Frequency reserve allocation method for demand response resources considering frequency security constraints

By constructing a dynamic frequency model of the power system and converting constraints using the HDMR method, the high-order nonlinear problem of frequency safety constraints in the power system is solved, and frequency stability and economical frequency modulation backup allocation are achieved in the new energy environment.

CN114530894BActive Publication Date: 2024-11-08ZHEJIANG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210193863.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-03-01
Publication Date
2024-11-08
Estimated Expiration
2042-03-01

AI Technical Summary

Technical Problem

With the increase in the penetration rate of new energy, the frequency stability of the power system is at risk, and the existing technology is difficult to effectively solve the high-order nonlinear problem of frequency safety constraints, resulting in difficulties in allocating demand response resources for frequency regulation and backup.

Method used

Build a dynamic frequency model of power system, convert complex high-order nonlinear frequency safety constraints into low-order quadratic constraints using high-dimensional model expression (HDMR) method, and solve them through McCormick envelope processing and commercial solver gurobi , optimize frequency modulation backup allocation.

Benefits of technology

On the premise of ensuring the frequency safety of the power system, it is realized within the bearable range of the frequency after power disturbance, and the total cost is reduced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114530894B_ABST
    Figure CN114530894B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for allocating frequency modulation standby for demand response resources considering frequency security constraints. Construct a power system, obtain the frequency response parameters and cost parameters of each generator set and demand response resource; construct a frequency dynamic model of the power system; construct a frequency modulation standby allocation optimization model for the power system; establish a high-order nonlinear frequency security constraint of the objective function; use a high-dimensional model expression method to convert it into a low-order quadratic frequency security constraint; use McCormick envelope processing and then use a commercial solver to solve, and allocate frequency modulation standby for demand response resources. The method of the present invention can ensure the safety of the power system operation in terms of frequency indicators, so that the power system can still ensure that the frequency operation is within an acceptable range after being subjected to a certain power disturbance.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a frequency modulation reserve allocation method, and in particular to a frequency modulation reserve allocation method for demand response resources considering frequency safety constraints. Background Art

[0002] With the advancement of the construction of new power systems with new energy as the main body, the penetration rate of new energy in the power system has gradually increased, the system's rotational inertia has decreased significantly, and the system frequency stability faces new risks. For a long time, the power system has relied on thermal power units to provide sufficient frequency regulation reserves to ensure the safety and stability of the system when the power is unbalanced. Therefore, when formulating the power system unit combination and cost scheduling plan, the frequency security constraint is one of the most typical constraints. However, the frequency security constraint is highly nonlinear and it is impossible to directly solve the optimization problem containing this constraint. With the gradual increase in the penetration rate of new energy, the scarcity of frequency regulation resources on the power generation side has become more prominent, and exploring the frequency regulation potential of flexible resources on the load side through demand response technology has become a research hotspot. At present, some demand-side response resources can be used for primary frequency regulation, but because the frequency dynamic characteristics of load equipment are different from those of traditional thermal power units, the system frequency response characteristics have changed, making the already high-order nonlinear frequency security constraints more complicated, which has brought difficulties to the dispatching system to allocate demand response resources for frequency regulation reserves. Summary of the invention

[0003] In order to solve the problems existing in the background technology, the present invention provides a demand response resource frequency modulation reserve allocation method considering frequency safety constraints. According to this method, the frequency modulation reserve plan is formulated to ensure the safety of the system operation on the frequency index, and the system can still ensure that the frequency operation is within an acceptable range after a certain power disturbance. It can ensure a certain cost under the premise of achieving safe and stable frequency of the power system.

[0004] The technical solution adopted by the present invention is:

[0005] The method comprises the following steps:

[0006] Step 1) constructing a power system, including a dispatching system, a generator set, a transmission network and demand response resources of the power system, and obtaining frequency response parameters and cost parameters of the generator set and the demand response resources.

[0007] Step 2) Construct a frequency dynamic model of the power system based on the frequency response parameters of the generator sets and demand response resources.

[0008] Step 3) According to the cost parameters of the generator sets and demand response resources, a frequency regulation reserve allocation optimization model of the power system is constructed, including the objective function and constraints.

[0009] According to the frequency dynamic model, the complex high-order nonlinear frequency safety constraints of the objective function are established.

[0010] Step 4) Use the high dimensional model representation (HDMR) method to transform the complex high-order nonlinear frequency safety constraints of the objective function into low-order quadratic frequency safety constraints.

[0011] Step 5) Use the McCormick envelope to process the low-order quadratic frequency security constraints of the objective function, and then use the commercial solver gurobi to solve the objective function, and allocate frequency reserve to the demand response resources based on the solution results.

[0012] In the step 1), a plurality of generators form a generator set.

[0013] Demand response resources include energy storage battery clusters, variable frequency air conditioner clusters and electric vehicle clusters; a number of energy storage batteries form an energy storage battery cluster, a number of variable frequency air conditioners form a variable frequency air conditioner cluster, and a number of electric vehicles form an electric vehicle cluster; electric vehicles are electric vehicles that are connected to the power system and are being charged. During the charging process, frequency modulation can be achieved by reducing the charging power of the electric vehicle or by reverse charging the demand response resources.

