quantify the user's electricity comfort and economy through the evaluation formula, set corresponding weights to reflect the different electricity preferences of different types of users, and obtain the user's evaluation of different scheduling arrangements. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is the fan output diagram;

[0033] Figure 2 is the load forecast curve;

[0034] Figure 3 is the wind power prediction curve;

[0035] Figure 4 is the Class A load dispatch curve;

[0036] Figure 5 is the Class B load dispatch curve;

[0037] Figure 6 is the Class C load dispatch curve;

[0038] Figure 7 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0039] Glossary: ​​Demand response refers to the behavior of electricity users to actively adjust their electricity consumption behavior according to the electricity price signals or economic incentive mechanisms issued by the power company to alleviate the tight electricity supply. According to different response mechanisms, DR can be divided into two types: price demand response (PDR) and incentive demand response (IDR). PDR guides electricity users to adjust their electricity consumption behavior for lower electricity prices through price signals (time-of-use electricity prices, real-time electricity prices, peak electricity prices, etc.), thereby achieving the effect of peak shaving and valley filling and promoting the consumption of renewable energy; IDR is based on the contract signed between the power company and the user, and encourages users to participate in the load reduction projects required by the power system through economic compensation or electricity price discounts. It is mainly used to deal with emergency situations such as power grid load peaks or system reliability accidents. IDR includes interruptible loads, demand-side bidding, emergency demand-side response, direct load control, etc. IDR is divided into the following types according to the length of time it takes to respond to power grid dispatch instructions:

[0040] 1) Class A IDR: The response time of this type of load is > 1h, and the user needs to be notified one day in advance.

[0041] 2) Class B IDR: The response time of this type of load is 15 minutes to 1 hour, and the user needs to be informed 1 hour to 4 hours in advance.

[0042] 3) Class C IDR: The response time of this type of load is 5min-15min, and the user needs to be informed 1h-15min in advance.

[0043] 4) Class D IDR: This type of load can respond in real time and determine the calling plan in the real-time scheduling stage.

[0044] The present invention is described in more detail below with reference to the embodiments.

[0045] Embodiment 1

[0046] S1. Construct a multi-time scale optimization dispatch model for microgrids;

[0047] S2. Use the General Algebraic Modeling System (GAMS) to optimize and solve the multi-time scale optimization scheduling model of the microgrid, and obtain the scheduling plan of each power equipment with the goal of maximizing the economic benefits within multiple time scales, that is, obtain the wind power output power, energy storage device charging and discharging power and load resource scheduling power in each time period.

[0048] S3. Optimize the scheduling order of the real-time demand response load (specifically Class D IDR in this embodiment) in the scheduling plan:

[0049] S31. The calculation formula for establishing the scheduling priority is as follows:

[0050]

[0051] Where: α i，t is the proportion of dispatchable capacity; Q i，t is the reducible capacity of the i-th Class D IDR in time period t; m is the number of Class D IDRs; β i，t is the user willingness. The more dispatch subsidies the user gets after dispatching than the load cost the user has to pay before dispatching, the higher the user willingness. The default value for industrial users is 1. LD (t) is the dispatch quantity at time t; α j，D is the unit scheduling cost of class D IDR; P LD (t) is the electricity consumption arrangement of Class D IDR users before dispatch; i，f is the unit electricity price of category i users before dispatch (i.e. without subsidies); i，t The scheduling priority.

[0052] S32. According to the calculation formula in step S31, the scheduling priority of each Class D IDR is calculated in turn; according to the scheduling priority, the scheduling order of the Class D IDR in the scheduling plan is determined.

[0053] The beneficial effects of this embodiment are:

[0054] 1. The user willingness is calculated based on the ratio of the predicted value of the user scheduling subsidy to the predicted value of the user load cost. The obtained user willingness can accurately reflect the user's willingness for the Class D IDR used by the user to be scheduled.

