A data-driven risk-averse optimization approach for a tradably controlled federal power plant

By adopting a tradable controlled federal power plant data-driven risk aversion optimization method in federal power plants, using a temperature-controlled load-based virtual generator and virtual battery equivalent model, combined with a tradable energy market, we solve renewable energy volatility and users The problems reflected by intention have achieved improvements in the stability of the power system and user participation.

CN114139402BActive Publication Date: 2024-11-15SOUTHEAST UNIV
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202111510595.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-11
Publication Date
2024-11-15
Estimated Expiration
2041-12-11

AI Technical Summary

Technical Problem

The prior art is difficult to effectively curb the fluctuations in renewable energy output and cannot fully reflect the user's own wishes, resulting in insufficient power system stability and user participation.

Method used

Using a data-driven risk avoidance optimization method for federal power plants that can be traded and controlled, the FPP two-stage optimization scheduling framework with smooth output fluctuations is constructed, and the temperature control load is equivalent to virtual power generation. The machine and virtual battery use the tradable energy market mechanism and combine the power adjustment characteristics of different time scales to optimize the scheduling scheme.

Benefits of technology

Effectively suppress renewable energy power fluctuations, improve user participation and satisfaction, guide users to participate in regulation through price signals, consider user comfort, cost and benefits, and improve power system stability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114139402B_ABST
    Figure CN114139402B_ABST
Patent Text Reader

Abstract

The present invention discloses a data-driven risk avoidance optimization method for a tradable federal power plant, comprising the following steps: S1, constructing a two-stage FPP optimization scheduling framework for smooth output fluctuations; S2, constructing a tradable energy market mechanism within the FPP; S3, constructing a two-stage FPP optimization scheduling model based on the Wasserstein distance, and solving to obtain the optimal scheduling scheme. The present invention proposes a two-stage joint optimization method for federal power plants at hour and minute time scales, and according to the difference in power regulation characteristics of temperature-controlled loads under different control modes, the temperature-controlled loads are respectively equivalent to virtual generators and virtual batteries to participate in the regulation of federal power plants, effectively suppressing the power fluctuations of renewable energy while fully tapping the potential of temperature-controlled load regulation. Different from traditional demand response, the proposed model can well solve the problem of renewable energy volatility, and the market-based tradable control can fully respect the wishes of users, providing technical support for large-scale renewable resource grid-connected scheduling.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of electric power technology, and in particular relates to a data-driven risk avoidance optimization method for a federal power plant that can be traded and controlled. Background Art

[0002] With the continuous development of science and technology, the transformation of traditional fossil energy to clean and renewable energy has become a global trend. With the increase in the penetration rate of renewable energy, the increased volatility on the power generation side has brought great challenges to the stability of the power system. At the same time, temperature control load, as a typical load-side flexible load, has become an important resource for power demand-side regulation and management. The coordinated control of power supply-side and load-side resources has become a major manifestation of today's new power system. So far, demand response technology has begun to be applied to actual power systems, but its regulation potential is limited and it is difficult to effectively smooth out the fluctuations in new energy output, and it cannot fully reflect the wishes of users themselves. Therefore, a method that can effectively solve the problem of renewable energy volatility and fully reflect the wishes of users is in line with actual needs.

[0003] In response to the above problems, a data-driven risk aversion optimization method for federal power plants with tradable control is designed. Summary of the invention

[0004] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a data-driven risk avoidance optimization method for a federal power plant with tradable control. Different from traditional demand response, the proposed model can well solve the problem of renewable energy volatility, and the market-based tradable control can fully respect the user's wishes, providing technical support for large-scale renewable resource grid-connected scheduling.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A data-driven risk aversion optimization method for a tradably controlled federal power plant includes the following steps:

[0007] S1. Construct a two-stage FPP optimization scheduling framework to smooth output fluctuations;

[0008] S2. Construct a tradable energy market mechanism within the FPP;

[0009] S3. Construct a two-stage FPP optimization scheduling model based on Wasserstein distance and solve it to obtain the optimal scheduling solution.

