CN110880772A - Electricity selling company response power grid control method based on industrial park load aggregation - Google Patents

Abstract

The invention belongs to the power system operation and control technology, and particularly relates to an electric power selling company response power grid control method based on industrial park load aggregation, which models the power regulation characteristic of a typical industrial park load, provides a proportional aggregation mode in which a user reports a regulation and control range in priority by combining a load regulation willingness and a limit regulation range, and establishes an aggregation control strategy of the electric power selling company on an industrial park; considering load production benefits, establishing control cost models of various types of loads, and simultaneously providing a coordination control strategy for the power selling company to participate in demand response in multiple parks; aiming at the scene of wind power fluctuation of a large power grid, the stability of the power grid is maintained through load response power fluctuation, and the correctness of the provided control strategy is verified. For a power grid company, the evaluation indexes of low wind abandoning and light abandoning rates are facilitated to be realized; for a new energy power generation party, the new energy access rate is improved, and the total economic benefit of load-network-source is improved on the whole.

Description

A response grid control method for power sales companies based on load aggregation in industrial parks

technical field

The invention belongs to the technical field of power system operation and control, and in particular relates to a method for controlling a power sales company's response to a power grid based on load aggregation in an industrial park.

Background technique

As the energy strategy gradually moves towards large-scale new energy power generation and networking, the impact of the uncertainty of new energy on the safe and stable operation of the power grid has become an urgent problem that needs to be solved. Considering economy and practicability, it is of great significance to use load demand side response, a new type of flexible resource, to participate in power grid peak regulation and frequency regulation, and to consume new energy. Among them, industrial loads, especially high-energy-consuming loads such as electrolysis and arc loads, are the main types. They have the characteristics of high power consumption and stable power, and have huge potential for power regulation. For the form of networking between industrial parks and large power grid systems, how to carry out load demand response and formulate a coordinated control method for multiple loads is an urgent problem to be solved.

At the same time, the electricity sales side business has gradually expanded, and electricity sales companies have also faced continuous challenges under the market opportunities. It is not enough to simply earn the difference in selling electricity. Expanding such as participating in grid auxiliary services and building comprehensive energy services is the core problem faced by electricity selling companies.

To sum up, it is a good model for multi-party participation and mutual benefit to study the aggregation strategy of power sales companies for industrial park loads to participate in demand response and assist the grid to stabilize power fluctuations caused by new energy access. Electricity sales companies participate in grid auxiliary services through aggregation control strategies to expand business content; industrial loads can obtain certain economic compensation when production conditions permit; grid companies can ensure stable and reliable operation of the grid at a lower cost; new energy power generation Fang has increased the access rate of new energy; on the whole, the overall economic benefit of the Netherlands-grid-source has been improved.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a power sales company by aggregating the load of the industrial park to obtain the load aggregation characteristic ΔP _Σ,t -n _t of the industrial park and the control cost model F _IP -ΔP _Σ of the industrial park, and combine the adjustable capacity and sales The power company's regulation cost is reported to the power grid, and then according to the grid regulation command, the power regulation command is sent to each industrial park and its different types of loads by the coordination method of bidding for the best, and finally the method of smoothing power fluctuations is realized.

In order to achieve the above-mentioned purpose, the technical solution adopted in the present invention is a method for controlling a response grid of an electricity sales company based on load aggregation in an industrial park, comprising the following steps:

Step 1. According to the power characteristics of electrolytic aluminum, submerged arc furnace, and polysilicon loads, normalize the load power of the industrial park, and analyze the power theoretical adjustment boundary of each load;

Step 2. Combining the load regulation willingness and the limit regulation range, a proportional aggregation model is proposed that prioritizes the regulation range reported by users, and the aggregation control strategy of the electricity sales company for the industrial park is established;

Step 3. According to the aggregation model, establish the response strategy of the electricity sales company to the power grid control command, so as to realize the rational distribution and aggregation control of the load power of the industrial park;

Step 4. Analyze the impact of the aggregation control strategy on the benefit of load production, establish a load control cost model, and combine the aggregation model to establish a load control cost model for the industrial park;

Step 5. Based on the load control cost model of the industrial park, a bidding and optimal coordination strategy of the electricity sales company for multiple industrial parks is proposed to realize the coordinated distribution and control of the power of the multiple parks.

In the above-mentioned control method for power sales companies to respond to power grids based on load aggregation in industrial parks, the specific implementation of step 1 includes:

Step 1.1. The load power characteristics of electrolytic aluminum are as follows:

Wherein, P _AL is the power of electrolytic aluminum, V _B is the DC bus voltage of the electrolytic cell, I _d is the direct current of the electrolytic cell, R _EC is the equivalent resistance of the electrolytic cell in series, and E is the equivalent potential of the electrolytic cell;

The relationship between the electrolytic aluminum DC bus voltage V _B and the high voltage bus voltage V _AL-AH connected to the electrolytic aluminum load is:

In the formula, L _SR is the equivalent value of the saturable reactor, V _AL-AH is the high-voltage bus voltage, and ω is the voltage angular frequency;

Considering that the electrolytic aluminum load is connected to an on-load voltage regulating transformer, its transformation ratio has a total of m ₁ level, then changing its transformation ratio k _AL can realize m ₁ level adjustment:

Assuming that the adjustment range of the equivalent value L _SR of the saturable reactor is [L _SR ^min , L _SR ^max ] when the production requirements are met, then by formula (2), the electrolytic aluminum DC busbar under the corresponding strain ratio k _AL-i can be solved The voltage drop V _B adjustment range is:

From the formula (1), the maximum adjustable power range of the electrolytic aluminum load can be obtained as follows:

Step 1.2. The power characteristics of the submerged arc furnace load are as follows:

Among them, P _SAF and Q _SAF are the active power and reactive power of the submerged arc furnace respectively, U _SAF is the low-voltage side voltage of the submerged arc furnace, and R _line and X _line are the equivalent resistance and equivalent reactance of the short grid;

It is assumed that the equivalent impedance of the arc is represented by the static arc resistance R _arc and the static arc reactance X _arc in a certain time section;

The low voltage side voltage U _SAF of the submerged arc furnace is obtained from the high voltage busbar voltage U _SAF-AH of the submerged arc furnace through the special transformer for the submerged arc furnace:

U _SAF = U _SAF-AH /k _SAF (9)

In the formula, k _SAF is the special transformer ratio of submerged arc furnace;

By fitting the actual production data, the arc impedance relationship of the submerged arc furnace is obtained:

X _arc =a ₁ R _arc ² +a ₂ R _arc +a ₃ (10)

Among them, a ₁ , a ₂ , and a ₃ are arc impedance fitting coefficients, which are constants;

When the constant impedance voltage regulation method is used to adjust the power of the submerged arc furnace, the transformation ratio of the special transformer for the load of the submerged arc furnace is set to have a total of _m2 levels; then by changing the transformation ratio of the special transformer for the submerged arc furnace, its voltage has a total of _m2 level adjustment scope:

The limited range of arc static resistance R _arc :

R _arc ^min ≤R _arc ≤R _arc ^max (12)

电弧部分功率因数存在下限

则电弧部分功率因数

的限制为：Arc part power factor upper limit

There is a lower limit for the power factor of the arc part

Then the power factor of the arc part

The limits are:

which is

in,

At the same time, each smelting stage should meet the minimum power constraint:

P _SAF,t ≥P _SAF,t ^min (15)

In the formula, P _SAF,t ^min is the minimum power of the submerged arc furnace;

Assuming that the rated voltage of the low-voltage side of the submerged arc furnace is U _SAF,N , by formulas (7), (8), (12), (15), when the constant voltage and impedance power adjustment method is adopted, the submerged heat in each smelting stage The adjustment ranges of the active power P _SAF,t and reactive power Q _SAF,t of the furnace are as follows:

c ₁ ≤P _SAF,t ≤ c ₂ (16)

d ₁ ≤Q _SAF,t ≤ d ₂ (17)

in:

将此时的功率调节范围转化为等效低压侧电压变化范围，即The power characteristic parameters of the submerged arc furnace under rated load operation are:

Convert the power adjustment range at this time into the equivalent low-voltage side voltage variation range, that is

From equations (10) and (18), when the impedance-voltage coordinated adjustment method is adopted, the voltage variation range of the low-voltage side of the submerged arc furnace load is:

Step 1.3. The polysilicon load power satisfies the following electrical relationship:

In the formula, P _PCS is the total AC power of the polysilicon load, U _val is the single-phase voltage of the polysilicon load, and R _PCS is the single-phase resistance of the polysilicon rod;

The energy conversion relationship in the production process of a polycrystalline silicon rod in the Δt time is as follows:

