Parameter Optimization Method for Wind Farms Integrated with VSC-HVDC Based on Superconducting Magnetic Energy Storage Coils

Abstract

The present invention belongs to the technical field of wind power grid connection, and relates to a parameter optimization method for a wind farm incorporating a superconducting magnetic energy storage coil into a VSC-HVDC. It solves the problem of reduced efficiency of the SMES coil caused by not optimizing the initial inductance L<subgt;SMES< / subgt> and initial current I<subgt;SMES0< / subgt> of the SMES magnetic energy storage coil, as well as the four PI control parameters (K<subgt;Pi< / subgt> and K<subgt;Ii< / subgt;, i = 1, 2). The steps of the present invention are to minimize the cumulative error of DC power, minimize the cumulative deviation of DC voltage, and minimize the initial energy of the SMES coil as the objective function, so that the system exhibits the smallest DC power fluctuation situation and is equipped with different weights, and determine the optimal values of the parameters of the superconducting magnetic energy storage coil according to the particle swarm algorithm. The use of SMES-FCL can reduce the DC power fluctuation caused by intermittent wind power generation and limit the fault current of the DC line.

Description

Technical Field

[0001] The present invention belongs to the technical field of wind power grid connection, and relates to a parameter optimization method for a wind farm integrating a superconducting magnetic energy storage coil into a VSC-HVDC. Background Art

[0002] For the control strategies of the converters on the wind farm side and the grid side of a DFIG, the VSC on the wind farm side consists of constant AC voltage and constant active power, and the VSC on the grid side consists of constant reactive power and constant DC voltage control. The frequency of the offshore AC grid can be controlled by a frequency controller to keep the frequency constant by adjusting the active power of the wind farm. The AC voltage of the wind farm is controlled by the outer voltage controller loop and the inner current control loop of the wind farm VSC.

[0003] As an important renewable energy source, wind power occupies an important position in the future energy development. The grid has put forward higher and higher requirements for the controllability of wind power. Due to the intermittency and volatility of wind power, the DC power fluctuates greatly when it is integrated into the VSC-HVDC, affecting the system stability. The key problems of high-voltage DC wind farms are DC power fluctuations and DC faults. To solve these two problems simultaneously, the existing technology integrates a superconducting magnetic energy storage coil with current-limiting ability (SMES-FCL) into the HVDC-Wind farm system. The SMES-FCL can reduce the DC power fluctuations caused by intermittent wind power generation and limit the fault current of the DC line.

[0004] The SMES-FCL circuit consists of two parts, namely the SMES part and the FCL part. As Figure 1 shown, the SMES and FCL are composed of DC-to-DC converters and are connected by a superconducting magnetic energy storage coil. The superconducting magnetic energy storage coil part is designed to alleviate the power fluctuations generated by the wind farm. Figure 2 is the controller of the superconducting magnetic energy storage coil. By controlling four PI control parameters (K Pi and K Ii , i = 1, 2), the opening and closing of switches S1 and S2 in the SMES part in Figure 1 are controlled, and the pulse width modulation PWM wave is modulated to make the superconducting magnetic energy storage coil switch between the charging and discharging modes.

[0005] In the prior art, the initial inductance L SMES and the initial current I SMES0 of the superconducting magnetic energy storage coil, as well as the four PI control parameters (K Pi and K Ii , i = 1, 2) are not optimized. If the parameters are not optimized, the initial energy storage of the coil cannot be determined, and the four PI control parameters (K Pi and K Ii, where \(i = 1, 2\)) cannot be accurately determined as required, resulting in the superconducting magnetic energy storage coil not being able to maximize its advantages, causing frequent fluctuations in the DC voltage and being unable to effectively suppress the fault current. Summary of the Invention

[0006] The present invention provides a parameter optimization method for integrating a wind farm with a superconducting magnetic energy storage coil into a VSC-HVDC, effectively solving the problem of reduced efficiency of the superconducting magnetic energy storage coil caused by the lack of optimization of the initial inductance \(L\) SMES and the initial current \(I\) SMES0 as well as the four PI control parameters (\(K\) Pi and \(K\) Ii , where \(i = 1, 2\)).

[0007] To solve the above problems, the technical solution adopted by the present invention is as follows:

[0008] A parameter optimization method for integrating a wind farm with a superconducting magnetic energy storage coil into a VSC-HVDC, where the superconducting magnetic energy storage coil includes: an initial inductance \(L\) SMES and an initial current \(I\) SMES0 as well as the four PI control parameters (\(K\) Pi and \(K\) Ii , where \(i = 1, 2\)). It is characterized in that: the initial inductance \(L\) SMES and the initial current \(I\) SMES0 as well as the four PI control parameters (\(K\) Pi and \(K\) Ii , where \(i = 1, 2\)) are optimized using the particle swarm optimization algorithm, and the optimization steps are as follows:

[0009] S1. Based on minimizing the cumulative error of DC power, minimizing the cumulative deviation of DC voltage, and minimizing the initial energy of the superconducting magnetic energy storage coil as the objective function, the system is made to exhibit the smallest DC power fluctuation situation;

[0010] S2. Different weights are assigned to minimizing the cumulative error of DC power, minimizing the cumulative deviation of DC voltage, and minimizing the initial energy of the superconducting magnetic energy storage coil; the optimization expression is:

[0011] \(\omega_1\cdot LAE\) P +\(\omega_2\cdot LAE\) V +\(\omega_3E\) SMES (1)

[0012] where: \(LAE\) p is to minimize the cumulative error of DC power, \(LAE\) V is to minimize the cumulative deviation of DC voltage, \(E\) SMES is to minimize the initial energy of the superconducting magnetic energy storage coil, and \(\omega_1\), \(\omega_2\), and \(\omega_3\) are the weights of the first, second, and third terms respectively;

[0013] S3. With the minimization of the initial energy of the superconducting magnetic energy storage coil as the objective function and the maximization of the suppression of the DC fault current as the purpose, different weights are assigned according to the three objective functions, and the optimal values of the parameters of the superconducting magnetic energy storage coil are determined according to the particle swarm algorithm.

[0014] The minimization of the cumulative error of the DC power LAE p The expression is:

[0015]

[0016] Among them, t0 and t sim are the initial simulation time and the final simulation time respectively, and ΔP DC is the DC power P DC_ref minus the actual DC power P DC .

[0017] The minimization of the cumulative deviation of the DC voltage LAE v The expression is:

[0018]

[0019] Among them, t0 and t sim are the initial simulation time and the final simulation time respectively, and ΔV DC is the DC voltage V DC_ref minus the actual DC power V DC .

[0020] The minimization of the initial energy of the superconducting magnetic energy storage coil E SMES The expression is:

[0021]

[0022] Among them, L SMES is the inductance of the superconducting magnetic energy storage coil, and I SMES0 is the instantaneous coil current.

[0023] The steps of the particle swarm algorithm are as follows:

[0024] S301. According to the initialization process, the random positions and velocities of the particle swarm are initially set;

[0025] S302. Calculate the fitness value of each particle;

[0026] S303. Compare the fitness value of each particle with the fitness value of the best position it has experienced to determine the current best position;

[0027] S304. Compare the fitness value of each particle with the fitness value of the best position globally experienced to determine the current global best position;

[0028] S305. Evolve the velocity and position of the particles according to equations (5) and (6).

[0029] v id (t + 1)= v id (t)+ c1 * r1 * (P id (t)- x id (t))+ c2 * r2 * (P gd (t)- x id (t)) (5)

[0030] x id (t + 1)= x id (t)+ v id (t + 1) 1 ≤ i ≤ n 1 ≤ d ≤ D (6)

[0031] Wherein, c1 and c2 are positive constants, called acceleration factors. c1 is the step size for adjusting the direction of the particle flying towards its own best position; c2 is the step size for adjusting the particle flying towards the global best position; r1 and r2 are random numbers between (0, 1).

[0032] S306. If the fitness value or the preset maximum number of generations is not reached, return to step S302.

[0033] The beneficial effects of the present invention are as follows: The key problems of a high - voltage DC wind farm are DC power fluctuations and DC faults. In order to solve these two problems simultaneously, a superconducting magnetic energy storage coil with current - limiting ability (SMES - FCL) is integrated into the HVDC - Wind farm system. Using SMES - FCL can reduce the DC power fluctuations caused by intermittent wind power generation and limit the fault current of the DC line. Brief Description of the Drawings

[0034] Figure 1 is the circuit diagram of SMES - FCL in the background technology;

[0035] Figure 2 is the circuit diagram of the superconducting magnetic energy storage coil controller in the background technology;

[0036] Figure 3 is the optimization flow chart of the present invention;

[0037] Figure 4 is the flow chart of the particle swarm algorithm. Detailed Embodiments

[0038] The technical solution of the present invention will be further described below with reference to the drawings and through specific embodiments:

[0039] Embodiment 1

[0040] The SMES and FCL are composed of a DC-to-DC converter and connected by a superconducting magnetic energy storage coil. The superconducting magnetic energy storage coil part is designed to mitigate the power fluctuations generated by the wind farm; when a short-circuit fault occurs in the DC line, the FCL operates to suppress the fault current. The initial inductance L SMES and the initial current I SMES0 are determined by the subsequent optimization method. Figure 2 is the controller of the superconducting magnetic energy storage coil. By controlling the four PI control parameters (K Pi and K Ii , i = 1, 2), it controls the opening and closing of switches S1 and S2 in the superconducting magnetic energy storage coil part in Figure 1 , modulates the pulse-width modulation PWM wave, and enables the superconducting magnetic energy storage coil to switch between the charging and discharging modes.