[0010] Furthermore, a two-stage FPP optimization model based on multi-time-scale coordinated control of temperature-controlled loads is adopted. At the hourly time scale, the temperature-controlled load with temperature adjustment function is simulated as a virtual generator; at the minute time scale, the temperature-controlled load with frequency control function is regarded as a virtual battery.

[0011] Furthermore, the S1 specifically includes the following steps:

[0012] S1.1. Constructing an hourly equivalent model of temperature control load

[0013] The temperature-controlled load controls its output power through temperature regulation and frequency regulation. According to the physical regulation characteristics of the two methods, they have different time scale regulation effects.

[0014] In the hourly time scale regulation, the power consumption change of the temperature-controlled load is based on the temperature regulation, which can be maintained for a long time but cannot respond quickly. Therefore, the temperature-controlled load is equivalent to a virtual generator on the hourly time scale, which can be described as:

[0015]

[0016]

[0017]

[0018] In formulas ①, ②, and ③, P g is the power of the virtual generator, is the minimum value of the virtual generator power, is the maximum value of the virtual generator power, ΔT max is the maximum value of the allowable increment (ΔT) of the internal temperature, and κ is a constant parameter related to the temperature control load;

[0019] S1.2. Constructing a minute-scale equivalent model of temperature control load

[0020] In minute-scale regulation, the power consumption change of the temperature-controlled load is based on frequency control, which can respond to signals quickly but cannot be maintained for a long time. Therefore, the temperature-controlled load is equivalent to a virtual battery on a minute scale, which can be described as:

[0021]

[0022] SOC(k+1)=αSOC(k)+βP b (k)+γ ⑤

[0023] SOC min ≤SOC(k+1)≤SOC max ⑥

[0024] In formulas ④, ⑤, and ⑥, Pb is the charge and discharge power of the virtual battery, YesP b The minimum value of the virtual battery charge and discharge power, YesP b The maximum value of the virtual battery charge and discharge power, α, β and γ are the virtual battery parameters, SOC is the charging state of the virtual battery, SOC min is the minimum value of SOC, SOC max is the maximum value of SOC;

[0025] S1.3. Hourly and minute-scale coupled coordinated control of temperature control load

[0026] In the minute-scale model of temperature-controlled load, some parameters of the equivalent virtual battery are determined by the preset internal temperature Determined, the virtual battery parameters in formulas ④, ⑤, and ⑥ are: β and γ depend on Determined in the hourly scheduling of temperature control load, it can be described as:

[0027]

[0028]

[0029]

[0030] In formulas ⑦, ⑧, and ⑨, T o is the external temperature, T max is the maximum internal temperature;

[0031] The initial SOC value of the virtual battery SOC 0 depending on It can be described as:

[0032]

[0033] Furthermore, the S2 specifically includes the following steps:

[0034] On a one-hour time horizon, the objective function for customers in the tradable energy market is to minimize costs and maximize comfort levels, described as:

[0035]

[0036] formula In the equation, λ is the bid price of the temperature control load in the tradable energy market, λ0 is the electricity price of the temperature control load, is the baseline power of the temperature control load, a is a constant parameter representing customer preference, the larger a is, the higher the user's tolerance for temperature adjustment, aΔT 2represents the user's temperature adjustment cost, aΔT 2 The smaller it is, the higher the customer's comfort level is. is the electricity cost, λP g It is the compensation of FPP for temperature control load;

[0037] For simplicity, let λP g > 0, according to the temperature control load hour time scale model, use P g / κ replaces ΔT, so the formula Re-describe as:

[0038]

[0039] When the formula When the minimum value is reached, the following conditions must be met:

[0040]

[0041] Correspondingly, P g The relationship between and λ can be described as:

[0042]

[0043] Since the power adjustment range of the virtual generator is limited to To ensure user comfort, the Lagrangian condition P g The relationship between and λ is as follows:

[0044]

[0045] Obviously, comfort and power adjustment range affect the optimal decision of temperature control load output. In the regulation of minute time scale, the internal temperature of temperature control load changes within a limited range. In addition, as the time scale becomes shorter, the operator's forecast error will also become smaller accordingly. The day-ahead plan is alleviated by the FPP's hour-ahead scheduling, and the hour-ahead plan is smoothed by the FPP's hour-in-hour scheduling.

