CN114676991B - Multi-energy complementary system optimal scheduling method based on source-load double-side uncertainty - Google Patents

Abstract

A multi-energy complementary system optimization scheduling method based on source load double-side uncertainty comprises the following steps: acquiring the number and the operation parameters of wind, light, water, thermal power units and energy storage in a multi-energy complementary system; acquiring historical data, prediction data and load prediction data of a wind power and photovoltaic system; introducing a complementation coefficient and net load fluctuation thereof, and quantifying the complementation degree; a fuzzy opportunity constraint method is introduced to correct wind power, photovoltaic output and load prediction data; establishing a multi-energy complementary system scheduling model, taking the system economy optimization as an objective function, and considering constraint conditions of each unit; and converting the fuzzy opportunity constraint into a clear equivalent class constraint based on the scheduling model, and solving by adopting a solver to obtain the optimal output condition of each unit. The invention comprehensively considers the influence of the complementary characteristics of various energy sources and the uncertainty of wind, light and load predicted values on the optimal scheduling of the system, and improves the performance in the aspects of energy source consumption, economy and reliability.

Description

Multi-energy complementary system optimal scheduling method based on source-load double-side uncertainty

Technical Field

The invention relates to the field of optimization scheduling research of a multi-energy complementary system, in particular to a multi-energy complementary system optimization scheduling method based on source-load double-side uncertainty.

Background

In recent years, with exhaustion of fossil energy represented by coal, petroleum and natural gas and increasing severity of human ecological environment, in order to cope with challenges of shortage of energy and increasing environmental deterioration, china is participating in the process of global climate control in a more active posture, and the attention of the energy internet in the world is promoted: the preferential development of clean energy sources such as solar energy, wind energy, water energy and the like is an important approach for protecting ecology and reducing fossil energy pollution, and the implementation of multi-energy complementation is a strategic requirement for completing the energy technical revolution by utilizing the natural complementation characteristics of various energy sources.

With the strong support of national policies, clean energy bases such as wind, light, water and the like are continuously expanded, and renewable energy sources are greatly focused and developed. Under the new situation, the power generation and the power consumption in the power system are diversified, the power generation specific gravity of new energy sources represented by wind power and photovoltaic is greatly increased, the installed capacity of the wind power generation in China reaches 28153 kilowatts from 12235 kilowatts in 2015 to 2020 by 2020, and the installed capacity of the photovoltaic is increased from 4352 kilowatts to 25334 kilowatts. By 2020, the wind power and photovoltaic installed capacity accounts for 12.76% and 11.52% of the total installed power supply in China. The rapid increase of the installed capacity of clean energy represented by wind power photovoltaics is a key point of attention on how to realize clean and efficient utilization of energy.

The introduction of the multi-energy complementary system improves the consumption of renewable energy sources, but the intermittence and the difficulty in predictability of the multi-energy complementary system bring great challenges to the optimal scheduling of the multi-energy complementary system. The prediction accuracy of the current uncertainty power supply output and load power output of wind power, photovoltaic and the like is improved through various technologies, but the actual prediction value and the output value still have small errors under the influence of natural weather and unknown factors. However, most of the time when the power grid day-ahead dispatching plan is formulated still adopts the predicted value as the planned value, and if a large deviation occurs, the operation reliability is affected. Therefore, when the analysis of the multi-energy complementary mechanism is carried out, the influence of uncertainty is required to be fully considered, and the method has important significance for energy planning, scheduling strategy formulation and new energy consumption of the power system.

The defects of the optimized scheduling related to the multi-energy complementary system in the prior art are as follows:

1) Most of researches are focused on two energy sources such as wind, light, wind, water and the like aiming at a multi-energy complementary system, and researches on three or more energy sources are less, and researches on the complementary degree of a complementary power generation system are little.

2) At present, the influence of uncertainty factors on a scheduling plan is studied, the foothold is mainly focused on the uncertainty of large-scale intermittent energy source access, and the influence of load-side uncertainty on the scheduling plan is studied less frequently.

3) Aiming at the existing uncertain processing method, the fuzzy opportunity constraint method has the advantages of not depending on sufficient information, and the actual result can be taken into consideration with the advantages of system risk and cost.

Disclosure of Invention

The invention provides a source load double-side uncertainty-based optimal scheduling method for a multi-energy complementary system, which schedules the multi-energy complementary system from the aspects of energy consumption, economy and reliability, and can provide effective technical support and reference for the scheduling of the multi-energy complementary system.

The technical scheme adopted by the invention is as follows:

a multi-energy complementary system optimization scheduling method based on source load double-side uncertainty comprises the following steps:

Step 1: acquiring the number and the operation parameters of wind, light, water, thermal power units and energy storage in a multi-energy complementary system; acquiring historical data, prediction data and load prediction data of a wind power and photovoltaic system;

Step 2: taking the complementary characteristics among multiple energy sources into consideration, introducing a complementary coefficient and the net load fluctuation thereof, and quantifying the complementary degree;

Step 3: aiming at the uncertainty of the predicted values of the power supply side and the load side, a fuzzy opportunity constraint method is introduced, and wind power, photovoltaic output and load prediction data are corrected;

step 4: establishing a multi-energy complementary system scheduling model, taking the system economy optimization as an objective function, and considering constraint conditions of each unit;

step 5: and (3) converting the fuzzy opportunity constraint into a clear equivalent class constraint based on the multi-energy complementary system scheduling model established in the step (4), and solving by adopting a solver to obtain the optimal output condition of each unit.

In the step 1, the constituent elements of the multi-energy complementary system comprise new energy sources and other devices which are introduced;

the introduced new energy sources specifically refer to wind power plants, photovoltaic power stations and hydroelectric units; other devices specifically refer to thermal power plants and energy storage power stations;

The operating parameters of the wind, light, water, thermal power generating units and energy storage concretely comprise the upper and lower output limits of each unit; the climbing rate, the starting and stopping time, the coal consumption coefficient and the energy storage electric quantity, the self-loss rate and the charging and discharging rate of the energy storage power station of the thermal power generating unit.

The historical data of wind power and photovoltaic specifically comprise measured wind speed, atmospheric pressure, active power output by a wind power plant, illumination intensity, an included angle between a photovoltaic panel and the ground and output power of a photovoltaic power station.

