CN113852091B - A reactive voltage regulation method for renewable energy grid-connected power based on MPC - Google Patents

Abstract

The present invention proposes a reactive voltage regulation method for grid-connected renewable energy based on MPC. After renewable energy is connected to the grid, it is checked whether the voltage of each node exceeds the limit, the sensitivity matrix of the node voltage to the active and reactive power injected into the node is solved, and the node with the most serious degree of exceeding the limit is selected for active and reactive regulation. First, the change of future input power is predicted based on the historical operating status information, and the node voltage after disturbance is predicted through the sensitivity matrix. Then, with the goal of minimizing the node voltage offset and maximizing the reactive margin of the regulating equipment, the power balance, the upper and lower limits of the node voltage, the capacity of the grouped capacitor, the SVC constraint and the DFIG constraint are used as the constraint conditions to obtain the optimal result that meets the constraints. Finally, the reactive output of the wind turbine, the number of parallel capacitor groups switched on and off and the reactive output of the SVC are allocated. This strategy smoothes the equipment output and reduces the network loss while ensuring the voltage qualification rate, thereby improving the economy of voltage control.

Description

A reactive voltage regulation method for renewable energy grid-connected power based on MPC

Technical Field

The present invention relates to the field of automatic voltage control, and in particular to a strategy study of participating in coordinated control of reactive power and voltage of a power grid after new energy sources are incorporated into the power grid.

Background technique

Modern life cannot do without electricity, and the development of social economy is closely related to the power industry. Today, traditional fossil energy is facing the problem of depletion, so the proportion of renewable energy power generation connected to my country's power grid is increasing. New energy is transmitted over long distances through multiple ultra-high voltage AC and DC lines, and AC and DC power grids containing a high proportion of renewable energy are more sensitive to voltage.

At present, the grid voltage is mainly regulated by automatic voltage control AVC to achieve safe and economic optimization of the grid. The AVC system is divided into an AVC master station at the grid end and an AVC substation at the production station end. The master station is deployed in the dispatching center. With the minimum grid loss as the optimization goal, it calculates the reactive power that each production station needs to output and sends it to the substation through dispatching. The substation performs reactive closed-loop control with the highest bus voltage qualification rate, the most reasonable reactive compensation of the compensation device, and the optimal reactive output of the generator as the goals. The master station and the substation cooperate with each other to jointly perform reactive optimization and voltage control.

When adding a wind farm to the power grid, we must consider the voltage fluctuation of the power farm itself and the voltage disturbance caused by the real-time fluctuation characteristics of wind power. As an important reactive power source, the wind farm composed of double-fed wind turbines should play its due role in stabilizing the grid voltage and compensating reactive power. We combine traditional reactive voltage regulation equipment with double-fed generators to regulate voltage.

Summary of the invention

The present invention introduces model predictive control (MPC) into the new energy grid for rapid voltage control, with the aim of sensing the voltage change trajectory in advance and changing the output of the voltage regulating equipment to prevent voltage from exceeding the limit and achieve smooth control.

The present invention is achieved through the following technical solutions:

A reactive voltage regulation method for renewable energy grid-connected power based on MPC, characterized by comprising:

Step 1: Collect the distribution network topology and line parameters and the voltage of each node, and check whether the voltage of each node exceeds the limit. If it exceeds the limit, execute step 2, otherwise enter the next collection cycle and repeat this step;

Step 2: Solve the sensitivity matrix of node voltage to injected reactive power, and then select the node with the most serious over-limit as the regulating node, predict the voltage in the future time domain by the disturbance, and obtain the optimization result after comprehensive optimization with reactive power as the regulating variable;

Step 3: According to the optimization results, the reactive output of the wind turbine, the number of shunt capacitor groups switched on and off and the reactive output of the SVC.

In the above-mentioned MPC-based new energy grid-connected reactive voltage regulation method, it is characterized in that:

In step 2, the optimization is based on the objective function and constraints:

Objective function: The comprehensive goal is to achieve the best voltage stability level at the wind turbine terminal, the most balanced reactive power margin, and the maximum reactive power margin of the dynamic compensation equipment;

Among them, w ₁ +w ₂ +w ₃ +w ₄ =1, which is the weight factor; w ₁ , w ₂ , w ₃ , w ₄ are the weight factors of voltage deviation value, wind turbine reactive margin, SVC reactive margin, and capacitor reactive margin respectively;

_Vg is the actual value of the wind turbine node voltage, is a reference value, taking the unit value as 1; it has a balanced reactive power margin to ensure that chain disconnection accidents are avoided;