[0046] Furthermore, the S3 specifically includes the following steps:

[0047] S3.1. Optimal FPP Scheduling in Hours

[0048] Before optimizing the scheduling on an hourly time scale, FPP needs to arrange the output power of virtual generators according to the prediction results of RES to smooth the fluctuation of renewable energy output. The optimization goal of FPP is to minimize the control cost of temperature control load. Therefore, the FPP optimization scheduling model on an hourly basis is as follows:

[0049]

[0050]

[0051] Since the model is a non-convex function, the model is appropriately transformed, and the converted FPP optimization scheduling model in hours is as follows:

[0052]

[0053]

[0054] formula In, λ t is the market settlement price at time t, Δt h is the time interval in hours, N is the set of temperature control loads, T is the set of hourly time periods, and When t When it is a segment boundary point, λ t , P g,i,t and M i,t The value of S g and m g To determine λ t The introduced variable, R1 is S g The set of R2 is m g The set of r1 is S g The index of r2 is m g The index of

[0055] S3.2. FPP Optimization Scheduling in Minutes

[0056] In the FPP optimization scheduling process in minutes, the temperature control load is regarded as a virtual battery, so the centralized control potential of the virtual battery in the second stage needs to be evaluated, and the relevant scheduling parameters are calculated as follows:

[0057]

[0058]

[0059]

[0060]

[0061]

[0062]

[0063]

[0064] According to the above parameters, a similar FPP optimization scheduling model based on the minute time scale is as follows:

[0065]

[0066]

[0067] formula In the m is the time slot on the minute time scale, P load is the forecast load in minutes, P w is the wind power in minutes, It is the fluctuation of renewable energy output;

[0068] S3.3. FPP two-stage optimization scheduling based on Wasserstein distance

[0069] The above two-stage optimization scheduling is given as follows:

[0070]

[0071] x is the decision variable of the first stage, X is the constraint of the optimization model of the first stage, c is the coefficient vector, ξ is the random variable of renewable energy, P is the probability distribution of ξ, and Ω is the fuzzy set of P;

[0072] Α(x,ξ) is expressed as:

[0073]

[0074] formula In the above equation, y is the decision variable of the second stage, q is the coefficient vector, T, H and W are coefficient matrices, and P and Wasserstein distance are used to determine the fuzzy set Ω. When η∈(0,1), the confidence level will not be lower than η.

[0075]

[0076] formula In the equation, D is the diameter of the probability distribution space of ξ. Obviously, the radius of the Wasserstein ball becomes shorter as S increases, which will reduce the range of fuzzy sets contained in the confidence interval of the Wasserstein ball. Therefore, ξ is restated as:

[0077]

[0078] According to the conversion conditions, the original two-stage optimization model is converted to the following model:

[0079]

[0080]

[0081] The model is solved to obtain the optimal temperature adjustment of the temperature control load in the first stage and the optimal power adjustment of the temperature control load in the second stage. The temperature control load is controlled according to the optimization calculation results to achieve the suppression of renewable energy power fluctuations by the temperature control load.

[0082] Beneficial effects of the present invention:

[0083] The data-driven risk avoidance optimization method for a tradable controlled federal power plant proposed in the present invention takes into account the characteristic that the power fluctuation of renewable energy decreases with the decrease of the prediction time scale, and proposes a two-stage joint optimization method for the federal power plant at the hour and minute time scales. According to the different power regulation characteristics of the temperature control load under different control modes, the temperature control load is respectively equivalent to a virtual generator and a virtual battery to participate in the federal power plant regulation, which effectively suppresses the power fluctuation of renewable energy while fully tapping the regulation potential of the temperature control load. The mechanism of the tradable energy market is utilized to guide users to participate in the federal power plant regulation through price signals, and the three factors of user comfort, cost and benefit are fully considered in the process of controlling the temperature load, so as to improve the enthusiasm and satisfaction of users in participating in the regulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0085] Figure 1 is a flow chart of an optimization scheduling method according to an embodiment of the present invention;