[0055] 2. Determine the dispatch priority according to the dispatchable capacity and user willingness, and adjust the dispatch order of the real-time demand response load in the dispatch plan according to the dispatch priority, which can effectively improve the user's electricity experience.

[0056] Embodiment 2

[0057] Constructing a multi-time scale optimization dispatch model for microgrids includes the following steps:

[0058] A1. Build a day-ahead scheduling model:

[0059] The first objective function is constructed with the goal of minimizing the day-ahead scheduling cost:

[0060] F1=minC1=C PW +C loss +C EES1 +C A +C g (1)

[0061]

[0062] Where: C1 is the day-ahead scheduling cost; C PW is the operating cost of wind turbines; C loss Penalty for load loss; C EES1 is the operating cost of the battery; C A is the dispatching cost of Class A IDR load; C g The cost of purchasing electricity from the large power grid; K loss P is the penalty coefficient for unit power of load loss; loss (t) is the load loss size of the microgrid at time t.

[0063] Wind turbine operating cost C PW The calculation formula is as follows;

[0064]

[0065]

[0066] Where, v is the measured wind speed; C P is the performance parameter of the wind turbine; ρ is the air density; A WT It is the projection of the swept area of ​​the wind turbine blades on the plane perpendicular to the wind speed; is the rated power of the wind turbine; V c is the cut-in wind speed; V r is the rated wind speed; V f K is the cut-out wind speed; WT is the wind turbine operating cost coefficient;

[0067] P WT (t) is the electric power output of the wind turbine during period t. The wind speed varies at different heights, so it is necessary to convert the measured wind speed at a specific height into the actual wind speed at the height of the wind turbine tower.

[0068] The dispatching cost of each type of load is:

[0069]

[0070] Where: C A , C B , C C , C D are the dispatching costs of class A IDR, class B IDR, class C IDR, and class D IDR loads, respectively, j，A , α j，B , α j，C , α j，D are the unit dispatching costs of various loads, P j，B，t , P j，B，t , P j，C，t , P j，D，tThe dispatching power for various loads.

[0071] Diesel generators are a commonly used small generator that generally provide backup or peak load services. When wind power, photovoltaic power, etc. cannot meet the load power supply, diesel generators participate in the regulation to ensure that important loads are not interrupted in power supply and improve power supply reliability. The cost model of diesel generators is as follows:

[0072]

[0073] Where: K DE is the unit cost coefficient, F DE is the fuel cost coefficient per unit of electricity generation, P DE.i (t) is the power generated by the i-th diesel generator at time t.

[0074] The refined modeling of energy storage equipment is the key to the reasonable dispatch and operation of the wind power storage integrated microgrid system. The joint operation of energy storage technology and renewable energy technology can effectively alleviate the intermittent and volatile nature of new energy output, improve the utilization rate of new energy, and play an indispensable role in the safe and stable operation of microgrids. The cost model of batteries and supercapacitors is as follows:

[0075]

[0076] Where: is the construction cost of energy storage resources; P EES is the rated capacity of the energy storage resource; N is the number of charge and discharge cycles of the energy storage resource; is the charging power; is the discharge power; C EES K is the energy storage management cost; EES is the unit operation and maintenance cost coefficient.

[0077] The constraints of the first objective function are set as follows:

[0078] Power balance constraints:

[0079]

[0080] Where: P PW (t) is the electric power output of the wind turbine at time t; P G (t) is the power purchased from the large power grid at time t; and is the discharge power and charge power of the battery at time t; P L (t) is the electric power consumed by the load at time t; P loss (t) is the load loss size of the microgrid at time t; P LA (t) is the scheduling arrangement of Class A IDR at time t.