[0014] In the step 2), the regulating power supporting the frequency safety of the power system is provided by the generator set and the demand response resources. According to the frequency response parameters of the generator set and the demand response resources, the frequency dynamic model of the power system is constructed as follows:

[0015] 2HsΔf(s)+DΔf(s)=ΔP G (s)+ΔP DR (s)-ΔP D (s)

[0016] Where H and D represent the inertia and load damping of the power system, respectively, s represents the operator in the frequency domain of the frequency dynamic model, Δf(s) represents the frequency offset of the power system, and ΔP G (s) represents the output change of the generator set, ΔP DR (s) represents the power stability support amount of the demand response resource, ΔP D (s) represents the power disturbance of the power system.

[0017] Represents the inertia link of the power system;

[0018] The output change of the generator set ΔP G (s) are as follows:

[0019]

[0020] Where R represents the droop control coefficient of the generator set, 1 / R represents the primary frequency regulation control parameter of the generator set; F HP Indicates the power ratio of the high-pressure turbine section of the generator set, T RH Represents the reheat time constant of the reheat turbine of the generator set, T G Indicates the speed governor parameters of the generator set, T CH Indicates the turbine parameters of the generator set; R, F HP , T RH , T G and T CH That is the frequency response parameter of the generator set.

[0021] It represents the primary frequency regulation link of the generator set in the frequency dynamic model. represents the speed governor link of the generator set in the frequency dynamic model, Represents the reheat turbine link of the generator set in the frequency dynamic model; the output change of the generator set ΔP G (s) is the multiplication result of the values ​​of the primary frequency regulation link, the speed governor link and the reheat turbine link.

[0022] The power stability power support amount ΔP of the demand response resource DR (s) is the power change of the variable frequency air conditioning cluster ΔP IAC (s), power change of the energy storage battery cluster ΔP B (s) and the power change of the electric vehicle cluster ΔP EV (s) is obtained by adding by an adder.

[0023] The inverter air conditioner cluster is installed in the room. The power change of the inverter air conditioner cluster is ΔP. IAC (s) are as follows:

[0024]

[0025] Among them, θ P represents the operating power constant coefficient of the variable frequency air conditioning cluster, A represents the equivalent primary frequency modulation control parameter of the variable frequency air conditioning cluster, which is equivalent to 1 / R of the generator set, T a Represents the equivalent speed regulator time constant, T a =(C room V room ρ A ) / H r , C room 、V room , A denote the specific heat capacity, volume and air density of the room where the variable frequency air conditioning cluster is installed, H r represents the equivalent heat transfer coefficient inside and outside the room, Tc represents the time constant of the compressor of the variable frequency air conditioning cluster, Φ IAC represents the equivalent controller parameter, Φ IAC =θ Q / H r ,θ Q represents the operating power constant coefficient of the variable frequency air conditioning cluster, C(s) represents the PI controller parameter of the variable frequency air conditioning cluster, C(s)=α+β / s, α and β represent the proportional coefficient and integral coefficient of the PI controller of the variable frequency air conditioning cluster respectively; θ P , A, T a 、T c , Φ IAC and C(s) are the frequency response parameters of the variable frequency air conditioning cluster.

[0026] Represents the thermal-electrical model link of the variable frequency air conditioning cluster in the frequency dynamic model, Represents the compressor link of the variable frequency air conditioning cluster in the frequency dynamic model; -θ P A represents the PI controller link of the variable frequency air conditioning cluster in the frequency dynamic model; the power change of the variable frequency air conditioning cluster ΔP IAC (s) is the result of the multiplication of the values ​​of the thermal-electrical model link, the compressor link and the PI controller link.

[0027] Power change of energy storage battery cluster ΔP B (s) are as follows:

[0028]

[0029] Among them, T B Represents the response time constant of the energy storage battery, R B Indicates the regulation coefficient of the energy storage battery, represents the response characteristics of the energy storage battery cluster, Represents the frequency response coefficient of the energy storage battery cluster; T B and R B That is the frequency response parameter of the energy storage battery cluster.

[0030] Power change of electric vehicle cluster ΔP EV (s) are as follows:

[0031]

[0032] Among them, T EV It represents the response time constant of the electric vehicle cluster. The response time of the bidirectional converter of the electric vehicle cluster is generally in the millisecond level, and the power battery of the electric vehicle cluster generally has no response delay; represents the response characteristics of the electric vehicle cluster; K EVrepresents the frequency response coefficient of the electric vehicle cluster, R EV Indicates the adjustment coefficient of electric vehicles; T EV and K EV That is the frequency response parameter of the electric vehicle cluster.

[0033] The output change of the generator set ΔP G and the power stability power support amount ΔP of the demand response resource DR That is the decision variable of the frequency dynamic model of the power system.