The predicted data of wind power and photovoltaic specifically comprises predicted output power of wind power and photovoltaic power stations in a scheduling period (24 h).

The load forecast data of the multi-energy complementary system specifically comprises forecast values of the electric load required by the user in a scheduling period (24 h).

In the step 2, according to the running characteristics of the multi-energy complementary system, complementary coefficients and net load fluctuation are introduced for quantification to more intuitively reflect the complementary degree between energy sources;

In order to fully exert the complementary characteristics among energy sources, the complementary degrees of different power supply combinations are compared, and a complementary coefficient I _C is defined and used for describing the complementary condition of multiple power supply power after superposition;

Wherein: i _C is the complementary coefficient; c ^k is the degree of complementarity; The per unit value of the power change of the energy i in the time scale is used; p _i,k+1 is the actual output value of the energy i; p _i,c is the installed capacity of energy i; n is a scheduling period (24 h); k is the time scale; p _i,k is the output value of the energy source i in the time scale k; a per unit value of the power change of the first energy source in a time scale; a per unit value of the power change of the second energy source in a time scale; The per unit value of the power change of the ith energy source in the time scale;

The complementary coefficient I _C is considered, and the payload fluctuation is also considered, wherein the payload fluctuation is defined as the residual load amount born by the thermal power generating unit after the wind, light and water storage force is subtracted in the system, namely:

P_d,t＝P_L,t-P_W,t-P_V,t-P_H,t-P_st

Wherein: σ is the net load fluctuation; p _d,t is the payload value at time t; p _L,t is the load output value at time t; p _W,t is the power generated by the wind turbine at time t; p _V,t is the generated power of the photovoltaic power station at the time t; p _H,t is the power generated by the hydroelectric generating set at the time t; p _st is the charge and discharge power of the energy storage device at time t; t is a scheduling period (24 h).

In the step 3, wind power, photovoltaic and load output are taken as uncertainty variables, and fuzzy parameters are taken into consideration of uncertainty of predicted values of a power supply side and a load sideSolving uncertainty by adopting a fuzzy opportunity constraint method, and correcting wind power, photovoltaic and load prediction data;

the fuzzy chance constraint method is described as follows:

the fuzzy opportunity constraint means that the constraint condition contains fuzzy parameters, when the method is adopted for optimal scheduling, the scheduling result is allowed to not meet the constraint condition to a certain extent, but the probability of the establishment of the scheduling result is not less than the set confidence level, namely the fuzzy opportunity constraint expression form is as follows:

Wherein x is a decision vector; f (x) is an objective function; ζ is a fuzzy parameter vector; g (x, ζ) is a constraint function; alpha is the confidence level; c _r { } is the trustworthiness of the event;

in order to fully embody the uncertainty of an intermittent power supply and a load, the wind-light output and the load power are used as decision variables, and the upper limit and the lower limit are represented by fuzzy parameters. The invention adopts triangle membership function to express the membership degree of the fuzzy parameter, and the triangle fuzzy parameter can not directly calculate the formula, so the invention adopts a numerical value equivalent method, namely:

is a fuzzy parameter of an intermittent power supply and a load; Predicting a force value for the day-ahead of the intermittent power supply and the load; alpha is the confidence level; p _F1,t is a first membership parameter corresponding to the triangle blur parameter; p _F2,t is a second membership parameter corresponding to the triangle blur parameter; p _F3,t is a third membership parameter corresponding to the triangle blur parameter;

The predicted output before the wind, light and load days is as follows:

Wherein: p _F1、P_F2、P_F3 is a triangle membership parameter; triangle fuzzy numbers corresponding to wind, light and load output respectively; Predicting force values for wind, light and load day ahead; p _fc,t is the predicted demand value of wind, light and load at time t; w _k is a scaling factor, determined from historical data, and 0<w ₁<1,w₂＝1,w₃ >1. Is a fuzzy parameter of an intermittent power supply and a load; predicting a force value for the wind power day before; predicting a force value for the photovoltaic day-ahead; Predicting a force value for the day before load; p _W1,t is a first membership parameter corresponding to the wind power fuzzy parameter; p _W2,t is a second membership parameter corresponding to the wind power fuzzy parameter; p _W3,t is a third membership parameter corresponding to the wind power fuzzy parameter; p _V1,t is a first membership parameter corresponding to the photovoltaic fuzzy parameter; p _V2,t is a second membership parameter corresponding to the photovoltaic fuzzy parameter; p _V3,t is a third membership parameter corresponding to the photovoltaic fuzzy parameter; p _L1,t is a first membership parameter corresponding to the load fuzzy parameter; p _L2,t is a second membership parameter corresponding to the load fuzzy parameter; p _L3,t is a third membership parameter corresponding to the load fuzzy parameter; w ₁ is a first scaling factor corresponding to the triangle blur parameter; w ₂ is a corresponding second scaling factor under the triangle blur parameters; w ₃ is a corresponding third scaling factor under the triangle blur parameters; p _fc,t(w₁,w₂,w₃) is a triangle membership parameter, and is obtained by multiplying a predicted value of wind, light and load and a proportionality coefficient.

In the step 4, a multi-energy complementary system scheduling model based on fuzzy opportunity constraint is established, the model considers complementary characteristics and economic benefits among various energy sources, and an uncertainty influence caused by intermittent energy sources and load predicted values is considered by adopting a fuzzy opportunity constraint method. Taking the optimal economical efficiency as an objective function, considering constraint conditions of the output of each unit, and specifically comprising the following steps:

① : objective function:

considering the running start-stop cost of the thermal power generating unit and the wind and light discarding punishment cost of the system, namely:

min f＝C₁+C₂

Wherein: c ₁ is the running start-stop cost of the thermal power generating unit; c ₂ is the wind and light discarding punishment cost; p _G,j,t is the output value of the thermal power unit j at the time t, and a _j、b_j and c _j respectively correspond to the coal consumption coefficient of the thermal power unit j; s _jt is the start-stop cost of the thermal power unit j; u _j,t is the start-stop state of the thermal power unit j at the time t; c _w and C _v respectively correspond to the wind discarding and light discarding punishment cost; t is a scheduling period (24 h); t is any time; n _G is the number of thermal power units; u _j,t-1 is the start-stop state of the thermal power unit j at the time t-1; triangle fuzzy number for wind power output; p _W,t is the actual output of the wind turbine; Triangle fuzzy number for photovoltaic output; p _V,t is the actual output of the photovoltaic power plant;