[0086] Figure 2 is a bidding curve diagram of customers in a tradable energy market according to an embodiment of the present invention;

[0087] Figure 3 is a schematic diagram of FPP scheduling under different time scales in an embodiment of the present invention;

[0088] Figure 4 is a schematic diagram of the output curve of the present invention;

[0089] Figure 5 is a schematic diagram of an output curve hours ago of an embodiment of the present invention;

[0090] Figure 6 Schematic diagram of output curve within one hour according to an embodiment of the present invention. DETAILED DESCRIPTION

[0091] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the 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.

[0092] like Figure 1 As shown, a data-driven risk aversion optimization method for a tradably controlled federal power plant includes the following steps:

[0093] S1. Construct a two-stage optimal dispatch framework for FPP (Federal Power Plant) to smooth output fluctuations;

[0094] S2. Construct a tradable energy market mechanism within the FPP;

[0095] S3. Construct a FPP two-stage optimization scheduling model based on Wasserstein distance (used to measure the distance between two probability distributions) and solve it to obtain the optimal scheduling solution.

[0096] The FPP composed of renewable energy and temperature-controlled load is connected to the main power grid. In order to make full use of the flexibility of the temperature-controlled load to solve the fluctuation of renewable energy output, this application adopts a two-stage optimization model of FPP based on multi-time scale coordinated control of the temperature-controlled load. At the hourly time scale (first stage), the temperature-controlled load with temperature adjustment function is simulated as a virtual generator; at the minute time scale (second stage), the temperature-controlled load with frequency control function is regarded as a virtual battery.

[0097] S1 specifically includes the following steps:

[0098] S1.1. Constructing an hourly equivalent model of temperature control load

[0099] The temperature-controlled load controls its output power through temperature regulation and frequency regulation. According to the physical regulation characteristics of the two methods, they have different time scale regulation effects.

[0100] In the hourly time scale regulation, the power consumption change of the temperature-controlled load is based on the temperature regulation, which can be maintained for a long time but cannot respond quickly. Therefore, the temperature-controlled load is equivalent to a virtual generator on the hourly time scale, which can be described as:

[0101]

[0102]

[0103]

[0104] In formulas ①, ②, and ③, Pg is the power of the virtual generator, is the minimum value of the virtual generator power, is the maximum value of the virtual generator power, ΔT max is the maximum value of the allowable increment (ΔT) of the internal temperature, and κ is a constant parameter related to the temperature control load.

[0105] S1.2. Constructing a minute-scale equivalent model of temperature control load

[0106] In minute-scale regulation, the power consumption change of the temperature-controlled load is based on frequency control, which can respond to signals quickly but cannot be maintained for a long time. Therefore, the temperature-controlled load is equivalent to a virtual battery on a minute scale, which can be described as:

[0107]

[0108] SOC(k+1)=αSOC(k)+βP b (k)+γ ⑤

[0109] SOC min ≤SOC(k+1)≤SOC max ⑥

[0110] In formulas ④, ⑤, and ⑥, P b is the charge and discharge power of the virtual battery, YesP b The minimum value of the virtual battery charge and discharge power, YesP b The maximum value of the virtual battery charge and discharge power, α, β and γ are the virtual battery parameters, SOC is the charging state of the virtual battery, SOC min is the minimum value of SOC, SOC max It is the maximum value of SOC.

[0111] S1.3. Hourly and minute-scale coupled coordinated control of temperature control load

[0112] In the minute-scale model of temperature-controlled load, some parameters of the equivalent virtual battery are determined by the preset internal temperature Determined, the virtual battery parameters in formulas ④, ⑤, and ⑥ are: β and γ depend on Determined in the hourly scheduling of temperature control load, it can be described as:

[0113]

[0114]

[0115]

[0116] In formulas ⑦, ⑧, and ⑨, T o is the external temperature, T max is the maximum internal temperature.