[0034] In the step 3), according to the cost parameters of the generator set and the demand response resource, the objective function of the frequency regulation reserve allocation optimization model of the power system is constructed as follows:

[0035]

[0036] Among them, F represents the total cost of the objective function, min represents the lowest value; C G , C IAC , C B and C EV They represent the operating costs of the generator set, variable frequency air conditioning cluster, energy storage battery cluster and electric vehicle cluster respectively; T is the number of time periods of the power system; N G 、N IAC 、N B and N EV They represent the number of generators in the generator set, variable frequency air conditioners in the variable frequency air conditioner cluster, energy storage batteries in the energy storage battery cluster, and electric vehicles in the electric vehicle cluster; c Gi and c Ri They represent the output cost and standby cost of the i-th generator in the generator set, and They represent the output and reserve of the i-th generator in the generator set during period t; c IACj represents the backup cost of the jth variable frequency air conditioner in the variable frequency air conditioner cluster, represents the standby of the jth variable frequency air conditioner in the variable frequency air conditioner cluster during period t; c Bw represents the backup cost of the wth energy storage battery in the energy storage battery cluster, represents the backup of the wth energy storage battery in the energy storage battery cluster during period t; c EVv represents the backup cost of the vth electric vehicle in the electric vehicle cluster, represents the reserve of the vth electric vehicle in the electric vehicle cluster during period t; c Gi and c Ri That is the cost parameter of the generator set, c IACj 、c Bw and c EVvIt is the cost parameter of demand response resources.

[0037] Generator output cost c Gi Using the traditional quadratic cost model, it is represented as follows:

[0038]

[0039] Among them, a i 、b i 、c i They respectively represent the operating cost coefficient of the generator.

[0040] The objective function is to minimize the operating cost of the power system throughout the entire period. The generation side cost of the power system includes the output cost of the generator set and the cost of the generator set providing frequency regulation backup. The demand side cost of the power system includes the cost of the variable frequency air conditioning cluster, energy storage battery cluster and electric vehicle cluster providing frequency regulation backup.

[0041] In the step 3), in the process of the power generation units of the power system transmitting power to the demand response resources through the transmission network, the constraints of the established frequency regulation reserve allocation optimization model are as follows:

[0042] a) Load power balance constraints:

[0043]

[0044] Among them, N G Indicates the number of generators in the generator set, represents the output of the i-th generator in the generator set during period t, It represents the sum of all load powers in the power system during period t.

[0045] b) Power flow transmission capacity constraints:

[0046]

[0047] Among them, T li represents the power transmission allocation coefficient of the node of the power system where the i-th generator in the generator group is located to the l-th branch in the transmission network; K represents the number of nodes in the power system, T lk represents the power transmission allocation coefficient of the kth node to the lth branch in the transmission network, represents the load of the kth node in period t, It represents the power flow transmission capacity of the lth branch in the transmission network.

[0048] c) Generator spinning reserve capacity constraints:

[0049]

[0050] in, represents the output upper limit of the i-th generator in the generator set during period t, and r represents the reserve factor of the power system; the spinning reserve capacity of the generator set is used to cope with fluctuations in the power system load.

[0051] d) Generator output and reserve capacity constraints:

[0052]

[0053]

[0054] in, It represents the lower limit of the output of the i-th generator in the generator set during the t period;

[0055] e) Demand response resource output capacity constraints:

[0056]

[0057]

[0058]

[0059] in, and They represent the upper and lower output limits of the jth variable frequency air conditioner in the variable frequency air conditioner cluster at time period t respectively; and They represent the upper and lower output limits of the wth energy storage battery in the energy storage battery cluster at time period t respectively; and They represent the upper and lower limits of the output of the vth electric vehicle in the electric vehicle cluster at time period t; represents the standby of the jth variable frequency air conditioner in the variable frequency air conditioner cluster during period t; represents the backup of the wth energy storage battery in the energy storage battery cluster during period t; It represents the reserve of the vth electric vehicle in the electric vehicle cluster during period t.

[0060] f) Generator and demand response resource ramping constraints:

[0061]

[0062]

[0063] in, represents the output of the i-th generator in the generator set at time t-1, Ru i and Rd iThey respectively represent the upward climbing rate and downward climbing rate of the i-th generator in the generator group; compared with the generator group, the demand response resources have less climbing restrictions and a faster climbing speed.

[0064] In the step 3), according to the frequency deviation of the power system output by the frequency dynamic model, the complex high-order nonlinear frequency security constraint of the objective function is obtained as follows:

[0065]

[0066]

[0067] Among them, Δf represents the maximum frequency deviation of the power system, which means that the power system generates power disturbance in time period t. When the generator set, variable frequency air conditioning cluster, energy storage battery cluster and electric vehicle cluster provide and The frequency regulation reserve makes the maximum frequency deviation of the power system; FR() represents the frequency dynamic model, and x represents the disturbance of power disturbance and FM standby A collection of; Indicates the frequency deviation limit of the power system; represents the reserve of the i-th generator in the generator set during period t, represents the standby of the jth variable frequency air conditioner in the variable frequency air conditioner cluster during period t, represents the backup of the wth energy storage battery in the energy storage battery cluster during period t, It represents the reserve of the vth electric vehicle in the electric vehicle cluster during period t.

[0068] The frequency regulation backup provided by both the source and load sides, that is, the frequency regulation backup provided by the generator set, variable frequency air conditioning cluster, energy storage battery cluster and electric vehicle cluster respectively, should be able to ensure that the maximum frequency deviation Δf of the power system is within the frequency deviation limit value. If the maximum frequency offset Δf exceeds the frequency offset limit If the frequency of the power system drops too severely, it will cause low-frequency load shedding in the power system and cause power outages.