对应于(ΔQ_out2+ΔQ_out3)，其中，η、K、L、T_out分别为反应吸热占比、硅棒与混合气体总传热系数、硅棒总长度、底盘及炉璧表面等效温度，均为常数；r为多晶硅硅棒半径，短时间内可视r为常量；In the formula, P _PCS is the total AC power of the polysilicon load, ΔQ _out1 represents the heat used to heat the reaction gas, and v ₁ ·Δt ·s ₁ ·ρ _g ·c·(T _x -T _g ) can be obtained from the gas specific heat capacity formula, Among them, v ₁ , s ₁ , ρ _g , c, and T _g are the inlet velocity, inlet area, mixed gas density, mixed gas specific heat capacity, and inlet temperature, all of which are constants; Tx is the surface temperature of the silicon rod, ΔQ _out2 and ΔQ _out3 represent the sustaining endothermic reaction and the heat dissipated through the bottom and walls of the reduction furnace due to thermal radiation, respectively

Corresponding to (ΔQ _out2 +ΔQ _out3 ), where η, K, L, and T _out are the ratio of reaction heat absorption, the total heat transfer coefficient of the silicon rod and the mixed gas, the total length of the silicon rod, and the equivalent surface of the chassis and the furnace wall. The temperature is constant; r is the radius of the polysilicon rod, and r can be regarded as a constant in a short time;

From equations (20) and (21), the power characteristic equation of polysilicon load with radius r can be obtained as follows:

In the formula, A, B, C, D, G, H are the polysilicon power characteristic fitting coefficients, which are constants and are obtained by fitting the data during the actual production rated operation; I is the polysilicon load single-phase current;

For the silicon rod surface temperature T _x , the control range is:

T _x ^min ≤T _x ≤T _x ^max (23)

When 1000℃ _≤Tx≤1100 ℃, the production can be guaranteed, and the optimum temperature is when Tx ₌ Tx _{, N} =1080℃; when participating in the adjustment, the polysilicon load generally participates in the downward adjustment of the power, then there _is Tx ^min =1000℃,T _x ^max =1080℃;

For cooling water, its flow rate is generally adjusted. If the cooling water flow rate adjustment rate is set to α, there are

α ^min ≤α≤α ^max (25)

where α ^min =90% and α ^max =100%. During rated operation, α=100%;

From formulas (20) and (21), we can get:

From equations (23) and (26), it can be determined that the adjustment range of polysilicon load power is:

The adjustment range of the effective value U _val of the single-phase voltage of the polysilicon load is:

Step 1.4, normalization of industrial park load;

Let the number of electrolytic aluminum load, submerged arc furnace load, and polysilicon load in the industrial park be N _AL , N _SAF , and N _PCS respectively, take the rated voltage of each load as the voltage reference value, and the voltage of each load as U ^* , each load The minimum and maximum voltage limits are

Step 1.4.1. For the electrolytic aluminum load i={1,2,...,N _AL } there are:

P _AL,i =m _1,i ·U ^*2 +m _2,i U ^* (30)

为第i个电解铝的电解槽直流母线电压额定值，R_EC,i为第i个电解铝的电解槽串联等效电阻，E_i为第i个电解铝的电解槽等效电势；In the formula, P _AL,i is the power of the ith electrolytic aluminum load,

is the DC bus voltage rating of the electrolytic cell of the ith electrolytic aluminum, R _EC,i is the series equivalent resistance of the electrolytic cell of the ith electrolytic aluminum, and E _i is the equivalent potential of the electrolytic cell of the ith electrolytic aluminum;

The limit variation range of the DC bus voltage per unit value U ^* of the electrolytic aluminum load is:

Step 1.4.2. Load i={N _AL +1,N _AL +2,…,N _AL +N _SAF } for submerged arc furnace:

P _SAF,i =m _1,i ·U ^*2 (33)

Q _SAF,i =n _i ·U ^*2 (34)

为第i个矿热炉的低压侧电压额定值，

为第i个矿热炉负荷额定运行时的功率特性参数，R_line,i和X_line,i为第i个矿热炉短网的等效电阻和等效电抗，R_arc,i和X_arc,i为第i个矿热炉的电弧静态电阻和电弧静态电抗；In the formula, P _SAF,i and Q _SAF,i are the active power and reactive power of the ith submerged arc furnace load, respectively,

is the low voltage side voltage rating of the i-th submerged arc furnace,

are the power characteristic parameters of the i-th submerged arc furnace during rated load operation, R _line,i and X _line,i are the equivalent resistance and equivalent reactance of the i-th submerged arc furnace short network, R _arc,i and X _{arc , i} is the arc static resistance and arc static reactance of the i-th submerged arc furnace;

The limit variation range of the per-unit value U ^* of the low-voltage side voltage of the submerged arc furnace is:

Step 1.4.3. Load on polysilicon i={N _AL +N _SAF +1,N _AL +N _SAF +2,…,N _AL +N _SAF +N _PCS }:

P _PCS,i =m _1,i ·U ^*2 (38)

为第i个多晶硅负荷单相电压额定值，R_PCS,i为第i个多晶硅棒单相电阻；In the formula, P _PCS,i is the i-th polysilicon load power,

is the single-phase voltage rating of the i-th polysilicon load, R _PCS,i is the single-phase resistance of the i-th polysilicon rod;

The limit variation range of the per-unit value of the single-phase voltage of the polysilicon load is:

Step 1.4.4. The total load power characteristic _P∑ -U ^* of the industrial park is:

P _∑ =M·U ^*2 +M'·U ^* (41)

in,

M and M' are the quadratic term coefficients and the primary term coefficients of the total load power characteristics, respectively, m _1,i and m _2,i are the quadratic term coefficients and primary term coefficients of the power characteristics of various types of loads. , (35), (39) are obtained.

In the above-mentioned control method for power sales companies to respond to power grids based on load aggregation in industrial parks, the implementation of step 2 includes the following steps:

Step 2.1. Give priority to the proportional aggregation model reported by the user:

电压调节死区为

考虑所述式(6)、(19)、(40)确定的各类型负荷的电压极限变化范围，负荷实际参与调节的电压变化范围是：

Set the voltage variation range that the load autonomously reports and wants to participate in regulation as

The voltage regulation deadband is

Considering the voltage limit variation range of each type of load determined by the formulas (6), (19) and (40), the voltage variation range that the load actually participates in regulation is:

Then the actual adjustable minimum/maximum value of the voltage of each load is

The actual allowable maximum up/down adjustment of each load voltage is:

When the power adjustment amount is allocated, the voltage between each load is adjusted in proportion to the size of the voltage adjustment range;

At the initial time, when not participating in the adjustment, each load is in the rated operation state, that is, U ^* _{i,0 =} ¹ ;

ΔU ^* _i,t =U ^* _i,t -U ^* _i,0 =U ^* _i,t -1 (44)

Then the load voltage regulation amount at adjacent moments is:

U ^* _i,t -U ^* _i,t-1 =ΔU ^* _i,t -ΔU ^* _i,t-1 (45)

which is

In the formula, n _t is the proportional adjustment coefficient, the adjustment coefficient of each load is consistent in each adjustment, and the actual adjustment amount is positively correlated with the adjustable range of each load;

Assume that the power regulation amount of the i-th load in the park at time t, relative to its initial power, is ΔP _load,i,t =P _load,i,t -P _load,i,0 , the total power regulation amount of the park load, relative to the initial power is ΔP _∑,t =P _∑,t -P _∑,0 ;

Then from the load regulation characteristics (30), (33), (38), the aggregation characteristics of each load can be obtained:

again

The load aggregation characteristics of the industrial park are:

in,

Substitute n _t =1 into equations (47) and (48) to obtain the maximum upward adjustment capacity ΔP _load,i,up ^max and ΔP _Σ,up ^max of each load and industrial park respectively; substituting n _t =-1 into Obtain the maximum downward adjustment capacity ΔP _load,i,down ^max and ΔP _Σ,down ^max of each load and industrial park respectively;

Step 2.2, the aggregation algorithm of the electricity sales company in response to the grid demand;

Step 2.2.1, offline aggregation stage;

Step 2.2.1.1. Solve the load power characteristics of the industrial park: Based on the measured data, determine the power characteristics of electrolytic aluminum, submerged arc furnace, and polysilicon load by formulas (1), (7), (8) and (20), (22) respectively. ;

Step 2.2.1.2. Solve the total load power characteristic: the load power characteristic parameters are obtained by formula (31), (35), (36), (39), and the total load power characteristic is obtained by formula (41);

Step 2.2.1.3. Solve the voltage limit adjustment range: Calculate the voltage limit variation range of each load from equations (4), (5), (19), (29) combined with equations (32), (37), (40).

Step 2.2.1.4. Determine the actual voltage adjustment range: record the voltage adjustment range and adjustment dead zone reported by each load independently, and solve the actual allowable maximum upward adjustment amount of the voltage of each load by formulas (42) and (43);

Step 2.2.1.5. Calculate the load aggregation characteristics and maximum adjustment capacity of the industrial park: Obtain the load aggregation characteristics ΔP _∑,t -n _t of the industrial park from equations (48) and (49); let n _t =1 or -1 to find Get the maximum upward and downward adjustment capacity ΔP _Σ,up ^max , ΔP _Σ,down ^max ;

Step 2.2.2, online calculation stage;

Assuming that the adjustment power of the industrial park at the previous moment of load control is ΔP _Σ,t-1 , and the initial adjustment time ΔP _Σ,0 =0, then the maximum upward and downward adjustment capacity of the industrial park at the current time t is ΔP _Σ,up,t ^max =ΔP _Σ,up ^max -ΔP _Σ,t-1 ;

Step 2.2.3. Report the adjustable capacity of the grid at the current moment;

The electricity sales company uploads the maximum upward and downward adjustment capacities ΔP _Σ,up,t ^max and ΔP _Σ,down,t ^max to the grid dispatch center at this time.