[0081] PDR constraints:

[0082]

[0083] The constraints for Class A IDR load are:

[0084]

[0085] The upper and lower limits of the energy storage device’s charge and discharge are:

[0086] S EES1.min ≤S EES1 (t)≤S EES.1max (11)

[0087] Ju You and They are the adjustable lower and upper limits of price-based demand response respectively; and They are the lower and upper limits of the diesel generator output respectively; and and are the upper and lower limits of the adjustable IDR for Class A respectively; S EES.1min and S EES1.max They are respectively the lower limit and upper limit of the charging and discharging power of the energy storage device.

[0088] A2. Constructing an intraday scheduling model:

[0089] The second objective function is constructed with the goal of minimizing the hourly scheduling cost within the day, and the third objective function is constructed with the goal of minimizing the 15-minute scheduling cost within the day:

[0090] F2=minC2=C PW +C loss +C EES1 +C B +C g (12)

[0091] F3=minC2=C PW +C loss +C EES2 +C DE +C C +C D +C g (13)

[0092] Where: F2 is the cost of hourly scheduling within a day; F3 is the cost of 15-minute scheduling within a day; C B is the scheduling cost of class B IDR; C EES1 is the operating cost of the battery; C EES2 is the operating cost of the supercapacitor; CDE is the operating cost of the diesel generator; C C is the scheduling cost of class C IDR; D is the scheduling cost of Class D IDR.

[0093] The hourly power balance constraint is:

[0094]

[0095] The 15-minute power balance constraint is:

[0096]

[0097] Where: P PW (t) is the electric power output of the wind turbine at time t; P G (t) is the power purchased from the large power grid at time t; and are the discharge power of the battery and supercapacitor at time t; P L (t) is the electric power consumed by the load at time t; P DE (t) is the electric power output of the diesel generator at time t; P loss (t) is the load loss size of the microgrid at time t; P LA (is the scheduling arrangement of Class B IDR at time t; P LC (t) is the scheduling arrangement of Class C IDR at time t; P LD (t) is the scheduling arrangement of Class D IDR at time t.

[0098] The upper and lower limits of the diesel generator output are:

[0099]

[0100] Ju You and They are the lower and upper limits of the diesel generator output respectively.

[0101] The constraints of Class B, C and D IDRs are the same as those of Class A IDRs, and they cannot exceed their adjustable upper and lower limits. The constraints of supercapacitors and batteries are the same, and they cannot exceed the lower and upper limits of charge and discharge power.

[0102] The progress of this embodiment is that a scheduling plan is obtained by establishing a multi-time scale optimization scheduling model for a microgrid, wherein the day-ahead scheduling model is based on typical scenario values ​​of wind power and regional load, and optimizes the scheduling of the microgrid with the goal of minimizing the intraday operating cost; the intraday scheduling model is based on the day-ahead scheduling plan, comprehensively considers the characteristic indicators of the dispatchable resources, performs economic optimization scheduling, and obtains the scheduling plan of the load-side demand response resources within the day in two stages.

[0103] Embodiment 3

[0104] Furthermore, the user's evaluation of the scheduling plan is calculated and expressed as:

[0105]

[0106] Where: R i is the user evaluation of the i-th IDR user; α1 is the user's economic preference coefficient; α2 is the user's comfort preference coefficient; t represents the scheduling time node, and there is a scheduling every 15 minutes; E cost，i is the electricity economy of the i-th IDR user; E cosy，i is the electricity comfort of the i-th IDR user; p i，f is the unit electricity price that the i-th IDR user needs to pay before dispatch; p i P is the unit electricity price that the i-th user needs to pay after dispatch; IDRD The scheduling arrangement for IDR users after scheduling; LD To optimize the electricity usage arrangements of IDR users before dispatch.

[0107] The improvement of this embodiment is that the user's electricity comfort and economy are quantified through an evaluation formula, and corresponding weights are set to reflect the different electricity preferences of different types of users, so as to obtain the user's evaluation of different scheduling arrangements.