[0069] The mapping relationship Δf=FR(x) between the frequency regulation reserve provided by both sides of the source and the load and the maximum frequency offset Δf of the power system is determined by the frequency dynamic model and is a high-order nonlinear implicit correlation.

[0070] The HDMR method is used to express the complex high-order nonlinear frequency safety constraints of the objective function by using a high-dimensional model to transform them into low-order quadratic frequency safety constraints as shown below:

[0071]

[0072]

[0073] Wherein, DDFR() represents the data driven frequency dynamic model converted by the HDMR method.

[0074] The HDMR method can decompose this high-dimensional nonlinear physical system into multiple low-dimensional problems.

[0075] In the step 5), the McCormick envelope is used to process the low-order quadratic frequency security constraints, and then the commercial solver gurobi is used to solve the objective function of the frequency regulation reserve allocation optimization model to obtain the operating cost of the demand response resources when the total cost of the objective function is the lowest, and then the reserve of the demand response resources is obtained. Based on the obtained reserve of the demand response resources, instructions are issued to the demand response resources through the dispatching system of the power system to perform frequency regulation reserve allocation.

[0076] The beneficial effects of the present invention are:

[0077] The present invention constructs a demand response resource frequency modulation reserve allocation method considering frequency safety constraints. According to this method, the frequency modulation reserve plan is formulated to ensure the safety of the system operation on the frequency index, and the system can still ensure that the frequency operation is within the tolerable range after a certain power disturbance. It can ensure a certain cost under the premise of achieving safe and stable frequency of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 Schematic diagram of a frequency dynamic model considering demand response resources for the present invention. DETAILED DESCRIPTION

[0079] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0080] The method of the present invention comprises the following steps:

[0081] Step 1) constructing a power system, including a dispatching system, a generator set, a transmission network and demand response resources of the power system, and obtaining frequency response parameters and cost parameters of the generator set and the demand response resources.

[0082] A number of generators form a generator set; demand response resources include energy storage battery clusters, variable frequency air conditioner clusters and electric vehicle clusters; a number of energy storage batteries form an energy storage battery cluster, a number of variable frequency air conditioners form a variable frequency air conditioner cluster, and a number of electric vehicles form an electric vehicle cluster; electric vehicles are electric vehicles that are connected to the power system and are being charged. During the charging process, frequency modulation can be achieved by reducing the charging power of the electric vehicle or reverse charging the demand response resources.

[0083] Step 2) According to the frequency response parameters of the generator sets and demand response resources, a frequency dynamic model of the power system is constructed, such as Figure 1 shown.

[0084] The regulating power that supports the frequency security of the power system is provided by the generator sets and demand response resources. According to the frequency response parameters of the generator sets and demand response resources, the frequency dynamic model of the power system is constructed as follows:

[0085] 2HsΔf(s)+DΔf(s)=ΔP G (s)+ΔP DR (s)-ΔP D (s)

[0086] Where H and D represent the inertia and load damping of the power system, respectively, s represents the operator in the frequency domain of the frequency dynamic model, Δf(s) represents the frequency offset of the power system, and ΔP G (s) represents the output change of the generator set, ΔP DR (s) represents the power stability support amount of the demand response resource, ΔP D (s) represents the power disturbance of the power system.

[0087] Represents the inertia link of the power system.

[0088] The output change of the generator set ΔP G (s) are as follows:

[0089]

[0090] Where R represents the droop control coefficient of the generator set, 1 / R represents the primary frequency regulation control parameter of the generator set; F HP Indicates the power ratio of the high-pressure turbine section of the generator set, T RH Represents the reheat time constant of the reheat turbine of the generator set, T G Indicates the speed governor parameters of the generator set, T CH Indicates the turbine parameters of the generator set; R, F HP , T RH , T G and T CH That is the frequency response parameter of the generator set.

[0091] It represents the primary frequency regulation link of the generator set in the frequency dynamic model. represents the speed governor link of the generator set in the frequency dynamic model, Represents the reheat turbine link of the generator set in the frequency dynamic model; the output change of the generator set ΔP G(s) is the multiplication result of the values ​​of the primary frequency regulation link, the speed governor link and the reheat turbine link.

[0092] Power stability support amount ΔP of demand response resources DR (s) is the power change of the variable frequency air conditioning cluster ΔP IAC (s), power change of the energy storage battery cluster ΔP B (s) and the power change of the electric vehicle cluster ΔP EV (s) is obtained by adding by an adder.

[0093] The inverter air conditioner cluster is installed in the room. The power change of the inverter air conditioner cluster is ΔP. IAC (s) are as follows:

[0094]

[0095] Among them, θ P represents the operating power constant coefficient of the variable frequency air conditioning cluster, A represents the equivalent primary frequency modulation control parameter of the variable frequency air conditioning cluster, which is equivalent to 1 / R of the generator set, T a Represents the equivalent speed regulator time constant, T a =(C room V room ρ A ) / H r , C room 、V room , A denote the specific heat capacity, volume and air density of the room where the variable frequency air conditioning cluster is installed, H r represents the equivalent heat transfer coefficient inside and outside the room, T c represents the time constant of the compressor of the variable frequency air conditioning cluster, Φ IAC represents the equivalent controller parameter, Φ IAC =θ Q / H r ,θ Q represents the operating power constant coefficient of the variable frequency air conditioning cluster, C(s) represents the PI controller parameter of the variable frequency air conditioning cluster, C(s)=α+β / s, α and β represent the proportional coefficient and integral coefficient of the PI controller of the variable frequency air conditioning cluster respectively; θ P , A, T a 、T c , Φ IAC and C(s) are the frequency response parameters of the variable frequency air conditioning cluster.