[0117] The initial SOC value of the virtual battery SOC 0 depending on It can be described as:

[0118]

[0119] S2 specifically includes the following steps:

[0120] In order to protect the privacy and preferences of end users, a tradable energy market is established within the FPP for temperature control load power regulation. Users with temperature control load equipment need to submit their bidding curves based on their preferences and economic interests. Within a one-hour time frame, the objective function of customers in the tradable energy market is to minimize costs and maximize comfort levels, which can be described as:

[0121]

[0122] formula In the equation, λ is the bid price of the temperature control load in the tradable energy market, λ0 is the electricity price of the temperature control load, is the baseline power of the temperature control load, a is a constant parameter representing customer preference, and aΔT 2 represents the user's temperature adjustment cost, aΔT 2 The smaller it is, the higher the customer's comfort level is. is the electricity cost, λP g It is the compensation of FPP for temperature control load.

[0123] For simplicity, since λ and P g have the same sign (i.e., λP g >0), according to the temperature control load time scale model, use P g / κ replaces ΔT, so the formula Re-describe as:

[0124]

[0125] When the formula When the minimum value is reached, the following conditions must be met:

[0126]

[0127] Correspondingly, P g The relationship between and λ can be described as:

[0128]

[0129] Since the power adjustment range of the virtual generator is limited to To ensure user comfort, the Lagrangian condition P g The relationship between and λ is as follows:

[0130]

[0131] Obviously, comfort and power adjustment range affect the optimal decision of temperature control load output. In the regulation of minute time scale, the internal temperature of temperature control load varies within a limited range. In addition, as the time scale becomes shorter, the forecast error of the operator (RES) will also become smaller accordingly. The day-ahead plan is alleviated by the hour-ahead scheduling of FPP, and the hour-ahead plan is smoothed by the intra-hour scheduling of FPP. For simplicity, it is assumed that the economic compensation of temperature control load regulation only occurs in the scheduling of hourly time scale.

[0132] The S3 specifically includes the following steps:

[0133] S3.1. Optimal FPP Scheduling in Hours

[0134] Before optimizing the scheduling on an hourly time scale, FPP needs to arrange the output power of virtual generators according to the prediction results of RES to smooth the fluctuation of renewable energy output. Under this condition, the optimization goal of FPP is to minimize the control cost of temperature control load. Therefore, the FPP optimization scheduling model on an hourly basis is as follows:

[0135]

[0136]

[0137] Since the model is a non-convex function, the model is appropriately transformed, and the converted FPP optimization scheduling model in hours is as follows:

[0138]

[0139]

[0140] formula In, λ t is the market settlement price at time t, Δt h is the time interval in hours, N is the set of temperature control loads, T is the set of hourly time periods, and When t When it is a segment boundary point, λ t , P g,i,t and M i,t The value of S g and m g To determine λt The introduced variable, R1 is S g The set of R2 is m g The set of r1 is S g The index of r2 is m g The index of .

[0141] S3.2. FPP Optimization Scheduling in Minutes

[0142] In the FPP optimization scheduling process in minutes, the temperature control load is regarded as a virtual battery, so the centralized control potential of the virtual battery in the second stage needs to be evaluated, and the relevant scheduling parameters are calculated as follows:

[0143]

[0144]

[0145]

[0146]

[0147]

[0148]

[0149]

[0150] According to the above parameters, a similar FPP optimization scheduling model based on the minute time scale is as follows:

[0151]

[0152]

[0153] formula In the m is the time slot on the minute time scale, P load is the forecast load in minutes, P w is the wind power in minutes, It is the fluctuation of renewable energy output.

[0154] S3.3. FPP two-stage optimization scheduling based on Wasserstein distance

[0155] The above two-stage optimization scheduling is given as follows:

[0156]

[0157] x is the decision variable of the first stage, X is the constraint of the first stage optimization model, c is the coefficient vector, ξ is the random variable of renewable energy, P is the probability distribution of ξ, and Ω is the fuzzy set of P.

[0158] Α(x,ξ) is expressed as:

[0159]

[0160] formula In the above equation, y is the decision variable of the second stage, q is the coefficient vector, T, H and W are coefficient matrices, and P and Wasserstein distance are used to determine the fuzzy set Ω. When η∈(0, 1), the confidence level will not be lower than η.

