CN109787268A - It is a kind of meter and photovoltaic injecting power method for estimating state - Google Patents

Abstract

The invention provides a state estimation method that takes into account photovoltaic injection power, including: (1) initializing network parameters, establishing an equivalent model of photovoltaic power station and an injection power measurement equation; (2) calculating the branch power of the distribution network measurement, branch current amplitude measurement, node injection power measurement and voltage amplitude measurement are converted into the amplitude and phase angle of the branch current; (3) build a measurement matrix, the branch current I, the inverter The amplitude modulation ratio M and the phase modulation angle A are used as state quantities, and an estimation algorithm is used to estimate the state and correct the state variables; (4) Repeat step (3) until the convergence conditions are met, obtain the state estimation result, and upload it to the real-time database. The invention improves the estimation accuracy, thereby laying a foundation for the intelligent, safe and stable operation of the distribution network.

Description

A State Estimation Method Considering Photovoltaic Injection Power

technical field

The invention relates to the technical field of distribution network operation and control, and relates to a state estimation method that takes into account photovoltaic injection power.

Background technique

With the continuous influx of distributed power sources such as photovoltaics and wind power into the power grid, the fluctuating and intermittent distributed power sources put forward higher requirements for the safe, stable and efficient operation of the power grid, and the effective estimation of the power grid is an important guarantee for the automation of power grid dispatching . Due to the ever-expanding scale of the power grid, a large amount of data is generated in real time, which is often disturbed by random errors, meter errors and model errors, resulting in inaccurate data. Under the condition of ensuring the observability of the system, the state estimation method improves the precision and accuracy of the measurement data according to the redundancy of the actual measurement device. Compared with traditional power flow calculation, state estimation is extended on the basis of power flow calculation. It is based on more data and involves data identification and screening, so the estimated electrical results are more accurate. Therefore, it is of great significance to study the state estimation method with photovoltaic power source to analyze the effect of the penetration of photovoltaic power source on the state estimation of the power grid.

The existing power system state estimation methods usually choose to treat the photovoltaic power source as the load, ignoring the type, model and state variables of the photovoltaic power source, that is, only regard the photovoltaic power source as a general load node. This estimation method can only obtain the voltage amplitude and phase angle of the node, which can neither determine the operating state of the photovoltaic power source nor distinguish the characteristics of the photovoltaic power source. In the prior art, the type, model and state variables of photovoltaic power sources are ignored, and the accuracy is low. Therefore, it is necessary to provide a distribution network state estimation method that takes into account the characteristics of photovoltaic power sources.

SUMMARY OF THE INVENTION

In order to overcome the deficiencies of the prior art, the present invention provides a state estimation method that takes into account the photovoltaic injection power, thereby improving the accuracy and improving the intelligence and stability of the power distribution network.

The present invention provides a distribution network state estimation method taking into account photovoltaic injection power, the method comprising:

(1) Initialize the network parameters, establish the equivalent model of the photovoltaic power station and the injection power measurement equation;

(2) Convert the branch power measurement, branch current amplitude measurement, node injection power measurement and voltage amplitude measurement of the distribution network into the amplitude and phase angle of the branch current;

(3) Constructing a measurement matrix, using the branch current I, the inverter amplitude modulation ratio M and the phase modulation angle A as state quantities, and using an estimation algorithm to estimate the state and correct the state variables;

(4) Step (3) is repeated until the convergence condition is satisfied, the state estimation result is obtained, and the real-time database is uploaded.

The working principle of the invention: According to the characteristics of the photovoltaic power source and the influence on the power flow of the power distribution network, a state estimation method suitable for the power distribution network with photovoltaic power source is studied, thereby laying a foundation for the intelligent, safe and stable operation of the power distribution network. Base. Considering the characteristics of photovoltaic power sources, specifically the output characteristics of photovoltaic cells and photovoltaic inverters, an accurate photovoltaic power generation equivalent model is established. The model is incorporated into the distribution network system, the measurement matrix is constructed, and the branch current, inverter amplitude modulation ratio and phase modulation angle are used as state quantities, the estimation method is used to estimate the state, and the state variables are corrected to obtain the distribution. Real-time operating status of power grid and photovoltaic power supply.

In the above scheme, for optimization, further, in the step (1), the equivalent model of the photovoltaic power station and the injection power measurement equation are:

Among them, P _ps is the active power of photovoltaic power generation, Q _ps is the reactive power of photovoltaic power generation; U is the amplitude of the AC side voltage, θ is the phase angle of the AC side voltage; U _cell is the voltage of the photovoltaic cell; _Na is the The number of photovoltaic cells connected in series in a single photovoltaic array; _Naa is the number of photovoltaic modules contained in all photovoltaic arrays; M is the inverter amplitude modulation ratio in the photovoltaic power generation model, A is the phase modulation angle in the photovoltaic power generation model; z _ab , φ _ab , z _bc , φ _bc , z _ac , and φ _ac are the equivalent impedance amplitude and phase angle of the conversion of the Y-type circuit of the AC part of the equivalent circuit of the photovoltaic power generation model into the Δ-type equivalent circuit.