[0096] Represents the thermal-electrical model link of the variable frequency air conditioning cluster in the frequency dynamic model, Represents the compressor link of the variable frequency air conditioning cluster in the frequency dynamic model; -θ PA represents the PI controller link of the variable frequency air conditioning cluster in the frequency dynamic model; the power change of the variable frequency air conditioning cluster ΔP IAC (s) is the result of the multiplication of the values ​​of the thermal-electrical model link, the compressor link and the PI controller link.

[0097] Power change of energy storage battery cluster ΔP B (s) are as follows:

[0098]

[0099] Among them, T B Represents the response time constant of the energy storage battery, R B Indicates the regulation coefficient of the energy storage battery, represents the response characteristics of the energy storage battery cluster, Represents the frequency response coefficient of the energy storage battery cluster; T B and R B That is the frequency response parameter of the energy storage battery cluster.

[0100] Power change of electric vehicle cluster ΔP EV (s) are as follows:

[0101]

[0102] Among them, T EV It represents the response time constant of the electric vehicle cluster. The response time of the bidirectional converter of the electric vehicle cluster is generally in the millisecond level, and the power battery of the electric vehicle cluster generally has no response delay; represents the response characteristics of the electric vehicle cluster; K EV represents the frequency response coefficient of the electric vehicle cluster, R EV Indicates the adjustment coefficient of electric vehicles; T EV and K EV That is the frequency response parameter of the electric vehicle cluster.

[0103] The output change of the generator set ΔP G and the power stability power support amount ΔP of the demand response resource DR That is the decision variable of the frequency dynamic model of the power system.

[0104] Step 3) According to the cost parameters of the generator sets and demand response resources, the frequency regulation reserve allocation optimization model of the power system is constructed, including the objective function and constraints. According to the frequency dynamic model, the complex high-order nonlinear frequency security constraints of the objective function are established.

[0105] According to the cost parameters of the generator sets and demand response resources, the objective function of the frequency regulation reserve allocation optimization model of the power system is constructed as follows:

[0106]

[0107] Among them, F represents the total cost of the objective function, min represents the lowest value; C G , C IAC , C B and C EV They represent the operating costs of the generator set, variable frequency air conditioning cluster, energy storage battery cluster and electric vehicle cluster respectively; T is the number of time periods of the power system; N G 、N IAC 、N B and N EV They represent the number of generators in the generator set, variable frequency air conditioners in the variable frequency air conditioner cluster, energy storage batteries in the energy storage battery cluster, and electric vehicles in the electric vehicle cluster; c Gi and c Ri They represent the output cost and standby cost of the i-th generator in the generator set, and They represent the output and reserve of the i-th generator in the generator set during period t; c IACj represents the backup cost of the jth variable frequency air conditioner in the variable frequency air conditioner cluster, represents the standby of the jth variable frequency air conditioner in the variable frequency air conditioner cluster during period t; c Bw represents the backup cost of the wth energy storage battery in the energy storage battery cluster, represents the backup of the wth energy storage battery in the energy storage battery cluster during period t; c EVv represents the backup cost of the vth electric vehicle in the electric vehicle cluster, represents the reserve of the vth electric vehicle in the electric vehicle cluster during period t; c Gi and c Ri That is the cost parameter of the generator set, c IACj 、c Bw and c EVv It is the cost parameter of demand response resources.

[0108] Generator output cost c Gi Using the traditional quadratic cost model, it is represented as follows:

[0109]

[0110] Among them, a i 、b i 、c i They respectively represent the operating cost coefficient of the generator.

[0111] The objective function is to minimize the operating cost of the power system throughout the entire period. The generation side cost of the power system includes the output cost of the generator set and the cost of the generator set providing frequency regulation backup. The demand side cost of the power system includes the cost of the variable frequency air conditioning cluster, energy storage battery cluster and electric vehicle cluster providing frequency regulation backup.

[0112] In the process of power generation units in the power system transmitting power to demand response resources through the transmission network, the constraints of the established frequency regulation reserve allocation optimization model are as follows:

[0113] a) Load power balance constraints:

[0114]

[0115] Among them, N G Indicates the number of generators in the generator set, represents the output of the i-th generator in the generator set during period t, It represents the sum of all load powers in the power system during period t.

[0116] b) Power flow transmission capacity constraints:

[0117]

[0118] Among them, T li represents the power transmission allocation coefficient of the node of the power system where the i-th generator in the generator group is located to the l-th branch in the transmission network; K represents the number of nodes in the power system, T lk represents the power transmission allocation coefficient of the kth node to the lth branch in the transmission network, represents the load of the kth node in period t, It represents the power flow transmission capacity of the lth branch in the transmission network.

[0119] c) Generator spinning reserve capacity constraints:

[0120]

[0121] in, represents the output upper limit of the i-th generator in the generator set during period t, and r represents the reserve factor of the power system; the spinning reserve capacity of the generator set is used to cope with fluctuations in the power system load.