In the above-mentioned control method for a power sales company based on load aggregation in an industrial park to respond to a power grid, the implementation of step 3 includes the following steps:

Step 3.1, load distribution principle;

When the power grid issues the control target power P _t ^net to the power sales company, the power sales company distributes power to each load in the industrial park;

Suppose the total power adjustment amount of the industrial park at time t is ΔP _∑,t , and solve it as follows:

ΔP _∑,t =ΔP _t ^net =P _t ^net -P _t-1 ^net (50)

The aggregation command n _t corresponding to the power adjustment amount ΔP _∑,t can be obtained from equation (48):

From equations (44) and (46), the voltage control objective of each load is:

The actual power of each load after participating in the adjustment can be obtained according to formulas (1), (7), (8) and formulas (20) and (22) respectively;

Step 3.2, load control instruction;

Step 3.2.1. Control instructions for electrolytic aluminum load;

Substituting the DC bus voltage of the electrolytic cell V _B,i,t =V _BN,i ·U ^* _i,t into the formula (2), the equivalent value of the saturable reactor _LSR,i of the electrolytic aluminum load can be solved as:

Substituting the DC bus voltage V _B,i,t =V _BN,i ·U ^* _i,t into the formula (1), the rectified current value I _d,i,t corresponding to the load can be solved as:

Step 3.2.2. Control instructions for submerged arc furnace load;

The electrode current command value I _ref is used to control the electrode lift; the electrode hydraulic lift model is as follows:

Among them, L is the distance between the electrode and the charge, which is equal to the arc length of the arc, L ₀ is the initial position of the electrode, and I _ref is the electrode current command value; v _up and v _down represent the maximum value of the rising and falling speeds of the electrode, both is a fixed value, c _up and c _down are the proportional coefficients between the current difference and the speed;

When changing the arc impedance, adjust the I _ref in the lift model according to the following formula, and then the electrode lift can be realized:

时，电极控制指令由下式求得：When the voltage on the low-voltage side of the submerged arc furnace is

, the electrode control command is obtained by the following formula:

Step 3.2.3. Control instructions for polysilicon load;

It can be seen from equation (20) that the polysilicon load power can be adjusted by changing the single-phase voltage U _val of the polysilicon load; the patch wave technology is used to control U _val , as follows:

In the formula, the voltage angular frequency ω is a constant, U ₁ and U ₂ are the patch wave voltage, and t ₁ is the patch wave time;

After obtaining the polysilicon load single-phase voltage U _val,i,t =U _valN,i ·U ^* _i,t , first determine the patchwork voltages U ₁ , U ₂ : the patch voltage values are U ₁ , U ₂ ∈ { 0,380,600,800,1500}V, take the patchwork voltages U ₁ and U ₂ as the two adjacent voltage levels U ₁ <U ₂ of the target voltage U _val,i,t respectively; then U _val,i,t , U ₁ , U ₂ are substituted into equation (58), and the time t ₁ of the patch voltage is calculated; U _{val, i, t is} substituted into equation (26), and the surface temperature T _x of the silicon rod and the cooling water flow rate α are solved: the silicon rod is preferentially maintained in the adjustment Surface temperature T _x =1080°C.

In the above-mentioned control method for a power sales company based on load aggregation in an industrial park to respond to a power grid, the implementation of step 4 includes the following steps:

Step 4.1, the influence of control strategy on the benefit of load production;

According to the different directions in which the industrial park load participates in the upward/downward power regulation, the load control cost is divided into:

Step 4.1.1. When the load power adjustment amount ΔP _load <0, its unit value loss F _v is calculated by the following formula:

F _v =(F _p -F _c )/ _CE (59)

where F _v is the value loss per unit load, in yuan/kWh, F _p is the selling price per unit load, in yuan/ton, F _c is the production cost per unit load, in yuan/ton, C _E is the electricity consumption per unit output, unit kWh/ton;

Step 4.1.2. When the load power adjustment amount ΔP _load > 0, model the loss of overload operation based on the life cycle model:

Let the maintenance cost of the load be λ _i , unit element, the rated maintenance life cycle is τ _i,N , unit h; when the load is running at time t, the life change due to overload operation is τ _i,t , the unit is h, then the load The change ρ _i,t of the converted maintenance cost per hour, in unit yuan/h, is calculated by the following formula:

Denote the load power as P _N,i , the unit is kW, then the converted maintenance cost of the unit power consumption of the load is the maintenance unit price F _re , the unit is yuan/kWh, which can be solved by the following formula:

To sum up, the control cost F ₀ of the load unit electricity, unit yuan/kWh, is:

Step 4.2, industrial park control cost model;

Suppose the control cost of each high-energy-consuming load in an industrial park is F _0,i , where i=1,2,...N _AL +N _SAF +N _PCS , which are in the order of electrolytic aluminum, submerged arc furnace, and polysilicon load. Numbering;

Then the relationship between the control cost of each load and the regulated power is as follows:

where F _load,i,t is the control cost of the ith load;

设工业园区的总控制代价为F_IP,t：Power regulation in industrial parks

Let the total control cost of the industrial park be F _IP,t :

When solving the functional relationship between the industrial park control cost F _IP,t and the regulation power ΔP _Σ,t F _IP,t -ΔP _Σ,t , the industrial park control cost is obtained by combining the load bureau and characteristics ΔP _load,i,t -n _t -Aggregate the model F _IP,t -nt, and then obtain the power regulation cost model F _IP,t -ΔP _Σ,t of the industrial park, namely:

The power regulation amount of the ith load in the industrial park is determined by the following formula:

From equations (65) and (66), the control cost-aggregation model F _IP,t -nt of the industrial park can be obtained as follows:

in,

When n _t = 1, the cost F _IP,t ^max when the industrial park provides the maximum upward adjustable power can be obtained; when n _t = -1, the cost when the industrial park provides the maximum downward adjustable power can be obtained. cost F _IP,t ^min ;

From equations (51) and (67), the power regulation cost model F _IP,t -ΔP _Σ,t of the industrial park can be obtained as follows:

in,

In the above-mentioned control method for power sales companies to respond to power grids based on industrial park load aggregation, the implementation of step 5 includes the following steps:

Step 5.1, the control cost model and coordinated control method of the electricity sales company;

The number of industrial parks managed by the electricity sales company is N _IP , and the industrial parks are sorted according to the lowest cost when the power is adjusted upward, and the numbers are recorded as u1, u2...uN _IP ; the industrial parks are ranked according to the lowest cost when the power is adjusted downward, The numbers are denoted as d1, d2...dN _IP , so the control cost model F _PSC -ΔP _PSC of the electricity sales company is:

In the formula, F _PSC is the control cost of the electricity sales company, ΔP _PSC is the total regulated power of the electricity retail company, that is, the sum of the regulated power of each industrial park, ΔP _Σ,up ^ui,max , ΔP _Σ,down ^di,max are respectively numbered as The maximum upward adjustment capacity of the industrial park of ui, and the maximum downward adjustment capacity of the industrial park numbered di;

According to the competitive bidding strategy, the power of each industrial park participates in the adjustment according to the adjustment demand and in the order of control cost from low to high; the relationship between the total adjusted power of the electricity sales company and the power of each park is as follows:

The total power of each industrial park can be obtained by Equation (70), and the aggregate command n _t of each park can be solved by Equation (51), and then Equations (52), (54), (57), (58) and (26) ) to obtain the voltage target value of each load and the actual control command of each load;

Step 5.2, the coordinated control algorithm of the electricity sales company for multiple parks;

Step 5.2.1. Offline aggregation stage;

由式(62)测算控制代价；Step 5.2.1.1. Each park load independently reports the voltage adjustment range

Calculate the control cost by formula (62);

Step 5.2.1.2. Evaluate and confirm the adjustment range of each load by formula (43), calculate the aggregation characteristic ΔP _Σ,t -n _t of each park by formula (48), and calculate the power adjustment cost model F of each park by formula (67). _IP - _ΔPΣ ;

Step 5.2.1.3. Calculate the control cost model F _PSC -ΔP _PSC of the electricity retail company by formula (69);

Step 5.2.2. Online aggregation calculation stage;

Calculate the maximum upward/downward adjustable capacity of each park at the current time t from the previous control time t-1 ΔP _Σ,up,t ^ui,max =ΔP _Σ,up ^ui,max -ΔP _Σ,t-1 , ΔP _{Σ ,down,t} ^di,max =ΔP _Σ,down ^di,max -ΔP _Σ,t-1 , the control cost of the electricity sales company at time t is F _PSC,t =F _PSC (ΔP _PSC -ΔP _Σ,i );

Step 5.2.2. Report to the power grid; update the adjustable capacity and cost model and report to the power grid;

Step 5.2.3. Receive grid power regulation demand;

Step 5.2.3.1. After receiving the power grid power adjustment demand ΔP _net,t , confirm the total power adjustment amount ΔP _Σ,t of each industrial park in turn according to formula (70) and the current adjustable capacity of each park;

Step 5.2.3.2. Allocate the load power in each industrial park according to step 3, and determine each control command;

Step 5.2.3.3. Record and update the current load status and wait for the next adjustment.