Further, the step (2) includes converting branch active power measurement, branch reactive power measurement, current amplitude measurement, node power injection measurement, and voltage amplitude measurement into branch current measurements. Amplitude and phase angle.

Further, in the step (2), the imaginary and real parts of the branch active power measurement, the imaginary and real parts of the branch reactive power measurement, the imaginary and real parts of the node injection power measurement and the imaginary and real parts of the voltage amplitude measurement are used. are expressed as branch current amplitude and branch current phase angle respectively; the current amplitude measurement is expressed as branch current.

Further, the estimation algorithm is to iteratively calculate the equivalent model of the photovoltaic power station and the injection power measurement equation in step (1) according to the least squares method.

Further, the estimation algorithm in the step (3) includes: constructing an equation according to the state variables: x=(I,δ,M,A), and establishing a Jacobian matrix according to the equation x=(I,δ,M,A) , the Jacobian matrix is a state estimation model considering multiple state quantities, and the Jacobian matrix is:

Among them, the subscript ps is the power corresponding to the photovoltaic field node, and The dimension of is consistent with the number of photovoltaic power stations in the distribution network.

In the present invention, the influence of the connection of the photovoltaic power source is considered, and the equivalent model of the photovoltaic power station and the injection power measurement equation are established; the branch power measurement, the branch current amplitude measurement, the node injection power measurement and the voltage amplitude are measured. The measurement is converted into the amplitude and phase angle of the branch current; the measurement matrix is constructed, and the current amount of each branch, the amplitude modulation ratio of the inverter and the phase modulation angle are used as state quantities, and the estimation algorithm is used to estimate the state and correct it. state variables to obtain accurate state estimation results. On the basis of absorbing the transformation method based on branch voltage measurement, a distribution network state estimation algorithm based on multiple state variables such as branch current is proposed. In addition to using the power measurement in the distribution network, it also uses the current amplitude Measurement and voltage amplitude measurement, because the latter two measurements occupy a large proportion in the distribution network measurement, so the algorithm effectively improves the measurement redundancy and the observability and accuracy of state estimation.

Beneficial effects of the present invention:

Effect 1: After the photovoltaic power source is integrated into the distribution network, the power distribution system changes from the previous single power supply mode to the multiple power supply mode, and the power flow changes accordingly;

The second effect is that the incorporation of photovoltaic power sources increases the order of the measurement matrix and improves the estimation accuracy;

The third effect is to lay the foundation for the intelligent, safe and stable operation of the distribution network.

Description of drawings

Fig. 1 is the photovoltaic power generation equivalent model of the present invention;

Fig. 2 is the equivalent model diagram of AC side in the present invention;

Fig. 3 is the π-type equivalent model of circuit in the present invention;

Fig. 4 is the single-line mode diagram of the power injection amount side in the present invention;

Fig. 5 is a flow chart of the state estimation method of the distribution network of the present invention.

Detailed ways

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

The present invention provides a state estimation method that takes into account the photovoltaic injection power, as shown in FIG. 5 , the method includes:

(1) Initialize the network parameters, establish the equivalent model of the photovoltaic power station and the injection power measurement equation;

(2) Convert the branch power measurement, branch current amplitude measurement, node injection power measurement and voltage amplitude measurement of the distribution network into the amplitude and phase angle of the branch current;

(3) Constructing a measurement matrix, using the branch current I, the inverter amplitude modulation ratio M and the phase modulation angle A as state quantities, and using an estimation algorithm to estimate the state and correct the state variables;

(4) Step (3) is repeated until the convergence condition is satisfied, the state estimation result is obtained, and the real-time database is uploaded.

The work flow of this embodiment: as shown in Figure 5, taking into account the influence of the photovoltaic power supply connection, establish an equivalent model considering the photovoltaic power station and the injection power measurement equation; measure the branch power and branch current amplitude. , The node injection power measurement and voltage amplitude measurement are converted into the amplitude and phase angle of the branch current; the measurement matrix is constructed, and the current amount of each branch, the inverter amplitude modulation ratio and the phase modulation angle are used as the state The state estimation algorithm is used to estimate the state variables, and the state variables are corrected, so as to obtain accurate state estimation results.