[0108] Embodiment 4

[0109] Select the wind turbine output data for the whole year of 2019 in a certain place, as shown in the attached Figure 1 As shown in the figure. Based on the model established in this paper, the dispatch of various resources at different time scales is studied. First, the day-ahead dispatch is carried out, and then the optimization dispatch of 1 hour and 15 minutes within the day is carried out in a rolling manner. Finally, the dispatch results are connected together to draw the dispatch result curve. The load forecast curves and wind power forecast curves of the day-ahead, 1 hour and 15 minutes within the day are shown in the figure. Figure 2 and attached Figure 3 shown.

[0110] (1) Analysis of results at multiple time scales

[0111] ① Analysis of the day-ahead dispatch results: Figure 5 is the dispatching curve of Class A load in different dispatching periods. Figure 4As shown in the figure: Class A load transfers the load from 08:00-20:00 to 00:00-20:00 and 20:00-24:00, so that a large amount of load is transferred from the peak load period to the low load period, achieving the purpose of peak elimination and valley filling. At the same time, a large amount of load is transferred to the evening, which can absorb wind power and reduce wind abandonment. Class A load is quickly dispatched around 08:00 and 20:00 to better adapt to the rapid changes in load demand during this period.

[0112] ② Analysis of 1h scheduling results within a day: Attachment Figure 5 is the dispatch curve of Class B load in different dispatch periods. Figure 6 As shown in the figure, Class B load transfers the load from 08:00 to 22:00 to 00:00 to 08:00. Since the load demand changes greatly around 08:00, Class B load is quickly dispatched during this period, cooperating with Class A load to complete the task of eliminating peak loads and filling valleys. During the 20:00 period, the load demand changes relatively little, and the dispatch of Class A load alone can meet the requirements, so Class B load changes more slowly during this period.

[0113] ③ Analysis of 15-minute scheduling results within a day: Attachment Figure 6 is the dispatch curve of class C and class D loads in different dispatch periods. Figure 7 As shown in the figure: Class C and Class D loads act as system backup and respond quickly in a short time. When the wind power output value decreases, the load is reduced to reduce the balancing pressure of the generator set. Class D loads are called to deal with the supply and demand imbalance caused by wind power output and load fluctuations, which reduces the balancing pressure of the generator set on a short time scale and ensures the safe and stable operation of the system.

[0114] (2) Considering the dispatch priority of Class D load

[0115] ① Since different types of Class D loads have different dispatch priorities under incentive demand response, it is necessary to sort the dispatch priorities of residential loads, commercial loads, and industrial loads respectively, so as to determine the dispatch order of different types of loads. Table 1 shows the dispatch priority ranking of Class D loads during their dispatch period.

[0116] Table 1 Scheduling priority and ranking of different IDRs

[0117]

[0118]

[0119] As shown in Table 1: the time periods with the dispatch priority of residential load being 1 are mostly concentrated in the morning and afternoon. At this time, the demand for residential load is low, there is sufficient adjustable capacity for dispatching and the dispatching is flexible, so the dispatching priority of residential load is higher; the time periods with the dispatch priority of industrial load being 1 are more, and most of them are concentrated in the noon period. At this time, the demand for residential load is low and the demand for industrial load is low, so the dispatching priority of industrial load is higher; the dispatching priority of commercial load is always 3. Since the commercial load has low volatility in each time period and low dispatching flexibility, the commercial load will only be dispatched when the dispatching of residential load and industrial load alone cannot meet the requirements.

[0120] ② In order to analyze the effectiveness of economic dispatch considering dispatch priority, two scenarios are designed and analyzed and compared. Scenario 1: Economic dispatch considering dispatch priority; Scenario 2: Economic dispatch not considering dispatch priority. Table 2 compares the dispatch results of the two scenarios.