Q _j is the reactive power output of the wind turbine, q is the number of wind turbines, Q _jmax and Q _jmin are the maximum and minimum reactive power values of the wind turbines, respectively;

Q _x is the reactive power output of SVC, s is the number of SVCs, Q _xmax and Q _xmin are the maximum and minimum reactive power values of SVC respectively;

_Qc is the reactive power output of the grouped switched capacitor bank, a is the number of capacitor banks, _Qcmax and _Qcmin are the maximum and minimum reactive power values of the capacitor bank respectively;

The constraints are:

_Vimin ≤V _i ≤V _imax

Q _cmin ≤Q _c ≤Q _cmax

Q _xmin ≤Q _x ≤Q _xmax

_Vimin and _Vimax represent the upper and lower limits of the node voltage respectively, _Vi is the actual voltage; _Pi is the active power injected into the node, _Qi is the reactive power injected into the node; _Gij is the branch conductance, _Bij is the branch susceptance; _θij is the phase angle between the voltages at nodes i and j.

In the above-mentioned MPC-based new energy grid-connected reactive voltage regulation method, step 2 specifically includes:

Step 2.1, determine the voltage sensitivity of the reactive power regulation equipment;

Step 2.2, obtain the Jacobian matrix of the power grid;

Step 2.3: Use MPC to predict future time domain voltage.

In the above-mentioned reactive voltage regulation method for renewable energy grid-connected based on MPC, in step 2.1, the reactive regulation equipment includes a group switching capacitor bank, a static VAR compensator SVC and a doubly-fed wind turbine DFIG;

In the group switching capacitor bank, the sensitivity of the key node voltage to the number of capacitor switching groups is:

_C0 is the capacity of a single capacitor; _Vj is the voltage at the node where the capacitor is connected, and ω is the angular frequency;

S _ij represents the voltage sensitivity of the j-th group head and tail transformer to the key node i;

In the static VAR compensator SVC, the sensitivity of the node voltage to reactive power is:

B is the susceptance of SVC, U is the voltage across SVC;

In a doubly-fed wind turbine DFIG, the sensitivity of node k to the reactive power output of node j is:

When K is greater than j,

When K is less than j,

_Xn is the branch impedance and _Un is the node voltage.

In the above-mentioned MPC-based renewable energy grid-connected reactive voltage regulation method, in step 2.2, the Jacobi matrix is obtained based on the following formula:

S＝J ^-1 .

In the above-mentioned reactive voltage regulation method based on MPC for renewable energy grid connection, in step 2.3, MPC is used to predict the future time domain voltage based on the linearized voltage second-level prediction model of dynamic control response parameters.

Wherein V(k) and V(k+1) represent the node voltages at the current time k and the next time k+1 respectively; _ΔQw , _ΔPL and _ΔQL represent the active and reactive power fluctuations of the wind farm and the load; _ΔQq , _ΔQsvc and _ΔQc represent the reactive power regulation of the wind turbine, SVC and capacitor bank respectively.

Therefore, the present invention has the following advantages: the method establishes a linearized voltage and network loss prediction model within a future time window based on the dynamic control response parameters calculated by real-time gradient, updates the voltage reference value online, establishes and solves a multi-objective rolling optimization model considering voltage deviation, network loss and equipment adjustment amount. This strategy makes up for the deficiency of the secondary voltage control of the AVC system in real-time performance. Compared with existing research, this strategy smoothes the equipment output and reduces network loss while ensuring the voltage qualification rate, thereby improving the economy of voltage control.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a flow chart of a reactive voltage regulation method for renewable energy grid-connected based on MPC.

FIG2 is a basic structure of a static VAR compensator SVC.

FIG3 is a basic structure of a double-fed asynchronous wind turbine generator DFIG.

Figure 4 is a branch power flow model with wind turbines connected to the grid.

Detailed ways

The technical solution of the present invention is further specifically described below through embodiments and in conjunction with the accompanying drawings.

Example:

This embodiment specifically includes the following steps:

Step 1: Determine the voltage sensitivity of the reactive power regulation device

Common reactive power regulation devices include:

1. Group switching capacitor bank

It is an important reactive power compensation device in the distribution network. It can change the reactive power output discretely by changing the number of capacitors. It can only output capacitive reactive power, but not inductive reactive power. Installing it on the load side can effectively alleviate the problem of low voltage at the end of the feeder. The compensation capacity of the capacitor group is:

Where ΔQ is the compensation power, Δn represents the number of capacitor switching groups, _VC is the voltage at the capacitor access node, ω is the angular frequency, and _C0 is the unit regulating capacitance.