② : power balance constraint:

Because wind power, photovoltaic power and load output contain uncertain variables, the power balance under certain conditions is meaningless, and therefore the influence caused by uncertainty needs to be considered when a scheduling plan is formulated, so that the power is approximately balanced on a certain confidence level, namely:

Wherein: p _Sc,t and P _Sd,t respectively correspond to the charge and discharge power of the energy storage device at the time t; c _r is the credibility of the event; alpha is the confidence level; p _H,t is the output value of the hydroelectric generating set at the time t; a fuzzy number which is a load predicted value;

③ : rotating the reserve constraint:

The system rotation reserve constraint contains uncertainty variables, and the uncertainty variables are expressed as follows by adopting fuzzy opportunity constraint:

The maximum output value of the thermal power unit j;

④ : wind power, photovoltaic and hydro-power output constraints:

0≤P_W,t≤P_W,max

0≤P_V,t≤P_V,max

P_H,min≤P_H,t≤P_H,max

wherein: p _W,max and P _V,max are respectively the upper limit of the output of a wind power plant and a photovoltaic power station; p _H,max and P _H,min are maximum and minimum values of the generating power of the hydroelectric generating set;

⑤ : thermal power generating unit constraint:

-r_d,j△t≤P_G,j,t-P_G,j,t-1≤r_u,j△t

Wherein: And The minimum output value and the maximum output value of the thermal power unit j are respectively; t _j,on is the minimum running time of the thermal power unit j; t _j,t-1 is the running time of the thermal power unit j at the time T-1; r _d,j、r_u,j is the upper and lower speed limits of the thermal power unit j when the load is increased or decreased respectively; t _j,off is the minimum downtime of the thermal power unit j; mu _j,t is the start-stop state of the thermal power unit j at the time t; u _j,t-1 is the start-stop state of the thermal power unit j at the time t-1; p _G,j,t-1 is the output value of the thermal power unit j at the time t-1; r _d,j is the lower limit of the climbing rate of the thermal power unit j; r _u,j is the upper limit of the climbing rate of the thermal power unit j; Δt is the run period;

⑥ : energy storage constraint:

u_sc,t+u_sd,t≤1

S_t,min≤S_t≤S_t,max

S_t,T＝S_t,1

Wherein: s _t is the electricity storage quantity at the time t; θ _i is the self-loss rate of the energy storage device; The charging efficiency at the time t; The discharge efficiency at time t; u _sc,t and u _sd,t are respectively the charge and discharge states of the energy storage device; s _t,min and S _t,max are upper and lower limits of the energy storage device capacity; s _t,1 and S _t,T are respectively the electric power values of the stored energy in the initial and final states; p _sc,max is the maximum value of the charging power; p _sd,max is the maximum value of the discharge power, and S _t-1 is the electricity storage amount at time t-1.

In the step 5, according to the multi-energy complementary system scheduling model based on fuzzy opportunity constraint in the step 4, fuzzy parameters and decision variables are separated during solving and converted into clear equivalent class constraint, and then a solver CPLEX is used for solving to obtain the optimal output condition of each unit;

Model equivalence:

According to the uncertain planning theory, when the confidence level alpha is more than or equal to 0.5, converting the power balance constraint and the rotation reserve constraint into clear equivalence classes under the triangular fuzzy parameters, namely:

Clear equivalence class of power balance constraints:

P _L2,t is a second membership parameter corresponding to the load fuzzy parameter; p _W2,t is a second membership parameter corresponding to the wind power fuzzy parameter; p _V2,t is a second membership parameter corresponding to the photovoltaic fuzzy parameter; p _L3,t is a third membership parameter corresponding to the load fuzzy parameter; p _W1,t is a first membership parameter corresponding to the wind power fuzzy parameter; p _V1,t is a first membership parameter corresponding to the photovoltaic fuzzy parameter;

Clear equivalence class of rotation reserve constraints:

Wherein: p _w1,t、P_w2,t、P_V1,t、P_V2,t is a triangle membership parameter of wind power photovoltaic; p _L2,t、P_L3,t is the triangle membership parameter of the load.

Model solving:

After modeling fuzzy constraint processing, the fuzzy scheduling model provided by the invention can be converted into a deterministic model, and a solver CPLEX is adopted to solve, so that the optimal scheduling strategy of the multi-energy complementary system, the output condition and the complementary degree of different unit combinations are obtained.

The invention discloses a source load double-side uncertainty-based optimal scheduling method for a multi-energy complementary system, which has the following technical effects:

1) The invention comprehensively considers the influence of the complementary characteristics of various energy sources and the uncertainty of wind, light and load predicted values on the optimal scheduling of the system, and improves the performance in the aspects of energy source consumption, economy and reliability.

2) The method adopts the fuzzy opportunity constraint method to treat uncertainty of both sides of the source load, and compared with the traditional deterministic model, the method has the advantages that the unit output plan is formulated to be closer to the actual running condition, and the result has more reference value. By selecting different confidence levels to give consideration to system risk and cost, an output scheme with optimal economy can be achieved under the condition of bearing a certain risk.

Drawings

FIG. 1 is a flow chart of the multi-energy complementary system scheduling of the present invention.

FIG. 2 is a diagram of a multi-energy complementary system scheduling model based on fuzzy opportunity constraints.

FIG. 3 is a graph of wind power, photovoltaic and load predictions for the results of the application.

FIG. 4 (1) is a graph of the net load of different power source combinations in the application results;

FIG. 4 (2) is a graph II of the payload of different power source combinations in the application result;

FIG. 4 (3) is a third graph of the payload of different power source combinations in the application results;

Fig. 4 (4) is a graph of the payload four of different power supply combinations in the application result.

FIG. 5 is a graph of unit output and load for each period of time in the application results.