[0122] d) Generator output and reserve capacity constraints:

[0123]

[0124]

[0125] in, It represents the lower limit of the output of the i-th generator in the generator set during period t.

[0126] e) Demand response resource output capacity constraints:

[0127]

[0128]

[0129]

[0130] in, and They represent the upper and lower output limits of the jth variable frequency air conditioner in the variable frequency air conditioner cluster at time period t respectively; and They represent the upper and lower output limits of the wth energy storage battery in the energy storage battery cluster at time period t respectively; and They represent the upper and lower limits of the output of the vth electric vehicle in the electric vehicle cluster at time period t; represents the standby of the jth variable frequency air conditioner in the variable frequency air conditioner cluster during period t; represents the backup of the wth energy storage battery in the energy storage battery cluster during period t; It represents the reserve of the vth electric vehicle in the electric vehicle cluster during period t.

[0131] f) Generator and demand response resource ramping constraints:

[0132]

[0133]

[0134] in, represents the output of the i-th generator in the generator set at time t-1, Ru i and Rd i They respectively represent the upward climbing rate and downward climbing rate of the i-th generator in the generator group; compared with the generator group, the demand response resources have less climbing restrictions and a faster climbing speed.

[0135] According to the frequency deviation of the power system output by the frequency dynamic model, the complex high-order nonlinear frequency security constraints of the objective function are obtained as follows:

[0136]

[0137]

[0138] Among them, Δf represents the maximum frequency deviation of the power system, which means that the power system generates power disturbance in time period t. When the generator set, variable frequency air conditioning cluster, energy storage battery cluster and electric vehicle cluster provide and The frequency regulation reserve makes the maximum frequency deviation of the power system; FR() represents the additional Figure 1 The frequency dynamic model shown, x represents the disturbance amount of power disturbance and FM standby A collection of; Indicates the frequency deviation limit of the power system; represents the reserve of the i-th generator in the generator set during period t, represents the standby of the jth variable frequency air conditioner in the variable frequency air conditioner cluster during period t, represents the backup of the wth energy storage battery in the energy storage battery cluster during period t, It represents the reserve of the vth electric vehicle in the electric vehicle cluster during period t.

[0139] The frequency regulation backup provided by both the source and load sides, that is, the frequency regulation backup provided by the generator set, variable frequency air conditioning cluster, energy storage battery cluster and electric vehicle cluster respectively, should be able to ensure that the maximum frequency deviation Δf of the power system is within the frequency deviation limit value. If the maximum frequency offset Δf exceeds the frequency offset limit If the frequency of the power system drops too severely, it will cause low-frequency load shedding in the power system and cause power outages.

[0140] The mapping relationship Δf=FR(x) between the frequency regulation reserve provided by both sides of the source and the load and the maximum frequency offset Δf of the power system is determined by the frequency dynamic model and is a high-order nonlinear implicit correlation.

[0141] Step 4) Use the high dimensional model representation (HDMR) method to transform the complex high-order nonlinear frequency safety constraints of the objective function into low-order quadratic frequency safety constraints.

[0142] The HDMR method is used to express the complex high-order nonlinear frequency safety constraints of the objective function by using a high-dimensional model to transform them into low-order quadratic frequency safety constraints as shown below:

[0143]

[0144]

[0145] Wherein, DDFR() represents the data driven frequency dynamic model converted by the HDMR method.

[0146] The HDMR method can decompose this high-dimensional nonlinear physical system into multiple low-dimensional problems.

[0147] Step 5) Use the McCormick envelope to process the low-order quadratic frequency security constraints of the objective function, and then use the commercial solver gurobi to solve the objective function, and allocate frequency reserve to the demand response resources based on the solution results.

[0148] The McCormick envelope is used to process low-order quadratic frequency security constraints, and then the commercial solver Gurobi is used to solve the objective function of the frequency regulation reserve allocation optimization model to obtain the operating cost of the demand response resources when the total cost of the objective function is the lowest, and then the reserve of the demand response resources is obtained. Based on the obtained reserve of the demand response resources, instructions are issued to the demand response resources through the dispatching system of the power system to allocate frequency regulation reserves. Specific embodiment:

[0150] The HDMR method is expressed using a high-dimensional model, and a large amount of power fluctuation data of the power system is randomly generated based on Gaussian sampling -ΔP D , spare parts for generator sets Backup for data and variable frequency air conditioning clusters Backup of data and energy storage battery clusters Data and backup for electric vehicles The data is composed of sample data, and a total of m groups of sample data are obtained. Then, the m groups of sample data are input into the frequency dynamic model for simulation calculation, and m groups of maximum frequency offset Δf data are output, that is, the frequency offset Δf(s) of the power system.

[0151] The m groups of maximum frequency offset Δf data are divided into a training set, a validation set, and a test set. The training set is input into the frequency dynamic model for training, and the validation set is input into the trained frequency dynamic model to test the error of the frequency dynamic model and select the model parameters with low error until the error of the frequency dynamic model is reduced to within the preset value δ, and the model parameters at this time are obtained as the model parameters of the frequency dynamic model; the test set is input into the verified frequency dynamic model to evaluate the error of the frequency dynamic model, and finally the maximum frequency offset Δf and the standby power of the generator set are output. Inverter air conditioning cluster backup Energy storage battery cluster backup and electric vehicle cluster backup A simple mapping relationship between Δf=DDFR(x).