[0161]

[0162] formula In the equation, D is the diameter of the probability distribution space of ξ. Obviously, the radius of the Wasserstein ball becomes shorter as S increases, which will reduce the range of fuzzy sets contained in the confidence interval of the Wasserstein ball. Therefore, ξ is restated as:

[0163]

[0164] According to the conversion conditions, the original two-stage optimization model is converted to the following model:

[0165]

[0166]

[0167] It is a linear programming problem. The commercial solver Gurobi can be used to solve the model directly to obtain the optimal temperature adjustment of the temperature control load in the first stage and the optimal power adjustment of the second stage. The temperature control load can be controlled according to the optimization calculation results to achieve the suppression of renewable energy power fluctuations by the temperature control load.

[0168] like Figure 4 , Figure 5 , Figure 6 As shown, the output curves of the system are shown respectively, that is, the output curves of the system before the day, before the hour and within the hour, and the resources within the FPP, the dispatching of the temperature control load through the tradable energy market, and the process of controlling the temperature control load.

[0169] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0170] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A data-driven risk avoidance optimization method for a tradable federal power plant, characterized in that: The steps include: S1. Construct a two-stage FPP optimization scheduling framework to smooth output fluctuations; S2. Construct a tradable energy market mechanism within the FPP; S3. Construct a two-stage FPP optimization scheduling model based on Wasserstein distance and solve it to obtain the optimal scheduling solution; The FPP two-stage optimization model based on multi-time scale coordinated control of temperature-controlled loads is adopted. At the hourly time scale, the temperature-controlled load with temperature adjustment function is simulated as a virtual generator; at the minute time scale, the temperature-controlled load with frequency control function is regarded as a virtual battery. The S1 specifically includes the following steps: S1.

1. Constructing an hourly equivalent model of temperature control load The temperature-controlled load controls its output power through temperature regulation and frequency regulation. According to the physical regulation characteristics of the two methods, they have different time scale regulation effects. In the hourly time scale regulation, the power consumption change of the temperature-controlled load is based on the temperature regulation, which can be maintained for a long time but cannot respond quickly. Therefore, the temperature-controlled load is equivalent to a virtual generator on the hourly time scale, which can be described as: In formulas ①, ②, and ③, P g is the power of the virtual generator, is the minimum value of the virtual generator power, is the maximum value of the virtual generator power, ΔT max is the maximum value of the allowable increment (ΔT) of the internal temperature, and κ is a constant parameter related to the temperature control load; S1.

2. Constructing a minute-scale equivalent model of temperature control load In minute-scale regulation, the power consumption change of the temperature-controlled load is based on frequency control, which can respond to signals quickly but cannot be maintained for a long time. Therefore, the temperature-controlled load is equivalent to a virtual battery on a minute scale, which can be described as: SOC(k+1)=αSOC(k)+βP b (k)+γ⑤ SOC min ≤SOC(k+1)≤SOC max ⑥ In formulas ④, ⑤, and ⑥, P b is the charge and discharge power of the virtual battery, YesP b The minimum value of the virtual battery charge and discharge power, YesP b The maximum value of the virtual battery charge and discharge power, α, β and γ are the virtual battery parameters, SOC is the charging state of the virtual battery, SOC min is the minimum value of SOC, SOC max is the maximum value of SOC; S1.