[0121] Table 2 Results of different scheduling strategies

[0122]

[0123]

[0124] As shown in Table 2: Since the dispatch priority of commercial loads is the lowest in economic dispatch considering dispatch priority, the dispatch capacity of commercial loads has decreased, while the dispatch capacity of residential loads and industrial loads has increased compared with economic dispatch not considering dispatch priority. The dispatch subsidy obtained by Class D loads under incentive-based demand response is only related to the dispatch capacity, so the dispatch subsidy obtained by commercial loads in Scenario 1 is lower than that in Scenario 2, and the dispatch subsidy obtained by residential loads and industrial loads is higher than that in Scenario 2. The user evaluation in Scenario 1 is higher than that in Scenario 2, indicating that users of various types of Class D loads, including residential loads, commercial loads, and industrial loads, are more satisfied with Scenario 1.

[0125] According to the above implementation cases, it can be seen that the present invention can flexibly dispatch various load resources through multi-time scale dispatch strategies, so as to achieve the best economic efficiency under the premise of meeting the power supply reliability. At the same time, setting the dispatch priority and user evaluation index can more reasonably dispatch the real-time demand response load and improve user evaluation.

[0126] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

Claims

1. A multi-time scale optimization scheduling method for microgrid based on source-load flexibility, characterized in that: The following steps are involved: Constructing a microgrid multi-time scale optimization scheduling model, wherein the microgrid multi-time scale optimization scheduling model includes a day-ahead scheduling model and an intra-day scheduling model; The day-ahead scheduling model is expressed as: F1=min C1=C PW +C loss +C EES1 +C A +C g Where: C1 is the day-ahead scheduling cost; C PW is the operating cost of wind turbines; C loss Penalty for load loss; C EES1 is the operating cost of the battery; C A is the dispatch cost of the demand response load with a response time of more than 1h; C g The cost of purchasing electricity from the large power grid; K loss P is the penalty coefficient for unit power of load loss; loss (t) is the load loss size of the microgrid at time t; The intraday scheduling model is expressed as follows: F2=minC2=C PW +C loss +C EES1 +C B +C g F3=minC2=C PW +C loss +C EES2 +C DE +C C +C D +C g Where: F2 is the cost of hourly scheduling within a day; F3 is the cost of 15-minute scheduling within a day; C PW is the operating cost of wind turbines; C loss Penalty for load loss; C EES1 is the operating cost of the battery; C EES2 is the operating cost of the supercapacitor; C DE is the operating cost of the diesel generator; C g Cost of purchasing electricity from the large power grid; C B is the dispatch cost of the demand response load with a response time of 15min-1h; C C is the dispatching cost of the demand response load with a response time of 5min-15min; C D the dispatch costs of loads for real-time demand response; Solving the multi-time scale optimization scheduling model of the microgrid to obtain a scheduling plan; According to the user's dispatch willingness and dispatchable capacity, the dispatch priority of each real-time demand response load in the dispatch plan is calculated and expressed as follows: Where: α i，t is the proportion of dispatchable capacity; β i，t is the user willingness; i，t is the scheduling priority; Q i，t is the curtailable capacity of the i-th real-time demand response load in time period t; m is the number of real-time demand response loads; ΔP LD (t) is the dispatch quantity at time t; α j，D is the unit dispatch cost of real-time demand response load; P LD (t) Electricity arrangement for real-time demand response load users before dispatch; i，f is the unit electricity price before class i dispatch; According to the scheduling priority, adjusting the scheduling order of each real-time demand response load in the scheduling plan; According to the user's electricity comfort and electricity economy, the user evaluation of the scheduling plan is calculated and expressed as: Where: R i is the evaluation of the ith user; α1 is the user's economic preference coefficient; α2 is the user's comfort preference coefficient; t represents the scheduling time node; E cost E is the economic efficiency of electricity use; cosy For the comfort of electricity use; i，f is the unit electricity price that the i-th user needs to pay before dispatch; p i P is the unit electricity price that the i-th user needs to pay after dispatch; IDRD To optimize the scheduling arrangement of users after scheduling; LD To optimize electricity usage arrangements before dispatch.

Images