[0152] Next, the simple mapping relationship Δf=DDFR(x) obtained by the HDMR method is used to express the high-dimensional model to replace the complex mapping relationship of the frequency safety constraint of the complex high-order nonlinearity That is, the low-order quadratic frequency safety constraints of the HDMR method are expressed by a high-dimensional model to replace the complex high-order nonlinear frequency safety constraints, thereby transforming the original implicit high-order nonlinear frequency safety constraint cost scheduling problem into a bilinear programming problem.

Claims

1. A method for allocating frequency reserve for demand response resources considering frequency security constraints, characterized in that: The following steps are involved: Step 1) constructing a power system, including a dispatching system, a generator set, a transmission network and demand response resources of the power system, and obtaining frequency response parameters and cost parameters of the generator set and the demand response resources; Step 2) constructing a frequency dynamic model of the power system according to the frequency response parameters of the generator sets and the demand response resources; Step 3) constructing a frequency regulation reserve allocation optimization model for the power system, including an objective function and constraints, according to the cost parameters of the generator sets and demand response resources; According to the frequency dynamic model, the high-order nonlinear frequency safety constraints of the objective function are established; Step 4) using a high-dimensional model representation (HDMR) method to convert the high-order nonlinear frequency safety constraints of the objective function into low-order quadratic frequency safety constraints; Step 5) Use McCormick envelope to process the low-order quadratic frequency security constraints of the objective function, and then use the commercial solver gurobi to solve the objective function, and allocate frequency reserve to demand response resources according to the solution results; In the step 1), a plurality of generators form a generator set; Demand response resources include energy storage battery clusters, variable frequency air conditioner clusters and electric vehicle clusters; a number of energy storage batteries form an energy storage battery cluster, a number of variable frequency air conditioners form a variable frequency air conditioner cluster, and a number of electric vehicles form an electric vehicle cluster; In the step 2), according to the frequency response parameters of the generator sets and the demand response resources, the frequency dynamic model of the power system is constructed as follows: 2HsΔf(s)+DΔf(s)=ΔP G (s)+ΔP DR (s)-ΔP D (s) Where H and D represent the inertia and load damping of the power system, respectively, s represents the operator in the frequency domain of the frequency dynamic model, Δf(s) represents the frequency offset of the power system, and ΔP G (s) represents the output change of the generator set, ΔP DR (s) represents the power stability support amount of the demand response resource, ΔP D (s(represents the power disturbance of the power system; The output change of the generator set ΔP G (s) are as follows: Where R represents the droop control coefficient of the generator set, 1 / R represents the primary frequency regulation control parameter of the generator set; F HP Indicates the power ratio of the high-pressure turbine section of the generator set, T RH Represents the reheat time constant of the reheat turbine of the generator set, T G Indicates the speed governor parameters of the generator set, T CH Indicates the turbine parameters of the generator set; R, F HP 、T RH 、T G and T CH That is the frequency response parameter of the generator set; In the step 3), according to the cost parameters of the generator set and the demand response resource, the objective function of the frequency regulation reserve allocation optimization model of the power system is constructed as follows: Among them, F represents the total cost of the objective function, min represents the lowest value; C G , C IAC , C B and C EV They represent the operating costs of the generator set, variable frequency air conditioning cluster, energy storage battery cluster and electric vehicle cluster respectively; T is the number of time periods of the power system; N G 、N IAC 、N B and N EV They represent the number of generators in the generator set, variable frequency air conditioners in the variable frequency air conditioner cluster, energy storage batteries in the energy storage battery cluster, and electric vehicles in the electric vehicle cluster; c Gi and c Ri They represent the output cost and standby cost of the i-th generator in the generator set, and They represent the output and reserve of the i-th generator in the generator set during period t; c IACj represents the backup cost of the jth variable frequency air conditioner in the variable frequency air conditioner cluster, represents the standby of the jth variable frequency air conditioner in the variable frequency air conditioner cluster during period t; c Bw represents the backup cost of the wth energy storage battery in the energy storage battery cluster, represents the backup of the wth energy storage battery in the energy storage battery cluster during period t; c EVv represents the backup cost of the vth electric vehicle in the electric vehicle cluster, represents the reserve of the vth electric vehicle in the electric vehicle cluster during period t; c Gi and c Ri That is the cost parameter of the generator set, c IACj 、c Bw and c EVv That is, the cost parameter of the demand response resource; Generator output cost c Gi It is represented using a quadratic cost model as follows: Among them, a i 、b i 、c i They represent the operating cost coefficient of the generator respectively; In the step 3), in the process of the power generation units of the power system transmitting power to the demand response resources through the transmission network, the constraints of the established frequency regulation reserve allocation optimization model are as follows: a) Load power balance constraints: Among them, N G Indicates the number of generators in the generator set, represents the output of the i-th generator in the generator set during period t, It represents the sum of all load powers in the power system during period t; b) Power flow transmission capacity constraints: Among them, T li represents the power transmission allocation coefficient of the node of the power system where the i-th generator in the generator group is located to the l-th branch in the transmission network; K represents the number of nodes in the power system, T lk represents the power transmission allocation coefficient of the kth node to the lth branch in the transmission network, represents the load of the kth node in period t, represents the power flow transmission capacity of the lth branch in the transmission network; c) Generator spinning reserve capacity constraints: in, represents the upper limit of the output of the i-th generator in the generator set during period t, and r represents the reserve coefficient of the power system; d) Generator output and reserve capacity constraints: in, It represents the lower limit of the output of the i-th generator in the generator set during the t period; e) Demand response resource output capacity constraints: in, and They represent the upper and lower output limits of the jth variable frequency air conditioner in the variable frequency air conditioner cluster at time period t respectively; and They represent the upper and lower output limits of the wth energy storage battery in the energy storage battery cluster at time period t respectively; and They represent the upper and lower limits of the output of the vth electric vehicle in the electric vehicle cluster at time period t; represents the standby of the jth variable frequency air conditioner in the variable frequency air conditioner cluster during period t; represents the backup of the wth energy storage battery in the energy storage battery cluster during period t; represents the reserve of the vth electric vehicle in the electric vehicle cluster during period t; f) Generator and demand response resource ramping constraints: in, represents the output of the i-th generator in the generator set at time t-1, Ru i and Rd i They respectively represent the upward climbing rate and downward climbing rate of the i-th generator in the generator set.