3. Hourly and minute-scale coupled coordinated control of temperature control load In the minute-scale model of temperature-controlled load, some parameters of the equivalent virtual battery are determined by the preset internal temperature Determined, the virtual battery parameters in formulas ④, ⑤, and ⑥ are: β and γ depend on It is determined in the hourly scheduling of the temperature control load and is described as: In formulas ⑦, ⑧, and ⑨, T o is the external temperature, T max is the maximum internal temperature; The initial SOC value of the virtual battery SOC 0 depending on Described as: The S2 specifically includes the following steps: On a one-hour time horizon, the objective function for customers in the tradable energy market is to minimize costs and maximize comfort levels, described as: formula In the equation, λ is the bid price of the temperature control load in the tradable energy market, λ0 is the electricity price of the temperature control load, is the baseline power of the temperature control load, a is a constant parameter representing customer preference, the larger a is, the higher the user's tolerance for temperature adjustment, aΔT 2 represents the user's temperature adjustment cost, aΔT 2 The smaller it is, the higher the customer's comfort level is. is the electricity cost, λP g It is the compensation of FPP for temperature control load; For simplicity, let λP g > 0, according to the temperature control load hour time scale model, use P g / κ replaces ΔT, so the formula Re-describe as: When the formula When the minimum value is reached, the following conditions must be met: Correspondingly, P g The relationship between and λ is described as: Since the power adjustment range of the virtual generator is limited to To ensure user comfort, the Lagrangian condition P g The relationship between and λ is as follows: Obviously, comfort and power adjustment range affect the optimal decision of temperature-controlled load output. In the regulation of minute time scale, the internal temperature of temperature-controlled load changes within a limited range. As the time scale becomes shorter, the operator's prediction error will also become smaller accordingly, which is alleviated by FPP's hour-ahead scheduling, and the hour-ahead plan is smoothed by FPP's hour-ahead scheduling. The S3 specifically includes the following steps: S3.

1. Optimal FPP Scheduling in Hours Before optimizing the scheduling on an hourly time scale, FPP needs to arrange the output power of virtual generators according to the prediction results of RES to smooth the fluctuation of renewable energy output. The optimization goal of FPP is to minimize the control cost of temperature control load. Therefore, the FPP optimization scheduling model on an hourly basis is as follows: Since the model is a non-convex function, the model is appropriately transformed, and the converted FPP optimization scheduling model in hours is as follows: formula In, λ t is the market settlement price at time t, Δt h is the time interval in hours, N is the set of temperature control loads, T is the set of hourly time periods, and When t When it is a segment boundary point, λ t , P g,i,t and M i,t The value of S g and m g To determine λ t The introduced variable, R1 is S g The set of R2 is m g The set of r1 is S g The index of r2 is m g The index of S3.

2. FPP Optimization Scheduling in Minutes In the FPP optimization scheduling process in minutes, the temperature control load is regarded as a virtual battery, so the centralized control potential of the virtual battery in the second stage needs to be evaluated, and the relevant scheduling parameters are calculated as follows: According to the above parameters, a similar FPP optimization scheduling model based on the minute time scale is as follows: formula In the m is the time slot on the minute time scale, P load is the forecast load in minutes, P w is the wind power in minutes, ξ tm It is the fluctuation of renewable energy output; S3.

3. FPP two-stage optimization scheduling based on Wasserstein distance The above two-stage optimization scheduling is given as follows: x is the decision variable of the first stage, X is the constraint of the optimization model of the first stage, c is the coefficient vector, ξ is the random variable of renewable energy, P is the probability distribution of ξ, and Ω is the fuzzy set of P; Α(x,ξ) is expressed as: formula In the above equation, y is the decision variable of the second stage, q is the coefficient vector, T, H and W are coefficient matrices, and P and Wasserstein distance are used to determine the fuzzy set Ω. When η∈(0,1), the confidence level will not be lower than η. formula In the equation, D is the diameter of the probability distribution space of ξ; obviously, the radius of the Wasserstein ball becomes shorter as S increases, which will reduce the range of fuzzy sets contained in the confidence interval of the Wasserstein ball. Therefore, ξ is restated as: According to the conversion conditions, the original two-stage optimization model is converted to the following model: The model is solved to obtain the optimal temperature adjustment of the temperature control load in the first stage and the optimal power adjustment of the temperature control load in the second stage. The temperature control load is controlled according to the optimization calculation results to achieve the suppression of renewable energy power fluctuations by the temperature control load.

Citation Information

Patent Citations

  • Multi-energy complementary system two-stage optimization scheduling method and system considering source-storage-load cooperation

    AU2020100983A4

  • Virtual power plant optimization trading strategy model based on two-stage stochastic programming

    CN108960510A

  • Method and device for optimizing new energy consumption capability of integrated energy system, and terminal equipment

    CN113268698A