The beneficial effects of the invention are as follows: based on the load production characteristics and control means, the power regulation characteristics of typical industrial park loads are modeled; the load regulation willingness and limit regulation range are integrated, and a proportional aggregation model that gives priority to the regulation range reported by the user is proposed. This establishes the aggregation control strategy of the power sales company for the industrial park to respond to the frequency regulation demand of the power grid; considering the load production benefit, the control cost model of various types of loads is established, and combined with the cost and the aggregation model, the power sales company's demand for multi-park participation is proposed The coordinated control strategy of response; for the scenario of wind power fluctuations in large power grids, the stability of the grid is maintained by the load response to power fluctuations, which verifies the correctness of the proposed control strategy.

It is a good model for multi-party participation and mutual benefit to study the aggregation strategy of power sales companies for industrial park loads to participate in demand response and assist the grid to stabilize power fluctuations caused by new energy access. Electricity sales companies participate in grid auxiliary services through aggregation control strategies to expand business content; industrial loads can obtain certain economic compensation when production conditions permit, which solves the current operating dilemma of high energy-consuming loads; The low cost ensures the stability and reliability of the power grid, which helps to achieve a lower evaluation index of "abandoning wind" and "abandoning light"; for new energy power generators, the access rate of new energy is increased, and more subsidies for new energy power generation are obtained. ; On the whole, the overall economic benefit of the load-grid-source has been improved.

Description of drawings

1 is an overall block diagram of an aggregate control strategy of a power sales company in response to a power grid according to an embodiment of the present invention;

Fig. 2 is the total load power characteristic of a typical industrial park according to an embodiment of the present invention;

Fig. 3 is the aggregation characteristic model of a typical industrial park according to an embodiment of the present invention;

Fig. 4 is a one-minute wind power fluctuation diagram in a region in Example 1 of the present invention;

Fig. 5(a) is a graph of aggregated command changes when an embodiment of the present invention responds to wind power fluctuations in Fig. 3;

Fig. 5(b) is a graph of the load voltage change of electrolytic aluminum when an embodiment of the present invention responds to the wind power fluctuation of Fig. 3;

Fig. 5(c) is a graph of the load voltage change of the submerged arc furnace when an embodiment of the present invention responds to the fluctuation of the wind power in Fig. 3;

Fig. 5(d) is a graph of the voltage change of r=75mm polysilicon when an embodiment of the present invention responds to the wind power fluctuation of Fig. 3;

Fig. 6(a) is a tie line fluctuation diagram of an embodiment of the present invention that does not adopt the aggregation control strategy for the wind power fluctuation in Fig. 3;

Fig. 6(b) is a tie line fluctuation diagram that adopts an aggregation control strategy for wind power fluctuation in Fig. 3 according to an embodiment of the present invention;

Fig. 7(a) is a graph showing the load power change of electrolytic aluminum when an embodiment of the present invention responds to the fluctuation of wind power in Fig. 3;

Fig. 7(b) is a graph showing the load power change of the submerged arc furnace when an embodiment of the present invention responds to the fluctuation of the wind power in Fig. 3;

Fig. 7(c) is a graph of polysilicon load power change when an embodiment of the present invention responds to wind power fluctuations in Fig. 3;

Fig. 8 is the power regulation cost F _IP - ΔP _Σ of a typical industrial park of the present invention;

9 is a characteristic diagram of the control cost of three industrial parks in Example 2 of the present invention;

Figure 10 is a one-minute wind power fluctuation diagram in a region in Example 2 of the present invention;

11 is the control cost characteristic of the power sales company of the present invention to the load aggregation of the industrial park in Example 2;

FIG. 12 is a control cost diagram of the electricity sales company in response to the wind power fluctuation in FIG. 9 in Example 2 of the present invention.

Detailed ways

The embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

This embodiment is implemented through the following technical solutions. The power sales company response grid control method based on industrial park load aggregation includes the following steps:

Step 1. According to the power characteristics of electrolytic aluminum, submerged arc furnace, and polysilicon loads, normalize the load power of the industrial park, and analyze the power theoretical adjustment boundary of each load;

Step 2. Combining the load regulation willingness and the limit regulation range, a proportional aggregation model is proposed that prioritizes the regulation range reported by users, and the aggregation control strategy of the electricity sales company for the industrial park is established;

Step 3. According to the aggregation model, establish the response strategy of the electricity sales company to the power grid control command, so as to realize the reasonable distribution and control of the load power of the industrial park;

Step 4. Analyze the impact of the control strategy on the benefit of load production, establish a load control cost model, and combine the aggregation model to establish a load control cost model for the industrial park;

Step 5. Based on the control cost model of the industrial park, a bidding and optimal coordination strategy of the electricity sales company for multiple industrial parks is proposed, so as to realize the coordinated distribution and control of the power of the multiple parks.

The above steps constitute the aggregate control strategy of the electricity sales company in response to the grid. By aggregating the load of the industrial park, the electricity retail company obtains the load aggregation characteristic ΔP _Σ,t -n _t of the industrial park and the control cost model F _IP -ΔP _Σ of the industrial park, and adjusts the adjustable capacity (the maximum upward adjustment capacity ΔP _Σ,up ^max ) , the maximum down regulation capacity ΔP _Σ,down ^max ) and the regulation cost of the electricity sales company (F _PSC -P _PSC ) are reported to the power grid, and then according to the grid regulation command, the power regulation command (electrolytic aluminum rectifier current value I _d , the electrode current command value I _ref of the submerged arc furnace, the splicing time ωt ₁ of polysilicon, the splicing voltages U ₁ , U ₂ ) are sent to each industrial park and its different types of loads, and finally realize the control of power fluctuations. Calm down. As shown in Figure 1.

During specific implementation, in step 1, the number of electrolytic aluminum load, submerged arc furnace load, and polysilicon load in the industrial park is recorded as N _AL , N _SAF , and N _PCS , respectively, and the rated voltage of each load is taken as the voltage reference value, and recorded The amount of each load voltage is U ^* , and the limit minimum and maximum value of each load voltage are recorded as

For electrolytic aluminum load (i={1,2,...,N _AL }) there are:

P _AL,i =m _1,i ·U ^*2 +m _2,i U ^* (30)

为第i个电解铝的电解槽直流母线电压额定值，R_EC,i为第i个电解铝的电解槽串联等效电阻，E_i为第i个电解铝的电解槽等效电势。In the formula, P _AL,i is the power of the ith electrolytic aluminum load,

is the DC bus voltage rating of the electrolytic cell of the ith electrolytic aluminum, R _EC,i is the series equivalent resistance of the electrolytic cell of the ith electrolytic aluminum, and E _i is the equivalent potential of the electrolytic cell of the ith electrolytic aluminum.

The limit variation range of the DC bus voltage per unit value U ^* of the electrolytic aluminum load is:

For submerged arc furnace load (i={N _AL +1,N _AL +2,…,N _AL +N _SAF }):

P _SAF,i =m _1,i ·U ^*2 (33)

Q _SAF,i =n _i ·U ^*2 (34)

为第i个矿热炉的低压侧电压额定值，

为第i个矿热炉负荷额定运行时的功率特性参数，R_line,i和X_line,i为第i个矿热炉短网的等效电阻和等效电抗，R_arc,i和X_arc,i为第i个矿热炉的电弧静态电阻和电弧静态电抗。In the formula, P _SAF,i and Q _SAF,i are the active power and reactive power of the ith submerged arc furnace load, respectively,

is the low voltage side voltage rating of the i-th submerged arc furnace,

are the power characteristic parameters of the i-th submerged arc furnace during rated load operation, R _line,i and X _line,i are the equivalent resistance and equivalent reactance of the i-th submerged arc furnace short network, R _arc,i and X _{arc , i} is the arc static resistance and arc static reactance of the i-th submerged arc furnace.

The limit variation range of the per-unit value U ^* of the low-voltage side voltage of the submerged arc furnace is:

For polysilicon loading (i={N _AL +N _SAF +1,N _AL +N _SAF +2,...,N _AL +N _SAF +N _PCS }):

P _PCS,i =m _1,i ·U ^*2 (38)

为第i个多晶硅负荷单相电压额定值，R_PCS,i为第i个多晶硅棒单相电阻。In the formula, P _PCS,i is the i-th polysilicon load power,

is the single-phase voltage rating of the i-th polysilicon load, and R _PCS,i is the single-phase resistance of the i-th polysilicon rod.

The limit variation range of the per-unit value of the single-phase voltage of the polysilicon load is:

Firstly, the power regulation characteristics of typical industrial park load electrolytic aluminum, submerged arc furnace and polysilicon are modeled. The industrial park contains an electrolytic aluminum plant with a total of one-phase electrolytic aluminum load with a rated power of 410MW; a submerged arc furnace load with a rated power of 76MW; a polysilicon plant with a load rated power of 122MW. Three types of typical load parameters are shown in Tables 1, 2, and 3 below.