Therefore, the sensitivity of the key node voltage to the number of capacitor switching groups is:

S _ij represents the voltage sensitivity of the j-th group head and tail transformer to the key node i.

2. Static VAR Compensator SVC

The reactive power generated is:

Among them, the The sensitivity of node voltage to reactive power can be obtained as:

3. Doubly-fed wind turbine DFIG

In a doubly-fed wind turbine, the stator is directly connected to the grid, and the rotor is connected to the grid through an AC-DC-AC converter. The AC-DC-AC converter consists of two voltage-type PWM converters connected back to back: the one close to the rotor is called the rotor-side converter, and the one close to the grid is called the grid-side converter.

P _mec is the mechanical power input by the wind turbine, P _s and Q _s are the active and reactive power emitted by the stator, P _c and Q _c are the active and reactive power input from the grid by the grid-side converter, and P _g and Q _g are the active and reactive power flowing into the grid from the wind turbine.

Ignoring the losses in the stator and rotor windings,

P _mec = P _s - _{P r}

_Ps ＝ _Pmec /(1-s)

P _r = sP _mec /(1-s)

Ignoring the switching loss and line loss of the power converter, according to the principle of energy conservation, we can get:

P _c =P _r

Ignoring system losses, according to the power flow direction specified in the figure:

_Pg ＝ _Ps - _Pc ＝ _Pmec

The reactive power of the doubly-fed induction generator is:

_Qg ＝ _Qs - _Qc

The operating range of reactive power on the stator side is usually constrained by three aspects: the rotor-side converter current and the stator current stability. The main factor is the rotor-side converter current, namely:

Where _Ls and _Lm are the stator inductance and excitation inductance respectively, _Us is the stator voltage peak, ω is the stator current angular frequency, _Ir and _Irmax are the rotor-side converter current and its maximum value.

The reactive power range of the stator side is obtained as follows:

because but

So we get

Q _gmax =Q _smax -Q _cmin

Q _gmin =Q _smin -Q _cmax

Assume that the reactive power output of the wind turbine connected to the distribution network node j is Q _wj , and the sensitivity of the voltage U _i at the node i to Q _wj is S _ij , that is, In the radial distribution network, as shown in Figure

Ignoring line losses, the voltage loss between nodes i and i-1 is:

Then the voltage at node i is:

Assume that node p is connected to a wind turbine and the output of the turbine is P _wp +jQ _wp . When node i is before p, the voltage of node i is:

When node i is after node p,

If a distributed wind turbine is connected,

The sensitivity of node k to the reactive power output of node j can be obtained as

When K is greater than j,

When K is less than j,

Step 2: Obtain the Jacobian matrix of the power grid

Jacobian Matrix

S＝J ^-1

Step 3: Use MPC to predict future time domain voltage

The linearized voltage second-level prediction model based on dynamic control response parameters can be expressed as

Where V(k) and V(k+1) represent the node voltages at the current time k and the next time k+1, respectively. _ΔQw , _ΔPL , and _ΔQL represent the active and reactive power fluctuations of the wind farm and the load. _ΔQq , _ΔQsvc , and _ΔQc represent the reactive power regulation of the wind turbine, SVC, and capacitor bank, respectively.

Step 4: Determine the objective function and constraints

Objective function: The comprehensive goal is to achieve the best voltage stability level at the wind turbine terminal, the most balanced reactive power margin, and the maximum reactive power margin of the dynamic compensation equipment.

Wherein w ₁ +w ₂ +w ₃ +w ₄ =1, which is the weight factor.

_Vg is the actual value of the wind turbine node voltage, As a reference value, the per unit value is 1. There is a balanced reactive power margin to ensure that chain disconnection accidents are avoided.

Q _j is the reactive power output of the wind turbine, q is the number of wind turbines, Q _jmax and Q _jmin are the maximum and minimum reactive power values of the wind turbines respectively.

Q _x is the reactive power output of SVC, s is the number of SVCs, Q _xmax and Q _xmin are the maximum and minimum reactive power values of SVC respectively.

_Qc is the reactive power output of the grouped switched capacitor bank, a is the number of capacitor banks, _Qcmax and _Qcmin are the maximum and minimum reactive power values of the capacitor bank, respectively.