Detailed Description

As shown in fig. 1, the optimal scheduling method of the multi-energy complementary system based on the double-side uncertainty of the source load is implemented, the complementary characteristics among multiple energy sources are considered, the multi-energy complementary power generation system is adopted to promote new energy consumption, meanwhile, the influence of the double uncertainties of the power source side and the load side on the optimal scheduling of the multi-energy complementary system is considered, a multi-energy complementary system scheduling model based on a fuzzy opportunity constraint method is established, the multi-energy complementary system is scheduled in terms of energy consumption, economy and reliability, and the optimal scheduling strategy of the system, the output condition and the complementation degree of different unit combinations are obtained. The method can provide effective technical support and reference opinion for the scheduling of the multi-energy complementary system. The method specifically comprises the following steps:

Step 1: determining the specific composition of a multi-energy complementary system, and acquiring the number and operation parameters of wind power, light, water, thermal power units and stored energy in the system, historical data, forecast data of wind power and photovoltaic and forecast data of system load;

Step 2: taking the complementary characteristics among multiple energy sources into consideration, and introducing a complementary coefficient and payload fluctuation thereof to quantify the complementary degree;

Step 3: aiming at the uncertainty of the predicted values of the power supply side and the load side, a fuzzy opportunity constraint method is introduced to correct wind power, photovoltaic output and load prediction data;

Step 4: and establishing a multi-energy complementary system scheduling model based on fuzzy opportunity constraint, taking the system economy optimization as an objective function, and considering constraint conditions of each unit.

Step 5: according to the multi-energy complementary system model, in order to simplify the solving process of the model, fuzzy opportunity constraints are converted into clear equivalent class constraints, and a commercial solver CPLEX is adopted for solving, so that the optimal output condition of each unit is obtained.

In the step 1, the system is analyzed to determine the constituent elements of the multi-energy complementary system, including the introduced new energy and other equipment, the number and the operation parameters of wind power, photovoltaic, hydroelectric power, thermal power generating units and energy storage devices in the system are obtained, the prediction data and the load prediction data of the wind power and the photovoltaic are input, and the multi-energy complementary system is optimally scheduled.

In the step 2, according to the running characteristics of the multi-energy system, in order to more intuitively reflect the complementation degree between energy sources, a complementation coefficient and net load fluctuation are introduced for quantification, specifically:

In order to fully exert the complementary characteristics among energy sources, the complementary degrees of different power supply combinations are compared, and a complementary coefficient I _C is defined, which describes the complementary condition of multiple power supply power after superposition;

Wherein: i _C is the complementary coefficient; c ^k is the degree of complementarity; The per unit value of the power change of the energy i in the time scale is used; p _i,k+1 is the actual output value of the energy i; p _i,c is the installed capacity of energy i;

The complementary coefficient is considered, and the net load fluctuation is also considered, wherein the net load fluctuation is defined as the residual load amount born by the thermal power generating unit after the wind, light and water storage force is subtracted in the system, namely:

P_d,t＝P_L,t-P_W,t-P_V,t-P_H,t-P_st

wherein: σ is the net load fluctuation; p _d,t is the payload value at time t; p _L,t is the load output value at time t;

P _W,t is the power generated by the wind turbine at time t; p _V,t is the generated power of the photovoltaic power station at the time t;

P _H,t is the power generated by the hydroelectric generating set at the time t; p _st is the charge and discharge power of the energy storage device at time t; in the step 3, wind power, photovoltaic and load output are taken as uncertainty variables in consideration of uncertainty of predicted values of a power source side and a load side, and fuzzy parameters are as follows And solving uncertainty by adopting a fuzzy opportunity constraint method, and correcting wind power, photovoltaic and load prediction data. The method comprises the following steps:

fuzzy chance constraint method description:

the fuzzy opportunity constraint means that the constraint conditions contain fuzzy parameters, when the method is adopted for optimal scheduling, the scheduling result is allowed to not meet the constraint conditions to a certain extent, but the probability of the establishment of the scheduling result is not less than the set confidence level, namely the fuzzy opportunity constraint expression form is that

Wherein x is a decision vector; f (x) is an objective function; ζ is a fuzzy parameter vector; g (x, ζ) is a constraint function; alpha is the confidence level; c _r { } is the trustworthiness of the event.

Uncertainty description:

in order to fully embody the uncertainty of an intermittent power supply and a load, the wind-light output and the load power are used as decision variables, and the upper limit and the lower limit are represented by fuzzy parameters. The invention adopts triangle membership function to express the membership degree of the fuzzy parameter, and the triangle fuzzy parameter can not directly calculate the formula, so the invention adopts a numerical value equivalent method, namely:

The predicted output before the wind, light and load days is as follows:

Wherein: p _F1、P_F2、P_F3 is a triangle membership parameter; triangle fuzzy numbers corresponding to wind, light and load output respectively; Predicting force values for wind, light and load day ahead; p _fc,t is the predicted demand value of wind, light and load at time t; w _k is a scaling factor, determined from historical data, and 0<w ₁<1,w₂＝1,w₃ >1.

In the step 4, the invention establishes a multi-energy complementary system scheduling model based on fuzzy opportunity constraint, as shown in fig. 2, the model considers complementary characteristics and economic benefits among various energy sources, and adopts a fuzzy opportunity constraint method to account for uncertainty influence caused by intermittent energy sources and load predicted values. Taking the optimal economical efficiency as an objective function, and considering constraint conditions of the output of each unit. The method comprises the following steps:

objective function:

The invention considers the running start-stop cost and the wind-abandoning and light-abandoning punishment cost of the thermal power generating unit, namely:

min f＝C₁+C₂

Wherein: c ₁ is the running start-stop cost of the thermal power generating unit; c ₂ is the wind and light discarding punishment cost; p _G,j,t is the output value of the thermal power unit j at the time t, and a _j、b_j and c _j respectively correspond to the coal consumption coefficient of the thermal power unit j; s _jt is the start-stop cost of the thermal power unit j; u _j,t is the start-stop state of the thermal power unit j at the time t; c _w and C _v respectively correspond to the wind discarding and light discarding punishment cost;

Power balance and rotational reserve constraints:

1) The power balance constraint is that the wind power, the photovoltaic power and the load output contain uncertain variables, so that the power balance under the determined condition is meaningless, and therefore, the influence caused by uncertainty is considered when a scheduling plan is formulated, so that the power is approximately balanced on a certain confidence level, namely:

wherein: p _Sc,t and P _Sd,t respectively correspond to the charge and discharge power of the energy storage device at the time t

2) Rotation reserve constraint

The system rotation reserve constraint contains uncertainty variables, and the uncertainty variables are expressed as follows by adopting fuzzy opportunity constraint:

element operation constraints:

1) Wind power, photovoltaic and hydro-electric power output constraints

0≤P_W,t≤P_W,max

0≤P_V,t≤P_V,max

P_H,min≤P_H,t≤P_H,max

Wherein: p _W,max and P _V,max are respectively the upper limit of the output of a wind power plant and a photovoltaic power station; p _H,max and P _H,min are maximum and minimum values of the generating power of the hydroelectric generating set;

2) Thermal power generating unit constraint

-r_d,j△t≤P_G,j,t-P_G,j,t-1≤r_u,j△t

Wherein: And The minimum output value and the maximum output value of the thermal power unit j are respectively; t _j,on is the minimum running time of the thermal power unit j; t _j,t-1 is the running time of the thermal power unit j at the time T-1; r _d,j、r_u,j is the upper and lower speed limits of the thermal power unit j when the load is increased or decreased respectively; t _j,off is the minimum downtime of the thermal power unit j;

3) Energy storage constraint

u_sc,t+u_sd,t≤1

S_t,min≤S_t≤S_t,max

S_t,T＝S_t,1

Wherein: s _t is the electricity storage quantity at the time t; θ _i is the self-loss rate of the energy storage device; The charging efficiency at the time t; The discharge efficiency at time t; u _sc,t and u _sd,t are respectively the charge and discharge states of the energy storage device; s _t,min and S _t,max are upper and lower limits of the energy storage device capacity; s _t,1 and S _t,T are respectively the electric power values of the stored energy in the initial and final states; p _sc,max is the maximum value of the charging power; p _sd,max is the maximum value of the discharge power;

In the step 5, according to the multi-energy complementary system scheduling model based on the fuzzy opportunity constraint, when solving the fuzzy opportunity constraint problem, the fuzzy parameters and the decision variables are separated and converted into clear equivalence classes, and then the solution is carried out by adopting a traditional method. The method comprises the following steps:

Model equivalence:

According to the uncertain planning theory, when the confidence level alpha is more than or equal to 0.5, converting the power balance constraint and the rotation reserve constraint into clear equivalence classes under the triangular fuzzy parameters, namely:

Clear equivalence class of power balance constraints:

Clear equivalence class of rotation reserve constraints:

Wherein: p _w1,t、P_w2,t、P_V1,t、P_V2,t is a triangle membership parameter of wind power photovoltaic; p _L2,t、P_L3,t is the triangle membership parameter of the load.

Model solving:

After modeling fuzzy constraint processing, the fuzzy scheduling model provided by the invention can be converted into a deterministic model, and a commercial solver CPLEX is adopted for solving, so that the optimal scheduling strategy of the multi-energy complementary system, the output condition and the complementary degree of different unit combinations are obtained.

Verification example:

(1) Basic data and parameters:

the system comprises 5 thermal power units shown in table 1, 1 wind power plant with total capacity of 300MW, 1 photovoltaic power station with total capacity of 50MW and 1 hydropower station with total capacity of 100 MW; the maximum energy storage electric quantity S _t in the energy storage device is 400MWh, the maximum charge and discharge power is 100MW, and the charge and discharge efficiency is 0.95.

TABLE 1 thermal power generating unit related data

A typical summer day of 24 hours is selected as a study object, wind power, photovoltaic and load prediction curves are shown in figure 3, fuzzy membership parameters corresponding to wind power, photovoltaic and load prediction values are shown in table 2, and the penalty cost of wind discarding and light discarding is 233 yuan (/ MWh).

TABLE 2 fuzzy parameters

(2) Multi-energy complementary system scheduling analysis:

According to the operation characteristics of the multi-energy system, the wind power and the photovoltaic, the wind power and the water power, the water power and the photovoltaic, the wind power, the photovoltaic and the water power are found to have the difference and also have the strong complementary characteristics, the wind power, the photovoltaic and the water power are respectively analyzed from the angles of complementation of each other and complementation of each other, the uncertainty of wind power and load is considered, and when the confidence level is 0.9, the dispatching result is shown in the following table 3.

Table 3 scheduling results under different combinations

As can be seen from table 3: the scheduling result of the wind-solar-water complementary power generation system established by the invention is optimal, namely, the corresponding complementary coefficient, the net load fluctuation and the power peak-valley difference of the thermal power unit are minimum.

The net load curves for the different power source combinations are shown in fig. 4 (1) to 4 (4). As can be seen from fig. 4 (1) to fig. 4 (4): when wind, light and water complement operation is performed, the overall curve is smooth, the net load curve tends to be straight in the load valley period (1:00-3:00), the peak value is obviously reduced in the peak period (10:00-14:00), 289.2MW is reduced compared with the initial load peak value, the peak-valley difference of the thermal power unit is 645.04MW, the net load fluctuation is 164.62, and the good peak clipping and valley filling purposes are achieved. The complementary coefficient of the wind-solar complementary analysis is relatively high and is 0.0197 through the complementary analysis of the wind-solar complementary analysis shown in the table 3 and the fig. 4 (1) to the fig. 4 (4), which shows that the complementarity is relatively low in the situation compared with other complementary situations; the wind-solar hybrid power generating unit has the advantages that the wind-solar hybrid power generating unit is high in peak period and low in valley period, so that the fluctuation is large, the peak-valley difference is 810.4MW, and the economic efficiency is poor due to frequent adjustment of the output force of the thermal power generating unit for suppressing the fluctuation.

When the system only considers wind-water complementation, the net load fluctuation curve is relatively smooth, the peak-valley difference is 799.9MW, the complementation coefficient is relatively small and is 0.0151, and the total capacity of the wind-fire installation is 400MW, so that the thermal power unit can reduce certain output at the moment, and the economical efficiency is better than 48.44MW.