Table 1 Electrolytic aluminum load parameters

In the table, V _AL-AHN is the rated value of the high-voltage busbar voltage connected to the electrolytic aluminum load, k _ALN is the rated value of the on-load voltage regulating transformer ratio, V _BN is the rated value of the DC busbar voltage of the electrolytic cell, and E is the equivalent value of the electrolytic cell. Potential, R _EC is the equivalent resistance of the electrolytic cell in series, L _SRN is the equivalent value of the saturable reactor.

Table 2 Load parameters of submerged arc furnace

In the table, V _SAF-AHN is the rated value of the high voltage busbar of the submerged arc furnace, k _SAFN is the rated value of the transformer ratio of the submerged arc furnace, U _SAFN is the rated value of the low voltage side of the submerged arc furnace, and R _line +jX _line is the short line The equivalent reactance rating of , R _arcN +jX _arcN is the equivalent impedance rating of the arc, I _refN is the electrode current command value rating, R _arc is the arc equivalent resistance value, X _arc =a ₁ R _arc ² +a ₂ R _arc +a ₃ is the arc impedance relationship of the submerged arc furnace, P _SAF,t ^min is the minimum power of the submerged arc furnace, and cosψ _arc is the power factor of the arc part.

Table 3 Polysilicon load parameters (silicon rod radius r = 5, 10, ..., 6 each of 75mm)

为额定情况下多晶硅负荷的单相电压随半径变化特性，P_PCS＝33.942·r+2.217为额定情况下多晶硅负荷的交流总功率随半径变化特性，η为反应吸热占比，K为硅棒与混合气体总传热系数，L为硅棒总长度，T_out为底盘及炉璧表面等效温度，T_X和T_Xn为硅棒表面温度及其额定值，α和α_N为冷却水流速调节率及其额定值。In the table, I ² =11.711r ³ +0.765·r ² is the variation of single-phase current with radius of polysilicon load under rated conditions,

is the variation characteristics of single-phase voltage of polysilicon load with radius under rated conditions, P _PCS =33.942·r+2.217 is the variation characteristics of AC total power of polysilicon load with radius under rated conditions, η is the ratio of reaction heat absorption, and K is silicon rod The total heat transfer coefficient with the mixed gas, L is the total length of the silicon rod, T _out is the equivalent temperature of the surface of the chassis and furnace wall, T _X and T _Xn are the surface temperature of the silicon rod and its rated value, α and α _N are the cooling water flow rate Regulation rate and its rating.

The load power characteristic parameters are obtained by substituting each parameter into equations (31), (35), (36), and (39). And the total load power characteristics are obtained from the following formula, as shown in Figure 2.

P _∑ =M·U ^*2 +M'·U ^* (41)

电压调节死区

如下表4所示。Each load autonomously reports the voltage variation range that participates in regulation

Voltage Regulation Dead Band

As shown in Table 4 below.

Table 4 Load voltage limit change constraints and autonomous reporting parameters

为负荷电压极限最小值，

为负荷电压极限最大值，

为负荷上报电压最小值，

为负荷上报电压最大值，

为负荷电压调节死区；多晶硅负荷数据为r＝75mm硅棒的参数。In the table,

is the minimum load voltage limit,

is the maximum load voltage limit,

Report the minimum voltage for the load,

Report the maximum voltage for the load,

The dead zone is adjusted for the load voltage; the polysilicon load data is the parameter of r=75mm silicon rod.

The actual adjustable minimum/maximum value of the voltage of each load is

The actual allowable maximum up/down adjustment of each load voltage is:

Therefore, the range of the actual upward/downward adjustable capacity of each load can be calculated from equations (42) and (43), see Table 5 below.

Table 5 Actual up/down adjustable voltage and power of load

为负荷实际允许最大向上调节量，

为负荷实际允许最大向下调节量，ΔP_load,i,up ^max为负荷实际向上可调容量，ΔP_load,i,down ^max为负荷实际向下可调容量。In the table,

For the actual allowable maximum upward adjustment of the load,

is the actual maximum allowable downward adjustment of the load, ΔP _load,i,up ^max is the actual upward adjustable capacity of the load, and ΔP _load,i,down ^max is the actual downward adjustable capacity of the load.

In step 2, it is assumed that each load is in the rated operating state at the initial time (when not participating in the adjustment), that is, U ^* _i,0 =1. The variation of the load voltage at time t and the voltage at the initial time is recorded as ΔU ^* _i,t , that is,

ΔU ^* _i,t =U ^* _i,t -U ^* _i,0 =U ^* _i,t -1 (44)

Then the load voltage regulation amount at adjacent moments is:

U ^* _i,t -U ^* _i,t-1 =ΔU ^* _i,t -ΔU ^* _i,t-1 (45)

which is

In the formula, n _t is the proportional adjustment coefficient, and the adjustment coefficient of each load is consistent in each adjustment, and the actual adjustment amount is positively correlated with the adjustable range of each load.

Let the power regulation amount of the i-th load in the park at time t (relative to its initial power) be ΔP _load,i,t =P _load,i,t -P _load,i,0 , the total power regulation amount of the park load (relative to the initial power) power) is ΔP _∑,t =P _∑,t -P _∑,0 .

Then the aggregation characteristics of each load can be obtained from the load regulation characteristic formulas (30), (33), (38):

again

The load aggregation characteristics of the industrial park are:

in,

The load aggregation characteristic ΔP _∑,t -n _t of the park can be obtained from equations (48) and (49), as shown in Fig. 3 .

Let n _t =1 or -1 to obtain the maximum upward and downward adjustment capacity ΔP _Σ,up ^max , ΔP _Σ,down ^max of the industrial park as 22.3415MW and -80.3150MW.

In step 3, for the aforementioned example, the wind power fluctuation occurs within 1 minute on the connecting line between the industrial park and the local power grid, as shown in FIG. 4 . The electricity sales company uses the total power adjustment ΔP _∑,t of the industrial park to respond to the power fluctuation of the tie line, that is,

ΔP _∑,t =ΔP _t ^net =P _t ^net -P _t-1 ^net (50)

The aggregation command n _t corresponding to the power adjustment amount ΔP _∑,t can be obtained from equation (48):

From equations (44) and (46), the voltage control objective of each load is:

From equation (51), the aggregation command of the electricity sales company at each time can be obtained, and then the target voltage of each load can be obtained from equation (52), according to equations (1), (7), (8) and (20), ( 22) Obtain the actual power of each load.

The aggregated commands and load voltage changes in response to wind power fluctuations in Figure 3 are shown in Figure 5(a), Figure (b), Figure (c), and Figure (d). Figures 6(a) and (b) show the tie line fluctuation diagrams before and after the aggregation control strategy is adopted. The tie line power fluctuation at each moment is reduced to around -0.3882MW, which has an obvious effect on the stabilization of wind power fluctuations.

When the aggregate control strategy is used to respond to power fluctuations, the power changes of each load are shown in Figure 7(a), Figure 7(b), and Figure 7(c). It can be seen from the figure that each load strictly meets the needs of self-reporting of the load at the time of control, which will not exceed the adjustment capacity of the load, and make full use of the adjustable capacity of each load.

In step 4, the production benefit of each type of load is firstly analyzed. According to the different directions in which the industrial park load participates in the upward/downward power regulation, the load control cost is divided into:

①When the load power adjustment amount ΔP _load < 0, the energy loss due to the power adjustment will affect the load output, thereby causing the load value loss, and the unit value loss F _v is calculated by the following formula:

F _v =(F _p -F _c )/ _CE (59)

Among them, F _v is the value loss per unit load (yuan/kWh), F _p is the selling price per unit load (yuan/ton), F _c is the production cost per unit load (yuan/ton), and C _E is the electricity consumption per unit output ( kWh/ton).

② When the load power adjustment amount ΔP _load > 0, although the load input energy increases, this part of energy cannot make the load increase correspondingly due to the capacity limitation of the load equipment, that is, at this time, the electric energy cost of the load is additionally increased, that is, the electricity price Pr _E (yuan/kWh).

In addition, since overload operation under high power conditions will increase the failure rate of the equipment and affect the life of the equipment, the following will model the loss of overload operation based on the life cycle model.

Let the maintenance cost of the load be λ _i (yuan), and the rated maintenance life cycle be τ _i,N (h). When the load is running at time t, the life change due to overload operation is τ _i,t (h), then the change in the hourly maintenance cost of the load ρ _i,t (yuan/h) can be calculated by the following formula:

Denote the load power as P _N,i (kW), then the converted maintenance cost of the unit power consumption of the load is the maintenance unit price F _re (yuan/kWh), which can be solved by the following formula:

To sum up, the control cost F ₀ (yuan/kWh) of load unit electricity is:

The economic parameters and control costs of typical park load control are shown in Table 6 below.