The constraints are:

_Vimin ≤V _i ≤V _imax

Q _cmin ≤Q _c ≤Q _cmax

Q _xmin ≤Q _x ≤Q _xmax

The specific embodiments described herein are merely examples of the spirit of the present invention. A person skilled in the art of the present invention may make various modifications or additions to the specific embodiments described or replace them in a similar manner, but this will not deviate from the spirit of the present invention or exceed the scope defined by the appended claims.

Claims (1)

1. The new energy grid-connected reactive voltage regulation method based on the MPC is characterized by comprising the following steps of:

Step 1, collecting topology and line parameters of a power distribution network and voltage of each node, checking whether the voltage of each node is out of limit, if so, executing step 2, otherwise, entering the next collecting period, and repeating the step;

step 2, solving a sensitivity matrix of node voltage to injected reactive power, then selecting the node with the most serious out-of-limit as an adjusting node, predicting the voltage in the future time domain through the disturbance quantity, and comprehensively optimizing by taking the reactive power as an adjusting variable to obtain an optimized result;

Step 3, according to the optimization result, the reactive power output of the wind motor, the switching number of the parallel capacitor bank and the SVC reactive power output;

in step 2, the optimization is based on objective functions and constraints:

objective function: the method comprises the steps that the maximum voltage stability level of a wind motor end, the maximum reactive margin balance and the maximum reactive margin of dynamic compensation equipment are taken as comprehensive targets;

Wherein w ₁+w₂+w₃+w₄ =1, is a weight factor; w ₁、w₂、w₃、w₄ is the weight factors of the voltage deviation value, the fan reactive margin, the SVC reactive margin and the capacitance reactive margin respectively;

V _g is the actual value of the wind turbine node voltage, Taking a per unit value of 1 as a reference value; the balanced reactive margin is provided to ensure that the occurrence of the interlocking off-grid accident is avoided;

q _j is reactive power output of the wind motor, Q is the number of the wind motor, and Q _jmax and Q _jmin are maximum reactive power value and minimum reactive power value of the wind turbine generator respectively;

Q _x is reactive power output of SVC, s is number of SVC, and Q _xmax and Q _xmin are maximum and minimum SVC reactive power values respectively;

Q _c is reactive power output of the group switching capacitor bank, a is the number of the capacitor bank, and Q _cmax and Q _cmin are the maximum reactive power value and the minimum reactive power value of the capacitor bank respectively;

The constraint conditions are as follows:

V_imin≤V_i≤V_imax

Q_cmin≤Q_c≤Q_cmax

Q_xmin≤Q_x≤Q_xmax

V _imin and V _imax respectively represent upper and lower limits of the node voltage, and V _i is the actual voltage; p _i is node injection active power, Q _i is node injection reactive power; g _ij is the branch conductance, B _ij is the branch susceptance; θ _ij is the phase angle between the voltages of nodes i and j;

The step 2 specifically comprises the following steps:

Step 2.1, determining the voltage sensitivity of the reactive power regulation equipment;

step 2.2, solving a jacobian matrix of the power grid;

step 2.3, predicting future time domain voltage by using MPC;

In step 2.1, the reactive power regulation device comprises a group switching capacitor bank, a static var compensator SVC and a doubly fed wind generator DFIG;

in the group switching capacitor bank, the sensitivity of the key node voltage to the number of the capacitor switching bank is as follows:

C ₀ is the capacity of a single capacitor; v _j is the voltage at the capacitor access node, ω is the angular frequency;

s _ij represents the voltage sensitivity of the j-th grouping head-tail transformer to the key node i;

in the static var compensator SVC, the sensitivity of the node voltage to reactive power is:

b is susceptance of SVC, U is voltage at two ends of SVC;

in the doubly-fed wind generator DFIG, the sensitivity of the node k to the reactive power output by the node j is as follows

When the K is larger than the j, the method comprises the steps of,

When the K is smaller than the j, the method comprises the steps of,

X _n is the branch impedance, U _n is the node voltage;

In step 2.2, the jacobian matrix is obtained based on the following formula

S＝J^-1；

In step 2.3, a linearization voltage second-level prediction model based on dynamic control response parameters is used for predicting future time domain voltage by using MPC

Wherein V (k) and V (k+1) respectively represent node voltages at the current moment k and the next moment k+1, and DeltaQ _w、ΔP_L、ΔQ_L represents the active and reactive power fluctuation quantity of the wind power plant and the load; Δq _q、ΔQ_svc、ΔQ_c represents the reactive power regulation of the wind turbine, SVC and capacitor bank, respectively.