When the system only considers light water complementation, the net load curve is consistent with the fluctuation of the load curve, the peak-valley difference is 713.28MW, and the complementation coefficient is well 0.0174. Since the installed capacity of the photo-water is relatively small, most thermal power units are used to cope with the load output, resulting in the worst economical efficiency compared with other situations.

Therefore, the wind, light, water and fire storage multi-energy complementary system adopted by the invention greatly improves the utilization rate of new energy and reduces the output of the thermal power unit.

(3) Source load double-side uncertain scheduling result:

to verify the effectiveness of the fuzzy opportunity constraint model proposed by the present invention, two scenarios were used for comparison, and the confidence level in the model was set to 0.9, as shown in table 4.

Scenario 1: the invention adopts the traditional scheduling mode, namely a deterministic model of system power balance, and simultaneously sets the reserve capacity to 10% of load prediction and 5% of wind-light output for coping with uncertainty of source load.

Scenario 2: the fuzzy opportunity constraint method provided by the invention is adopted to cope with uncertainty of intermittent power supply and load prediction.

Table 4 scheduling results in different scenarios

As can be seen from table 4: the model provided by the invention has the advantages of optimal scheduling result, 47.72 ten thousand yuan for the running start-stop cost of the thermal power unit, 7.76% for the waste wind and waste light rate and 56.21 ten thousand yuan for the economic cost of the system. Compared with a deterministic scheduling model, the thermal power generating unit operation start-stop cost in the scheduling model considering uncertainty is reduced by 10.47 ten thousand yuan, the system economic cost is reduced by 33.4%, the standby capacity is reduced by 1118.03MW, and the wind-solar energy absorption rate is improved by 19.54%. Mainly because the traditional deterministic model is established when the error is zero, the rotation reserve capacity required for ensuring the normal operation of the system is larger, so that the economic cost of the system is increased. The fuzzy opportunity constraint provided by the invention is that the power balance is satisfied under a certain confidence level, and the model has wider optimization range and better economy.

When the confidence level is 0.9, the unit output and load curves of each period are shown in fig. 5. As can be seen from fig. 5: the load curve is sometimes higher than the unit output, and the higher part is discharged by the energy storage device to meet the load requirement. When the load is in the peak period (10; 00-13:00), the output of the thermal power generating units 1, 3 and 4 reaches the maximum value; when the load is in the second peak period (18:00-21:00), the thermal power generating units 1 and 5 reach full firing; the output of each unit already contains the spare capacity for wind, light and load prediction uncertainty, and for 5 thermal power units, as the power of the 1 st unit is close to that of the 2 nd unit, the output of the 2 nd thermal power unit is always 0 on the premise of optimal economy, and the load demand is met by adjusting the output of other 4 units. The photovoltaic output is not shown in the figure, since it is relatively small.

To sum up: the invention comprehensively considers the influence of the complementary characteristics of various energy sources and the uncertainty of wind, light and load predicted values on the optimal scheduling of the system, establishes a multi-energy complementary system scheduling model based on fuzzy opportunity constraint, and improves the energy source consumption, economy and reliability. The application result shows that the complementation degree is best when the wind, light and water are combined, and the complementation coefficient is 0.0098 at minimum; compared with the traditional deterministic model, the economic cost of the model system is reduced by 33.4%, and the wind and light consumption is improved by 19.54%.

Claims (4)

1. The optimal scheduling method of the multi-energy complementary system based on the source load double-side uncertainty is characterized by comprising the following steps:

Step 1: acquiring the number and the operation parameters of wind, light, water, thermal power units and energy storage in a multi-energy complementary system; acquiring historical data, prediction data and load prediction data of a wind power and photovoltaic system;

Step 2: taking the complementary characteristics among multiple energy sources into consideration, introducing a complementary coefficient and the net load fluctuation thereof, and quantifying the complementary degree;

Step 3: aiming at the uncertainty of the predicted values of the power supply side and the load side, a fuzzy opportunity constraint method is introduced, and wind power, photovoltaic output and load prediction data are corrected;

step 4: establishing a multi-energy complementary system scheduling model, taking the system economy optimization as an objective function, and considering constraint conditions of each unit;

Step 5: based on the multi-energy complementary system scheduling model established in the step 4, the fuzzy opportunity constraint is converted into a clear equivalent class constraint, and a solver is adopted to solve, so that the optimal output condition of each unit is obtained;

in the step2, a complementary coefficient I _C is defined to describe the complementary situation of the superimposed power supplies;

Wherein: i _C is the complementary coefficient; c ^k is the degree of complementarity; The per unit value of the power change of the energy i in the time scale is used; p _i,k+1 is the actual output value of the energy i; p _i,c is the installed capacity of energy i; n is a scheduling period (24 h); k is the time scale; p _i,k is the output value of the energy source i in the time scale k; a per unit value of the power change of the first energy source in a time scale; a per unit value of the power change of the second energy source in a time scale; The per unit value of the power change of the ith energy source in the time scale;

The complementary coefficient I _C is considered, and the payload fluctuation is also considered, wherein the payload fluctuation is defined as the residual load amount born by the thermal power generating unit after the wind, light and water storage force is subtracted in the system, namely:

P_d,t＝P_L,t-P_W,t-P_V,t-P_H,t-P_st

Wherein: σ is the net load fluctuation; p _d,t is the payload value at time t; p _L,t is the load output value at time t; p _W,t is the power generated by the wind turbine at time t; p _V,t is the generated power of the photovoltaic power station at the time t; p _H,t is the power generated by the hydroelectric generating set at the time t; p _st is the charge and discharge power of the energy storage device at time t; t is a scheduling period;

In the step 3, wind power, photovoltaic and load output are taken as uncertainty variables, and fuzzy parameters are taken into consideration of uncertainty of predicted values of a power supply side and a load side Solving uncertainty by adopting a fuzzy opportunity constraint method, and correcting wind power, photovoltaic and load prediction data;

in the step 4, a multi-energy complementary system scheduling model based on fuzzy opportunity constraint is established, the model takes economical efficiency optimization as an objective function, and constraint conditions of output of each unit are considered, and the method specifically comprises the following steps:

① : objective function:

considering the running start-stop cost of the thermal power generating unit and the wind and light discarding punishment cost of the system, namely:

minf＝C₁+C₂

Wherein: c ₁ is the running start-stop cost of the thermal power generating unit; c ₂ is the wind and light discarding punishment cost; p _G,j,t is the output value of the thermal power unit j at the time t, and a _j、b_j and c _j respectively correspond to the coal consumption coefficient of the thermal power unit j; s _jt is the start-stop cost of the thermal power unit j; u _j,t is the start-stop state of the thermal power unit j at the time t; c _w and C _v respectively correspond to the wind discarding and light discarding punishment cost; t is a scheduling period; t is any time; n _G is the number of thermal power units; u _j,t-1 is the start-stop state of the thermal power unit j at the time t-1; triangle fuzzy number for wind power output; p _W,t is the actual output of the wind turbine; Triangle fuzzy number for photovoltaic output; p _V,t is the actual output of the photovoltaic power plant;

② : power balance constraint:

Because wind power, photovoltaic power and load output contain uncertain variables, the power balance under certain conditions is meaningless, and therefore the influence caused by uncertainty needs to be considered when a scheduling plan is formulated, so that the power is approximately balanced on a certain confidence level, namely:

Wherein: p _Sc,t and P _Sd,t respectively correspond to the charge and discharge power of the energy storage device at the time t; c _r is the credibility of the event; alpha is the confidence level; p _H,t is the output value of the hydroelectric generating set at the time t; a fuzzy number which is a load predicted value;

③ : rotating the reserve constraint:

The system rotation reserve constraint contains uncertainty variables, and the uncertainty variables are expressed as follows by adopting fuzzy opportunity constraint:

The maximum output value of the thermal power unit j;

④ : wind power, photovoltaic and hydro-power output constraints:

0≤P_W,t≤P_W,max

0≤P_V,t≤P_V,max

P_H,min≤P_H,t≤P_H,max

wherein: p _W,max and P _V,max are respectively the upper limit of the output of a wind power plant and a photovoltaic power station; p _H,max and P _H,min are maximum and minimum values of the generating power of the hydroelectric generating set;

⑤ : thermal power generating unit constraint:

-r_d,j△t≤P_G,j,t-P_G,j,t-1≤r_u,j△t

Wherein: And The minimum output value and the maximum output value of the thermal power unit j are respectively; t _j,on is the minimum running time of the thermal power unit j; t _j,t-1 is the running time of the thermal power unit j at the time T-1; r _d,j、r_u,j is the upper and lower speed limits of the thermal power unit j when the load is increased or decreased respectively; t _j,off is the minimum downtime of the thermal power unit j; mu _j,t is the start-stop state of the thermal power unit j at the time t; u _j,t-1 is the start-stop state of the thermal power unit j at the time t-1; p _G,j,t-1 is the output value of the thermal power unit j at the time t-1; r _d,j is the lower limit of the climbing rate of the thermal power unit j; r _u,j is the upper limit of the climbing rate of the thermal power unit j; Δt is the run period;

⑥ : energy storage constraint:

u_sc,t+u_sd,t≤1

S_t,min≤S_t≤S_t,max

S_t,T＝S_t,1

Wherein: s _t is the electricity storage quantity at the time t; θ _i is the self-loss rate of the energy storage device; The charging efficiency at the time t; The discharge efficiency at time t; u _sc,t and u _sd,t are respectively the charge and discharge states of the energy storage device; s _t,min and S _t,max are upper and lower limits of the energy storage device capacity; s _t,1 and S _t,T are respectively the electric power values of the stored energy in the initial and final states; p _sc,max is the maximum value of the charging power; p _sd,max is the maximum value of the discharge power, and S _t-1 is the electricity storage amount at time t-1.

2. The optimal scheduling method for the multi-energy complementary system based on source load double-side uncertainty as claimed in claim 1, wherein the method is characterized by comprising the following steps: the fuzzy chance constraint method is described as follows:

the fuzzy opportunity constraint means that the constraint condition contains fuzzy parameters, when the method is adopted for optimal scheduling, the scheduling result is allowed to not meet the constraint condition to a certain extent, but the probability of the establishment of the scheduling result is not less than the set confidence level, namely the fuzzy opportunity constraint expression form is as follows:

Wherein x is a decision vector; f (x) is an objective function; ζ is a fuzzy parameter vector; g (x, ζ) is a constraint function; alpha is the confidence level; c _r { } is the trustworthiness of the event;

In order to fully embody the uncertainty of an intermittent power supply and a load, wind-light output and load power are used as decision variables, and the upper limit and the lower limit are represented by fuzzy parameters; the membership of the fuzzy parameter is expressed by adopting a triangle membership function, and as the fuzzy parameter of the triangle cannot be directly calculated by a formula, a numerical value equivalent method is adopted, namely:

is a fuzzy parameter of an intermittent power supply and a load; Predicting a force value for the day-ahead of the intermittent power supply and the load; alpha is the confidence level; p _F1,t is a first membership parameter corresponding to the triangle blur parameter; p _F2,t is a second membership parameter corresponding to the triangle blur parameter; p _F3,t is a third membership parameter corresponding to the triangle blur parameter;

The predicted output before the wind, light and load days is as follows:

Wherein: p _F1、P_F2、P_F3 is a triangle membership parameter; triangle fuzzy numbers corresponding to wind, light and load output respectively; Predicting force values for wind, light and load day ahead; p _fc,t is the predicted demand value of wind, light and load at time t; w _k is a scaling factor, determined from historical data, and 0<w ₁<1,w₂＝1,w₃ >1; is a fuzzy parameter of an intermittent power supply and a load; predicting a force value for the wind power day before; predicting a force value for the photovoltaic day-ahead; Predicting a force value for the day before load; p _W1,t is a first membership parameter corresponding to the wind power fuzzy parameter; p _W2,t is a second membership parameter corresponding to the wind power fuzzy parameter; p _W3,t is a third membership parameter corresponding to the wind power fuzzy parameter; p _V1,t is a first membership parameter corresponding to the photovoltaic fuzzy parameter; p _V2,t is a second membership parameter corresponding to the photovoltaic fuzzy parameter; p _V3,t is a third membership parameter corresponding to the photovoltaic fuzzy parameter; p _L1,t is a first membership parameter corresponding to the load fuzzy parameter; p _L2,t is a second membership parameter corresponding to the load fuzzy parameter; p _L3,t is a third membership parameter corresponding to the load fuzzy parameter; w ₁ is a first scaling factor corresponding to the triangle blur parameter; w ₂ is a corresponding second scaling factor under the triangle blur parameters; w ₃ is a corresponding third scaling factor under the triangle blur parameters; p _fc,t(w₁,w₂,w₃) is a triangle membership parameter, and is obtained by multiplying a predicted value of wind, light and load and a proportionality coefficient.