2. According to claim 1, a method for allocating frequency reserve for demand response resources considering frequency security constraints is characterized by: The power stability power support amount ΔP of the demand response resource DR (s) is the power change of the variable frequency air conditioning cluster ΔP IAC (s), power change of the energy storage battery cluster ΔP B (s) and the power change of the electric vehicle cluster ΔP EV (s) is obtained by adding by an adder; The inverter air conditioner cluster is installed in the room. The power change of the inverter air conditioner cluster is ΔP. IAC (s) are as follows: Among them, θ P represents the operating power constant coefficient of the variable frequency air conditioning cluster, A represents the equivalent primary frequency modulation control parameter of the variable frequency air conditioning cluster, T a Represents the equivalent speed regulator time constant, T a =(C room V room ρ A ) / H r , C room 、V room , A denote the specific heat capacity, volume and air density of the room where the variable frequency air conditioning cluster is installed, H r represents the equivalent heat transfer coefficient inside and outside the room, T c represents the time constant of the compressor of the variable frequency air conditioning cluster, Φ IAC represents the equivalent controller parameter, Φ IAC =θ Q / H r ,θ Q represents the operating power constant coefficient of the variable frequency air conditioning cluster, C(s) represents the PI controller parameter of the variable frequency air conditioning cluster, C(s)α+β / s, α and β represent the proportional coefficient and integral coefficient of the PI controller of the variable frequency air conditioning cluster respectively; θ P , A 、T a 、T c , Φ IAC and C(s) are the frequency response parameters of the variable frequency air conditioning cluster; Power change of energy storage battery cluster ΔP B (s) are as follows: Among them, T B Represents the response time constant of the energy storage battery, R B Indicates the regulation coefficient of the energy storage battery, represents the response characteristics of the energy storage battery cluster, Represents the frequency response coefficient of the energy storage battery cluster; T B and R B That is the frequency response parameter of the energy storage battery cluster; Power change of electric vehicle cluster ΔP EV (s) are as follows: Among them, T EV represents the response time constant of the electric vehicle cluster represents the response characteristics of the electric vehicle cluster; K EV represents the frequency response coefficient of the electric vehicle cluster, R EV Indicates the adjustment coefficient of electric vehicles; T EV and K EV That is the frequency response parameter of the electric vehicle cluster.

3. According to claim 1, a method for allocating frequency reserve for demand response resources considering frequency security constraints is characterized in that: In the step 3), according to the frequency deviation of the power system output by the frequency dynamic model, the high-order nonlinear frequency security constraint of the objective function is obtained as follows: Among them, Δf represents the maximum frequency deviation of the power system, which means that the power system generates power disturbance in time period t. When the generator set, variable frequency air conditioning cluster, energy storage battery cluster and electric vehicle cluster provide and The frequency regulation reserve makes the maximum frequency deviation of the power system; FR () represents the frequency dynamic model, x represents the disturbance of power disturbance and FM standby A collection of; Indicates the frequency deviation limit of the power system; represents the reserve of the i-th generator in the generator set during period t, represents the standby of the jth variable frequency air conditioner in the variable frequency air conditioner cluster during period t, represents the backup of the wth energy storage battery in the energy storage battery cluster during period t, It represents the reserve of the vth electric vehicle in the electric vehicle cluster during period t.

4. The method for allocating frequency modulation reserve for demand response resources considering frequency security constraints according to claim 1, characterized in that: In the step 5), the McCormick envelope is used to process the low-order quadratic frequency security constraints, and then the commercial solver gurobi is used to solve the objective function of the frequency regulation reserve allocation optimization model to obtain the operating cost of the demand response resources when the total cost of the objective function is the lowest, and then the reserve of the demand response resources is obtained. Based on the obtained reserve of the demand response resources, instructions are issued to the demand response resources through the dispatching system of the power system to perform frequency regulation reserve allocation.