Specifically, as shown in Fig. 1, in the step (1), the equivalent model of the photovoltaic power station and the injection power measurement equation are:

Among them, P _ps is the active power of photovoltaic power generation, Q _ps is the reactive power of photovoltaic power generation; U is the amplitude of the AC side voltage, θ is the phase angle of the AC side voltage; U _cell is the voltage of the photovoltaic cell; _Na is the The number of photovoltaic cells connected in series in a single photovoltaic array; _Naa is the number of photovoltaic modules contained in all photovoltaic arrays; M is the inverter amplitude modulation ratio in the photovoltaic power generation model, A is the phase modulation angle in the photovoltaic power generation model; z _ab , φ _ab , z _bc , φ _bc , z _ac , and φ _ac are respectively the equivalent impedance amplitude and phase angle of the Y-type circuit of the AC part of the equivalent circuit of the photovoltaic power generation model converted into the Δ-type equivalent circuit as shown in Figure 2 .

Specifically, the step (2) includes converting branch active power measurement, branch reactive power measurement, current amplitude measurement, node power injection measurement, and voltage amplitude measurement into branch current measurements. Amplitude and phase angle.

Specifically, in the step (2), the imaginary and real parts of the branch active power measurement, the imaginary and real parts of the branch reactive power measurement, the imaginary and real parts of the node injection power measurement, and the imaginary and real parts of the voltage amplitude measurement are expressed as branch current amplitude and branch current phase angle respectively; the current amplitude measurement is expressed as branch current.

The various measurements in the power distribution system in step (2) cooperate with each other, and they include branch active power measurement, branch reactive power measurement, current amplitude measurement, and power injection measurement, that is, pseudo-measurement. , voltage amplitude measurement. The line can be equivalent to a π-type model, and Fig. 3 is a line model of the branch from busbar k to busbar m.

Specifically, the estimation algorithm in the step (3) includes: constructing an equation according to state variables: x=(I,δ,M,A), and establishing a Jacobian matrix according to the equation x=(I,δ,M,A) , the Jacobian matrix is a state estimation model considering multiple state quantities, and the Jacobian matrix is:

Among them, the subscript ps is the power corresponding to the photovoltaic field node, and The dimension of is consistent with the number of photovoltaic power stations in the distribution network.

In addition, the estimation algorithm is to iteratively calculate the equivalent model of the photovoltaic power station and the injection power measurement equation in step (1) according to the least squares method.

The state variable of the least squares state estimation needs to be obtained indirectly by means of the measurement equation, that is, the functional relationship between the state vector and the measurement vector. Considering the measurement noise equation, the relationship between them is:

z=h(x)+v;

Among them, z is the measurement vector of m dimension; h(x) is the measurement function vector:

h ^T (x)=[h ₁ (x),h ₂ (x),...,h _m (x)];

v is the measurement noise vector, and the expression of v is:

v ^T =[v ₁ ,v ₂ ,..., _vm ];

The objective function of the least squares state estimation is:

J(x) = [zh(x)] ^T R ^-1 [zh(x)];

Among them, R ^-1 =1/σ ² ; for a given measurement vector z, the state estimation vector x is the value that minimizes the objective function J(x). To obtain the extreme value, it is necessary to differentiate J(x)=[zh(x)] ^T R ^-1 [zh(x)], we get:

in, is the Jacobian matrix.

The state variable selects the branch current, that is, the amplitude and phase, the inverter amplitude modulation ratio and the phase modulation angle. This state estimation algorithm selects I, δ, M, and A as state variables, that is, x=(I, δ, M, A), thereby establishing the Jacobian matrix

The iterative equation of the state variable is:

Δx ^(k+1) = x ^(k) +Δx ^(k) ;

Among them, G(x ^(k) )= ^HT (x ^(k) )R ^-1 H(x ^(k) ) is the gain matrix; x ^(k) and x ^(k+1) are the kth and The state variable obtained by k+1 iterations. When |Δx ^(k+1) | meets the convergence criteria or reaches the number of iterations, the estimation ends, and the results are uploaded to the real-time database.

In the branch power measurement in this embodiment, the power from bus k to bus m can be expressed as:

P _km +jQ _km =V _km (I _km ) ^* =V _km I _km [cos(δ _k −α _km )+jsin(δ _k −α _km )];

By separating the real and imaginary parts, the active and reactive branch measurement equations can be expressed as:

P _km =V _km I _km cos(δ _k -α _km );

Q _km =V _km I _km sin(δ _k −α _km );

For branch current amplitude measurement, the current from busbar k to busbar m can be expressed as:

I _km (measurement)=I _km ;

The injection power measurement, as shown in Figure 4, assumes that the power is injected into k busbars, and there are n busbars connected to it at the same time. Current flows from busbars 1...m into busbar k, and from busbar k into busbar m+1...n. That is, busbars 1...m are upstream busbars of busbar k, and busbars m+1...n are downstream busbars of busbar k.

By transforming, we can get:

By separating the real and imaginary parts, the equation becomes:

Among them: Re(x): the real part of x; Im(x): the imaginary part of x.