3. The optimal scheduling method for the multi-energy complementary system based on source load double-side uncertainty as claimed in claim 1, wherein the method is characterized by comprising the following steps: the step 5 specifically comprises the following steps:

Model equivalence:

According to the uncertain planning theory, when the confidence level alpha is more than or equal to 0.5, converting the power balance constraint and the rotation reserve constraint into clear equivalence classes under the triangular fuzzy parameters, namely:

Clear equivalence class of power balance constraints:

P _L2,t is a second membership parameter corresponding to the load fuzzy parameter; p _W2,t is a second membership parameter corresponding to the wind power fuzzy parameter; p _V2,t is a second membership parameter corresponding to the photovoltaic fuzzy parameter; p _L3,t is a third membership parameter corresponding to the load fuzzy parameter; p _W1,t is a first membership parameter corresponding to the wind power fuzzy parameter; p _V1,t is a first membership parameter corresponding to the photovoltaic fuzzy parameter;

Clear equivalence class of rotation reserve constraints:

wherein: p _w1,t、P_w2,t、P_V1,t、P_V2,t is a triangle membership parameter of wind power photovoltaic; p _L2,t、P_L3,t is a triangle membership parameter of the load;

Model solving:

after modeling fuzzy constraint processing, the fuzzy scheduling model can be converted into a deterministic model, and a solver CPLEX is adopted to solve the fuzzy scheduling model, so that the optimal scheduling strategy of the multi-energy complementary system, the output condition and the complementary degree of different unit combinations are obtained.

4. The multi-energy complementary system scheduling model based on fuzzy opportunity constraint is characterized in that: the model takes the optimal economical efficiency as an objective function, considers constraint conditions of the output of each unit, and specifically comprises the following steps:

① : objective function:

considering the running start-stop cost of the thermal power generating unit and the wind and light discarding punishment cost of the system, namely:

minf＝C₁+C₂

Wherein: c ₁ is the running start-stop cost of the thermal power generating unit; c ₂ is the wind and light discarding punishment cost; p _G,j,t is the output value of the thermal power unit j at the time t, and a _j、b_j and c _j respectively correspond to the coal consumption coefficient of the thermal power unit j; s _jt is the start-stop cost of the thermal power unit j; u _j,t is the start-stop state of the thermal power unit j at the time t; c _w and C _v respectively correspond to the wind discarding and light discarding punishment cost; t is a scheduling period; t is any time; n _G is the number of thermal power units; u _j,t-1 is the start-stop state of the thermal power unit j at the time t-1; triangle fuzzy number for wind power output; p _W,t is the actual output of the wind turbine; Triangle fuzzy number for photovoltaic output; p _V,t is the actual output of the photovoltaic power plant;

② : power balance constraint:

Because wind power, photovoltaic power and load output contain uncertain variables, the power balance under certain conditions is meaningless, and therefore the influence caused by uncertainty needs to be considered when a scheduling plan is formulated, so that the power is approximately balanced on a certain confidence level, namely:

Wherein: p _Sc,t and P _Sd,t respectively correspond to the charge and discharge power of the energy storage device at the time t; c _r is the credibility of the event; alpha is the confidence level; p _H,t is the output value of the hydroelectric generating set at the time t; a fuzzy number which is a load predicted value;

③ : rotating the reserve constraint:

The system rotation reserve constraint contains uncertainty variables, and the uncertainty variables are expressed as follows by adopting fuzzy opportunity constraint:

The maximum output value of the thermal power unit j;

④ : wind power, photovoltaic and hydro-power output constraints:

0≤P_W,t≤P_W,max

0≤P_V,t≤P_V,max

P_H,min≤P_H,t≤P_H,max

wherein: p _W,max and P _V,max are respectively the upper limit of the output of a wind power plant and a photovoltaic power station; p _H,max and P _H,min are maximum and minimum values of the generating power of the hydroelectric generating set;

⑤ : thermal power generating unit constraint:

-r_d,j△t≤P_G,j,t-P_G,j,t-1≤r_u,j△t

Wherein: And The minimum output value and the maximum output value of the thermal power unit j are respectively; t _j,on is the minimum running time of the thermal power unit j; t _j,t-1 is the running time of the thermal power unit j at the time T-1; r _d,j、r_u,j is the upper and lower speed limits of the thermal power unit j when the load is increased or decreased respectively; t _j,off is the minimum downtime of the thermal power unit j; mu _j,t is the start-stop state of the thermal power unit j at the time t; u _j,t-1 is the start-stop state of the thermal power unit j at the time t-1; p _G,j,t-1 is the output value of the thermal power unit j at the time t-1; r _d,j is the lower limit of the climbing rate of the thermal power unit j; r _u,j is the upper limit of the climbing rate of the thermal power unit j; Δt is the run period;

⑥ : energy storage constraint:

u_sc,t+u_sd,t≤1

S_t,min≤S_t≤S_t,max

S_t,T＝S_t,1

Wherein: s _t is the electricity storage quantity at the time t; θ _i is the self-loss rate of the energy storage device; The charging efficiency at the time t; The discharge efficiency at time t; u _sc,t and u _sd,t are respectively the charge and discharge states of the energy storage device; s _t,min and S _t,max are upper and lower limits of the energy storage device capacity; s _t,1 and S _t,T are respectively the electric power values of the stored energy in the initial and final states; p _sc,max is the maximum value of the charging power; p _sd,max is the maximum value of the discharge power, and S _t-1 is the electricity storage amount at time t-1.