Table 6 Economic parameters of high energy-consuming load control in typical parks

The power regulation cost model F _IP,t -ΔP _Σ,t of the industrial park is as follows:

in,

Substituting the parameters in Table 6 into equations (62) and (67) can solve the power regulation cost F _IP -ΔP _Σ of the aforementioned typical industrial park, as shown in Fig. 8 .

In step 5, the number of industrial parks managed by the electricity sales company is recorded as N _IP , and the industrial parks are sorted according to the smallest cost when the power is adjusted upward, and the numbers are recorded as u1, u2...uN _IP ; the industrial parks are adjusted downward according to the power The order of the time cost is the smallest, and the numbers are recorded as d1, d2...dN _IP , so the control cost model F _PSC -ΔP _PSC of the electricity sales company is:

Among them, F _PSC is the control cost of the electricity retail company, ΔP _PSC is the total regulated power of the electricity retail company, that is, the sum of the regulated power of each industrial park, ΔP _Σ,up ^ui,max , ΔP _Σ,down ^di,max are respectively numbered ui The maximum upward adjustment capacity of the industrial park, and the maximum downward adjustment capacity of the industrial park numbered di.

According to the competitive bidding strategy, the power of each industrial park participates in the regulation according to the regulation demand and in the order of control cost from low to high. The relationship between the total regulated power of the electricity sales company and the power of each park is as follows:

An electricity sales company manages three industrial parks, and the high energy load of each park is as follows:

Park 1: Phase I electrolytic aluminum load, rated power is 410MW; Phase I submerged arc furnace load, rated power is 76MW; Phase I electrolytic aluminum load, rated power is 122MW, grouped according to the radius of production silicon rods, (r=5, 10,15,…,75mm), 6 silicon rods per group.

Park 2: Two-phase electrolytic aluminum load, with rated power of 410MW and 620MW respectively; first-phase submerged arc furnace load, rated power of 76MW.

Park 3: Two-phase submerged arc furnace load, with rated power of 76MW and 90MW respectively; two-phase electrolytic aluminum load, rated power of 122MW and 203MW respectively, grouped according to the radius of production silicon rods, r=5, 10, 15, …, 75mm, 6 and 8 silicon rods in each group, respectively.

Then, the power adjustment cost F _IP -ΔP _Σ of each park can be solved by equation (67), as shown in Figure 9. Industrial Park 1 has a maximum upward adjustment capacity of 22.335MW and a maximum downward adjustment capacity of -80.315MW; Industrial Park 2 has a larger adjustment capacity, with a maximum upward adjustment capacity of 52.6303MW and a maximum downward adjustment capacity of -136.4597MW. The control cost is relatively large; the maximum upward adjustment capacity of Industrial Park 3 is 13.3976MW, and the maximum downward adjustment capacity is -93.9124MW, and the control cost is relatively small.

The tie line between the place and the local power grid fluctuates within 1 minute due to wind power fluctuations, as shown in Figure 10.

The electricity sales company adopts the multi-park coordinated control strategy shown in Equation (70), that is, when participating in the upward adjustment of power, it prioritizes the adjustment of the parks with lower control costs, that is, in the order of the first, second, and third parks; when participating in the downward adjustment of the power At the time of adjustment, priority is given to adjusting the parks with the lowest control cost, that is, adjusting in the order of parks three, one, and two. The control cost model F _PSC -ΔP _PSC of the electricity sales company can be obtained as shown in Figure 11.

To stabilize the wind power fluctuation shown in Figure 11, within the 60s control time, the control cost of the electricity sales company at each moment is shown in Figure 12 (yuan/MWh). During the 60s control time, the total amount of electricity to stabilize the power is 2.1765MWh, and the total control cost is 189.1082 yuan. The average control cost is 86.8869 yuan/MWh.

To sum up, based on the modeling of the power regulation characteristics of typical industrial park loads, this embodiment proposes the limit regulation range of each load; and combined with the load regulation willingness, a proportional aggregation model that prioritizes the regulation range reported by users is proposed. This establishes the aggregation control strategy of the power sales company for the industrial park to respond to the frequency regulation demand of the power grid; by quantifying the impact of various types of load control on production benefits, a cost model for industrial park control is established; The park participates in the coordinated control strategy of demand response.

It should be understood that the parts not described in detail in this specification belong to the prior art.

Although the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, those skilled in the art should understand that these are only examples, and various modifications or changes may be made to these embodiments without departing from the principles and principles of the present invention and substance. The scope of the present invention is limited only by the appended claims.

Claims (6)

1. a power sales company response grid control method based on industrial park load aggregation, is characterized in that, comprises the following steps: Step 1. According to the power characteristics of electrolytic aluminum, submerged arc furnace, and polysilicon loads, normalize the load power of the industrial park, and analyze the power theoretical adjustment boundary of each load; Step 2. Combining the load regulation willingness and the limit regulation range, a proportional aggregation model is proposed that prioritizes the regulation range reported by users, and the aggregation control strategy of the electricity sales company for the industrial park is established; Step 3. According to the aggregation model, establish the response strategy of the electricity sales company to the power grid control command, so as to realize the rational distribution and aggregation control of the load power of the industrial park; Step 4. Analyze the impact of the aggregation control strategy on the benefit of load production, establish a load control cost model, and combine the aggregation model to establish a load control cost model for the industrial park; Step 5. Based on the load control cost model of the industrial park, a bidding and optimal coordination strategy of the electricity sales company for multiple industrial parks is proposed to realize the coordinated distribution and control of the power of the multiple parks. 2. The power sales company response grid control method based on industrial park load aggregation as claimed in claim 1, wherein the specific implementation of step 1 comprises: Step 1.1. The load power characteristics of electrolytic aluminum are as follows:

Wherein, P _AL is the power of electrolytic aluminum, V _B is the DC bus voltage of the electrolytic cell, I _d is the direct current of the electrolytic cell, R _EC is the equivalent resistance of the electrolytic cell in series, and E is the equivalent potential of the electrolytic cell; The relationship between the electrolytic aluminum DC bus voltage V _B and the high voltage bus voltage V _AL-AH connected to the electrolytic aluminum load is:

In the formula, L _SR is the equivalent value of the saturable reactor, V _AL-AH is the high-voltage bus voltage, and ω is the voltage angular frequency; Considering that the electrolytic aluminum load is connected to an on-load voltage regulating transformer, its transformation ratio has a total of m ₁ level, then changing its transformation ratio k _AL can realize m ₁ level adjustment:

Assuming that the adjustment range of the equivalent value L _SR of the saturable reactor is [L _SR ^min , L _SR ^max ] when the production requirements are met, then by formula (2), the electrolytic aluminum DC busbar under the corresponding strain ratio k _AL-i can be solved The voltage drop V _B adjustment range is:

From the formula (1), the maximum adjustable power range of the electrolytic aluminum load can be obtained as follows:

Step 1.2. The power characteristics of the submerged arc furnace load are as follows:

Among them, P _SAF and Q _SAF are the active power and reactive power of the submerged arc furnace respectively, U _SAF is the low-voltage side voltage of the submerged arc furnace, and R _line and X _line are the equivalent resistance and equivalent reactance of the short grid; It is assumed that the equivalent impedance of the arc is represented by the static arc resistance R _arc and the static arc reactance X _arc in a certain time section; The low voltage side voltage U _SAF of the submerged arc furnace is obtained from the high voltage busbar voltage U _SAF-AH of the submerged arc furnace through the special transformer for the submerged arc furnace: U _SAF = U _SAF-AH /k _SAF (9) In the formula, k _SAF is the transformation ratio of the special transformer for submerged arc furnace; By fitting the actual production data, the arc impedance relationship of the submerged arc furnace is obtained: X _arc =a ₁ R _arc ² +a ₂ R _arc +a ₃ (10) Among them, a ₁ , a ₂ , and a ₃ are arc impedance fitting coefficients, which are constants; When the constant impedance voltage regulation method is used to adjust the power of the submerged arc furnace, the transformation ratio of the special transformer for the load of the submerged arc furnace is set to have a total of m ₂ levels _; scope:

The limited range of arc static resistance R _arc : R _arc ^min ≤R _arc ≤R _arc ^max (12) Arc part power factor upper limit

There is a lower limit for the power factor of the arc part

Then the power factor of the arc part

The limits are:

which is

in,

At the same time, each smelting stage should meet the minimum power constraint: P _SAF,t ≥P _SAF,t ^min (15) In the formula, P _SAF,t ^min is the minimum power of the submerged arc furnace; Assuming that the rated voltage of the low-voltage side of the submerged arc furnace is U _SAF,N , by formulas (7), (8), (12), (15), when the constant voltage and impedance power adjustment method is adopted, the submerged heat in each smelting stage The adjustment ranges of the active power P _SAF,t and reactive power Q _SAF,t of the furnace are as follows: c ₁ ≤P _SAF,t ≤ c ₂ (16) d ₁ ≤Q _SAF,t ≤ d ₂ (17) in:

The power characteristic parameters of the submerged arc furnace under rated load operation are:

Convert the power adjustment range at this time into the equivalent low-voltage side voltage variation range, that is