Voltage amplitude measurement, assuming that the voltage amplitude measurement is set on bus k, and there are n branches connecting bus k and bus 0, all branch currents flow out from bus 0, we can get,

Among them, x is the system state variable, P _k is the injected active power of bus k, Q _k is the injected reactive power of bus k, P _km is the injected active power from bus k to bus m, and Q _km is the injected active power from bus k to bus m. The injected reactive power of bus m, V _k is the voltage amplitude of bus k, δ _k is the voltage phase angle of bus k, is the complex voltage form of bus k, I _km is the branch current amplitude from bus k to bus m, z _km is the impedance modulus value of bus k to bus m.

During photovoltaic injection power measurement:

The invention discloses a state estimation method considering photovoltaic injection power. The method takes into account the influence of photovoltaic power supply access, and establishes an equivalent model considering photovoltaic power station and injection power measurement equation; Amplitude measurement, node injection power measurement, and voltage amplitude measurement are converted into the amplitude and phase angle of the branch current; The modulation angle is used as a state quantity, and an estimation algorithm is used to estimate the state and correct the state variable, so as to obtain an accurate state estimation result. On the basis of absorbing the transformation method based on branch voltage measurement, a distribution network state estimation algorithm based on multiple state variables such as branch current is proposed. In addition to using the power measurement in the distribution network, it also uses the current amplitude Measurement and voltage amplitude measurement, because the latter two measurements occupy a large proportion in the distribution network measurement, so the algorithm effectively improves the measurement redundancy and the observability and accuracy of state estimation.

While the invention has been shown and described with reference to specific embodiments, those skilled in the art will appreciate that forms and Various changes in details.

Claims (6)

1. A state estimation method considering photovoltaic injection power is characterized in that: the method comprises the following steps:

(1) initializing network parameters, and establishing a photovoltaic power station equivalent model and an injection power measurement equation;

(2) converting branch power measurement, branch current amplitude measurement, node injection power measurement and voltage amplitude measurement of the power distribution network into amplitude and phase angle of branch current;

(3) constructing a measurement matrix, taking the branch current I, the amplitude modulation ratio M of the inverter and the phase modulation angle A as state quantities, carrying out state estimation by adopting an estimation algorithm, and correcting a state variable;

(4) and (4) repeating the step (3) until a convergence condition is met, obtaining a state estimation result, and uploading the state estimation result to a real-time database.

2. The state estimation method taking into account photovoltaic injection power according to claim 1, characterized in that: in the step (1), the equivalent model and the injection power measurement equation of the photovoltaic power station are as follows:

wherein, P_psActive power, Q, for photovoltaic power generation_psReactive power for photovoltaic power generation; u is the amplitude of the voltage at the AC side, and theta is the phase angle of the voltage at the AC side; u shape_cellIs the voltage of the photovoltaic cell; n is a radical of_aThe number of photovoltaic cells connected in series in a single photovoltaic array; n is a radical of_aaThe number of photovoltaic modules contained in all photovoltaic arrays; m is the inverter amplitude modulation ratio in the photovoltaic power generation model, and A is the phase modulation angle in the photovoltaic power generation model; z is a radical of_ab、φ_ab、z_bc、φ_bc、z_ac、φ_acAnd converting the Y-shaped circuits of the alternating current parts of the equivalent circuits of the photovoltaic power generation models into equivalent impedance amplitude and phase angle of the delta-shaped equivalent circuits.

3. The state estimation method taking into account photovoltaic injection power according to claim 1, characterized in that: and the step (2) comprises the steps of measuring the active power of the branch, measuring the reactive power of the branch, measuring the current amplitude, measuring the node power injection and measuring the voltage amplitude, and converting the voltage amplitude into the amplitude and the phase angle of the branch current.

4. A state estimation method taking account of photovoltaic injected power according to claim 3, characterized in that: in the step (2), the virtual real part for measuring the active power of the branch, the virtual real part for measuring the reactive power of the branch, the virtual real part for measuring the node-injected power and the virtual real part for measuring the voltage amplitude are respectively represented as a branch current amplitude and a branch current phase angle; the current magnitude measurement is expressed as a branch current.

5. The state estimation method taking into account photovoltaic injection power according to claim 1, characterized in that: and (3) performing iterative computation on the photovoltaic power station equivalent model and the injection power measurement equation in the step (1) according to a least square method.

6. The state estimation method taking into account photovoltaic injection power according to claim 5, characterized in that: the estimation algorithm comprises: and (3) constructing an equation according to the state variables: x ═ I, δ, M, a), a jacobian matrix is established according to the equation x ═ I, δ, M, a, the jacobian matrix being a state estimation model that considers a plurality of state quantities, the jacobian matrix being:

wherein, the subscript ps is the power corresponding to the photovoltaic electric field node,andthe dimension of the grid is consistent with the number of the photovoltaic power stations of the power distribution network.