From equations (10) and (18), when the impedance-voltage coordinated adjustment method is adopted, the voltage variation range of the low-voltage side of the submerged arc furnace load is:

Step 1.3. The polysilicon load power satisfies the following electrical relationship:

In the formula, P _PCS is the total AC power of the polysilicon load, U _val is the single-phase voltage of the polysilicon load, and R _PCS is the single-phase resistance of the polysilicon rod; The energy conversion relationship in the production process of a polycrystalline silicon rod in the Δt time is as follows:

In the formula, P _PCS is the total AC power of the polysilicon load, ΔQ _out1 represents the heat used to heat the reaction gas, and v ₁ ·Δt ·s ₁ ·ρ _g ·c·(T _x -T _g ) can be obtained from the gas specific heat capacity formula, Among them, v ₁ , s ₁ , ρ _g , c, and T _g are the inlet velocity, inlet area, mixed gas density, mixed gas specific heat capacity, and inlet temperature, all of which are constants; T _x , is the surface temperature of the silicon rod , ΔQ _out2 and ΔQ _out3 represent the sustaining endothermic reaction and the heat dissipated through the reduction furnace chassis and furnace walls due to thermal radiation, respectively

Corresponding to (ΔQ _out2 +ΔQ _out3 ), where η, K, L, and T _out are the ratio of reaction heat absorption, the total heat transfer coefficient of the silicon rod and the mixed gas, the total length of the silicon rod, and the equivalent surface of the chassis and the furnace wall. The temperature is constant; r is the radius of the polysilicon rod, and r can be regarded as a constant in a short time; From equations (20) and (21), the power characteristic equation of polysilicon load with radius r can be obtained as follows:

In the formula, A, B, C, D, G, H are the polysilicon power characteristic fitting coefficients, which are constants and are obtained by fitting the data during the actual production rated operation; I is the polysilicon load single-phase current; For the silicon rod surface temperature T _x , the control range is: T _x ^min ≤T _x ≤T _x ^max (23) When 1000℃ _≤Tx≤1100 ℃, the production can be guaranteed, and the optimum temperature is when Tx ₌ Tx _{, N} =1080℃; when participating in the adjustment, the polysilicon load generally participates in the downward adjustment of the power, then there _is Tx ^min =1000℃,T _x ^max =1080℃; For cooling water, its flow rate is generally adjusted. If the cooling water flow rate adjustment rate is set to α, there are

α ^min ≤α≤α ^max (25) where α ^min =90% and α ^max =100%. During rated operation, α=100%; From formulas (20) and (21), we can get:

From equations (23) and (26), it can be determined that the adjustment range of polysilicon load power is:

The adjustment range of the effective value U _val of the single-phase voltage of the polysilicon load is:

Step 1.4, normalization of industrial park load; Let the number of electrolytic aluminum load, submerged arc furnace load, and polysilicon load in the industrial park be N _AL , N _SAF , and N _PCS respectively, take the rated voltage of each load as the voltage reference value, and the voltage of each load as U ^* , each load The minimum and maximum voltage limits are

Step 1.4.1. For the electrolytic aluminum load i={1,2,...,N _AL } there are: P _AL,i =m _1,i ·U ^*2 +m _2,i U ^* (30)

In the formula, P _AL,i is the power of the ith electrolytic aluminum load, V _BN,i is the DC bus voltage rating of the ith electrolytic aluminum electrolytic cell, and R _EC,i is the ith electrolytic aluminum electrolytic cell series connection Equivalent resistance, E _i is the equivalent potential of the electrolytic cell of the i-th electrolytic aluminum; The limit variation range of the DC bus voltage per unit value U ^* of the electrolytic aluminum load is:

Step 1.4.2. Load i={N _AL +1,N _AL +2,…,N _AL +N _SAF } for submerged arc furnace: P _SAF,i =m _1,i ·U ^*2 (33) Q _SAF,i =n _i ·U ^*2 (34)

In the formula, P _SAF,i and Q _SAF,i are the active power and reactive power of the ith submerged arc furnace load, respectively,

is the low voltage side voltage rating of the i-th submerged arc furnace,

are the power characteristic parameters of the i-th submerged arc furnace during rated load operation, R _line,i and X _line,i are the equivalent resistance and equivalent reactance of the i-th submerged arc furnace short network, R _arc,i and X _{arc , i} is the arc static resistance and arc static reactance of the i-th submerged arc furnace; The limit variation range of the per-unit value U ^* of the low-voltage side voltage of the submerged arc furnace is:

Step 1.4.3. Load on polysilicon i={N _AL +N _SAF +1,N _AL +N _SAF +2,…,N _AL +N _SAF +N _PCS }: P _PCS,i =m _1,i ·U ^*2 (38)

In the formula, P _PCS,i is the i-th polysilicon load power,

is the single-phase voltage rating of the i-th polysilicon load, R _PCS,i is the single-phase resistance of the i-th polysilicon rod; The limit variation range of the per-unit value of the single-phase voltage of the polysilicon load is:

Step 1.4.4. The total load power characteristic _PΣ -U ^* of the industrial park is: P _Σ = M · U ^*2 +M' · U ^* (41) in,

M and M' are the quadratic term coefficients and the primary term coefficients of the total load power characteristics, respectively, m _1,i and m _2,i are the quadratic term coefficients and primary term coefficients of the power characteristics of various types of loads. , (35), (39) are obtained. 3. The power sales company response grid control method based on industrial park load aggregation as claimed in claim 2, wherein the realization of step 2 comprises the following steps: Step 2.1. Give priority to the proportional aggregation model reported by the user: Set the voltage variation range that the load autonomously reports and wants to participate in regulation as

The voltage regulation deadband is

Considering the voltage limit variation range of each type of load determined by the formulas (6), (19) and (40), the voltage variation range that the load actually participates in regulation is:

Then the actual adjustable minimum/maximum value of the voltage of each load is

The actual allowable maximum up/down adjustment of each load voltage is:

When the power adjustment amount is allocated, the voltage between each load is adjusted in proportion to the size of the voltage adjustment range; At the initial time, when not participating in the adjustment, each load is in the rated operation state, that is, U ^* _{i,0 =} ¹ ; ΔU ^* _i,t =U ^* _i,t -U ^* _i,0 =U ^* _i,t -1 (44) Then the load voltage regulation amount at adjacent moments is: U ^* _i,t -U ^* _i,t-1 =ΔU ^* _i,t -ΔU ^* _i,t-1 (45) which is

In the formula, n _t is the proportional adjustment coefficient, the adjustment coefficient of each load is consistent in each adjustment, and the actual adjustment amount is positively correlated with the adjustable range of each load; Assume that the power regulation amount of the i-th load in the park at time t, relative to its initial power, is ΔP _load,i,t =P _load,i,t -P _load,i,0 , the total power regulation amount of the park load, relative to the initial power is ΔP _Σ,t =P _Σ,t -P _Σ,0 ; Then from the load regulation characteristics (30), (33), (38), the aggregation characteristics of each load can be obtained:

again

The load aggregation characteristics of the industrial park are:

in,

Substitute n _t =1 into equations (47) and (48) to obtain the maximum upward adjustment capacity ΔP _load,i,up ^max and ΔP _Σ,up ^max of each load and industrial park respectively; substituting n _t =-1 into Obtain the maximum downward adjustment capacity ΔP _load,i,down ^max and ΔP _Σ,down ^max of each load and industrial park respectively; Step 2.2, the aggregation algorithm of the electricity sales company in response to the grid demand; Step 2.2.1, offline aggregation stage; Step 2.2.1.1. Solve the load power characteristics of the industrial park: Based on the measured data, determine the power characteristics of electrolytic aluminum, submerged arc furnace, and polysilicon load by formulas (1), (7), (8) and (20), (22) respectively. ; Step 2.2.1.2. Solve the total load power characteristic: the load power characteristic parameters are obtained by formula (31), (35), (36), (39), and the total load power characteristic is obtained by formula (41); Step 2.2.1.3. Solve the voltage limit adjustment range: Calculate the voltage limit variation range of each load from equations (4), (5), (19), (29) combined with equations (32), (37), (40).

Step 2.2.1.4. Determine the actual voltage adjustment range: record the voltage adjustment range and adjustment dead zone reported by each load independently, and solve the actual allowable maximum upward adjustment amount of the voltage of each load by formulas (42) and (43); Step 2.2.1.5. Calculate the load aggregation characteristics and maximum adjustment capacity of the industrial park: Obtain the load aggregation characteristics ΔP _∑,t -n _t of the industrial park from equations (48) and (49); let n _t =1 or -1 to find Get the maximum upward and downward adjustment capacity ΔP _Σ,up ^max , ΔP _Σ,down ^max ; Step 2.2.2, online calculation stage; Assuming that the adjustment power of the industrial park at the previous moment of load control is ΔP _Σ,t-1 , and the initial adjustment time ΔP _Σ,0 =0, then the maximum upward and downward adjustment capacity of the industrial park at the current time t is ΔP _Σ,up,t ^max =ΔP _Σ,up ^max -ΔP _Σ,t-1 ; Step 2.2.3. Report the adjustable capacity of the grid at the current moment; The electricity sales company uploads the maximum upward and downward adjustment capacities ΔP _Σ,up,t ^max and ΔP _Σ,down,t ^max to the grid dispatch center at this time. 4. The power sales company response grid control method based on industrial park load aggregation as claimed in claim 3, wherein the realization of step 3 comprises the following steps: Step 3.1, load distribution principle; When the power grid issues the control target power P _t ^net to the power sales company, the power sales company distributes power to each load in the industrial park; Suppose the total power adjustment amount of the industrial park at time t is ΔP _∑,t , and solve it as follows: ΔP _∑,t =ΔP _t ^net =P _t ^net -P _t-1 ^net (50) The aggregation command n _t corresponding to the power adjustment amount ΔP _∑,t can be obtained from equation (48):

From equations (44) and (46), the voltage control objective of each load is:

The actual power of each load after participating in the adjustment can be obtained according to formulas (1), (7), (8) and formulas (20) and (22) respectively; Step 3.2, load control instruction; Step 3.2.1. Control instructions for electrolytic aluminum load; Substituting the DC bus voltage of the electrolytic cell V _B,i,t =V _BN,i ·U ^* _i,t into the formula (2), the equivalent value of the saturable reactor _LSR,i of the electrolytic aluminum load can be solved as:

Substituting the DC bus voltage V _B,i,t =V _BN,i ·U ^* _i,t into the formula (1), the rectified current value I _d,i,t corresponding to the load can be solved as:

Step 3.2.2. Control instructions for submerged arc furnace load; The electrode current command value I _ref is used to control the electrode lift; the electrode hydraulic lift model is as follows:

Among them, L is the distance between the electrode and the charge, which is equal to the arc length of the arc, L ₀ is the initial position of the electrode, and I _ref is the electrode current command value; v _up and v _down represent the maximum value of the rising and falling speeds of the electrode, both is a fixed value, c _up and c _down are the proportional coefficients between the current difference and the speed; When changing the arc impedance, adjust the I _ref in the lift model according to the following formula, and then the electrode lift can be realized:

When the voltage on the low-voltage side of the submerged arc furnace is

, the electrode control command is obtained by the following formula:

Step 3.2.3. Control command of polysilicon load; It can be seen from equation (20) that the polysilicon load power can be adjusted by changing the single-phase voltage U _val of the polysilicon load; the patch wave technology is used to control U _val , as follows:

In the formula, the voltage angular frequency ω is a constant, U ₁ and U ₂ are the patch wave voltage, and t ₁ is the patch wave time; After obtaining the polysilicon load single-phase voltage U _val,i,t =U _valN,i ·U ^* _i,t , first determine the patchwork voltages U ₁ , U ₂ : the patch voltage values are U ₁ , U ₂ ∈ { 0,380,600,800,1500}V, take the patchwork voltages U ₁ and U ₂ as the two adjacent voltage levels U ₁ <U ₂ of the target voltage U _val,i,t respectively; then U _val,i,t , U ₁ , U ₂ are substituted into equation (58), and the time t ₁ of the patch voltage is calculated; U _{val, i, t is} substituted into equation (26), and the surface temperature T _x of the silicon rod and the cooling water flow rate α are solved: the silicon rod is preferentially maintained in the adjustment Surface temperature T _x =1080°C. 5. The power sales company response grid control method based on industrial park load aggregation as claimed in claim 4, wherein the realization of step 4 comprises the following steps: Step 4.1, the influence of control strategy on the benefit of load production; According to the different directions in which the industrial park load participates in the upward/downward power regulation, the load control cost is divided into: Step 4.1.1. When the load power adjustment amount ΔP _load <0, its unit value loss F _v is calculated by the following formula: F _v =(F _p -F _c )/ _CE (59) where F _v is the value loss per unit load, in yuan/kWh, F _p is the selling price per unit load, in yuan/ton, F _c is the production cost per unit load, in yuan/ton, C _E is the electricity consumption per unit output, unit kWh/ton; Step 4.1.2. When the load power adjustment amount ΔP _load > 0, model the loss of overload operation based on the life cycle model: Let the maintenance cost of the load be λ _i , unit element, the rated maintenance life cycle is τ _i,N , unit h; when the load is running at time t, the life change due to overload operation is τ _i,t , the unit is h, then the load The change ρ _i,t of the converted maintenance cost per hour, in unit yuan/h, is calculated by the following formula:

Denote the load power as P _N,i , the unit is kW, then the converted maintenance cost of the unit power consumption of the load is the maintenance unit price F _re , the unit is yuan/kWh, which can be solved by the following formula:

To sum up, the control cost F ₀ of the load unit electricity, unit yuan/kWh, is:

Step 4.2, industrial park control cost model; Suppose the control cost of each high-energy-consuming load in an industrial park is F _0,i , where i=1,2,...N _AL +N _SAF +N _PCS , which are in the order of electrolytic aluminum, submerged arc furnace, and polysilicon load. Numbering; Then the relationship between the control cost of each load and the regulated power is as follows:

where F _load,i,t is the control cost of the ith load; Power regulation in industrial parks

Let the total control cost of the industrial park be F _IP,t :

When solving the functional relationship between the industrial park control cost F _IP,t and the regulation power ΔP _Σ,t F _IP,t -ΔP _Σ,t , the industrial park control cost is obtained by combining the load bureau and characteristics ΔP _load,i,t -n _t -Aggregate the model F _IP,t -nt, and then obtain the power regulation cost model F _IP,t -ΔP _Σ,t of the industrial park, namely:

The power regulation amount of the ith load in the industrial park is determined by the following formula:

From equations (65) and (66), the control cost-aggregation model F _IP,t -nt of the industrial park can be obtained as follows:

in,

When n _t = 1, the cost F _IP,t ^max when the industrial park provides the maximum upward adjustable power can be obtained; when n _t = -1, the cost when the industrial park provides the maximum downward adjustable power can be obtained. cost F _IP,t ^min ; From equations (51) and (67), the power regulation cost model F _IP,t -ΔP _Σ,t of the industrial park can be obtained as follows:

in,

6. The power sales company response grid control method based on industrial park load aggregation as claimed in claim 5, wherein the realization of step 5 comprises the following steps: Step 5.1, the control cost model and coordinated control method of the electricity sales company; The number of industrial parks managed by the electricity sales company is N _IP , and the industrial parks are sorted according to the lowest cost when the power is adjusted upward, and the numbers are recorded as u1, u2...uN _IP ; the industrial parks are ranked according to the lowest cost when the power is adjusted downward, The numbers are denoted as d1, d2...dN _IP , so the control cost model F _PSC -ΔP _PSC of the electricity sales company is:

In the formula, F _PSC is the control cost of the electricity sales company, ΔP _PSC is the total regulated power of the electricity retail company, that is, the sum of the regulated power of each industrial park, ΔP _Σ,up ^ui,max , ΔP _Σ,down ^di,max are respectively numbered as The maximum upward adjustment capacity of the industrial park of ui, and the maximum downward adjustment capacity of the industrial park numbered di; According to the competitive bidding strategy, the power of each industrial park participates in the adjustment according to the adjustment demand and in the order of control cost from low to high; the relationship between the total adjusted power of the electricity sales company and the power of each park is as follows:

The total power of each industrial park can be obtained by Equation (70), and the aggregate command n _t of each park can be solved by Equation (51), and then Equations (52), (54), (57), (58) and (26) ) to obtain the voltage target value of each load and the actual control command of each load; Step 5.2, the coordinated control algorithm of the electricity sales company for multiple parks; Step 5.2.1. Offline aggregation stage; Step 5.2.1.1. Each park load independently reports the voltage adjustment range

Calculate the control cost by formula (62); Step 5.2.1.2. Evaluate and confirm the adjustment range of each load by formula (43), calculate the aggregation characteristic ΔP _Σ,t -n _t of each park by formula (48), and calculate the power adjustment cost model F of each park by formula (67). _IP - _ΔPΣ ; Step 5.2.1.3. Calculate the control cost model F _PSC -ΔP _PSC of the electricity retail company by formula (69); Step 5.2.2. Online aggregation calculation stage; Calculate the maximum upward/downward adjustable capacity of each park at the current time t from the previous control time t-1 ΔP _Σ,up,t ^ui,max =ΔP _Σ,up ^ui,max -ΔP _Σ,t-1 , ΔP _{Σ ,down,t} ^di,max =ΔP _Σ,down ^di,max -ΔP _Σ,t-1 , the control cost of the electricity sales company at time t is F _PSC,t =F _PSC (ΔP _PSC -ΔP _Σ,i ); Step 5.2.2. Report to the power grid; update the adjustable capacity and cost model and report to the power grid; Step 5.2.3. Receive grid power regulation demand; Step 5.2.3.1. After receiving the power grid power adjustment demand ΔP _net,t , confirm the total power adjustment amount ΔP _Σ,t of each industrial park in turn according to formula (70) and the current adjustable capacity of each park; Step 5.2.3.2. Allocate the load power in each industrial park according to step 3, and determine each control command; Step 5.2.3.3. Record and update the current load status and wait for the next adjustment.

Images