[0041] The operating principle of the superconducting magnetic energy storage coil can be described by the power flow equation:

[0042] P DC = P WVSC + P SMES (11)

[0043] where P DC , P WVSC and P SMES are the active powers in the DC line, the VSC on the wind farm side, and the superconducting magnetic energy storage coil respectively.

[0044] P SMES = V WVSC I SMES (12)

[0045] where V WVSC and I SMES are the VSC voltage on the wind farm side and the superconducting magnetic energy storage coil current respectively. By controlling I SMES , electric energy can be injected into or flow out of the superconducting magnetic energy storage coil. Therefore, the flow direction of P SMES can be controlled to reduce the DC power fluctuations.

[0046] In the FCL circuit, switches SW1 and SW2 are closed, and switches SW3, SW4, DW3, and DW4 are open, enabling the current to flow from the wind farm side converter to the DC line. For the operation of the FCL circuit, the impedance of the superconducting magnetic energy storage coil on the DC side during a fault:

[0047]

[0048] where V SMES is the superconducting magnetic energy storage coil voltage. The instantaneous coil voltage v SMES is given by the following formula:

[0049]

[0050] where i SMES is the instantaneous coil current, and l SMES is the inductance of the superconducting magnetic energy storage coil. When a DC fault occurs, i SMES increases immediately. This results in a rapid change rate of i SMES and v SMES . Therefore, the impedance of the SMES will be very high, which can reduce the fault current.

[0051] Subtract the actual DC power P DC_ref from the reference DC power P DC as the input signal of the proportional-integral PI-A controller to obtain the output signal of PI-A. Add the signal of PI-A to the difference between the actual DC I DC and the reference DC I DC_ref to obtain the input signal of PI-B. Therefore, the output of ΔD changes periodically. Then add it to the duty cycle of 0.5 to control the pulse width modulation PWM of switches S1 and S2. When D is greater than 0.5, it is in the charging mode, and when D is less than 0.5, it is in the discharging mode.

[0052] Use the following method to optimize the target parameters to minimize the initial stored energy of the initial inductance and initial coil current of the superconducting magnetic energy storage coil. The SMES-FCL can suppress the DC fault current maximally and keep the system running stably.

[0053] The optimization method is as follows:

[0054] Step 1: Based on minimizing the cumulative error of DC power, minimizing the cumulative deviation of DC voltage, and minimizing the initial energy of the superconducting magnetic energy storage coil as the objective function, make the system exhibit the smallest DC power fluctuation situation;

[0055] Among them, the minimization of the cumulative error of DC power LAE p The expression is:

[0056] [[ID=4,2]]

[0057] where t0 and t sim are the initial simulation time and the final simulation time respectively, and ΔP DC is the DC power P DC_ref minus the actual DC power P DC .

[0058] The minimization of the cumulative deviation of DC voltage LAE v The expression is:

[0059]

[0060] t0 and t sim are the initial simulation time and the final simulation time respectively, and ΔV DC is the DC voltage V DC_ref minus the actual DC power V DC .

[0061] Minimize the initial energy E of the superconducting magnetic energy storage coil SMES The expression is:

[0062]

[0063] Among them, L SMES is the inductance of the superconducting magnetic energy storage coil, and I SMES0 is the instantaneous coil current.

[0064] Step 2: Assign different weights to the minimization of the cumulative error of DC power, the minimization of the cumulative deviation of DC voltage, and the minimization of the initial energy of the superconducting magnetic energy storage coil; The optimization expression is:

[0065] ω1·LAE P +ω2·LAE V +ω3E SMES (1)

[0066] Among them: LAE p is the minimization of the cumulative error of DC power, LAE v is the minimization of the cumulative deviation of DC voltage, E SMES is the minimization of the initial energy of the superconducting magnetic energy storage coil, and ω1, ω2, and ω3 are the weights of the first, second, and third terms respectively. Since the main function of SMES-FCL is to limit the fault current, ω2 is set to 1.0, ω1 = 0.4, and ω3 = 0.004. Constraint (a) is Figure 2 the range of the gains K Pi and K Ii of the PI-A and PI-B controllers in.

[0067] (a) 0.1 ≤ K Pi ≤ 30, 0.1 ≤ K Ii ≤ 30, i = 1, 2.

[0068] (b) 0.01 ≤ L SMES ≤ 10H

[0069] (c) 0.01 ≤ I SMES0 ≤ 10KV

[0070] S3. With the minimization of the initial energy of the superconducting magnetic energy storage coil as the objective function and the maximization of suppressing the DC fault current as the purpose, different weights are assigned to the three objective functions, and the optimal values of the parameters of the superconducting magnetic energy storage coil are determined according to the particle swarm algorithm.

[0071] Particle swarm calculation method for the optimal parameters of the superconducting magnetic energy storage coil:

[0072] 1. Based on the minimization of the cumulative DC power error, the minimization of the cumulative DC voltage deviation, and the minimization of the initial energy of the superconducting magnetic energy storage coil as the objective functions, the system exhibits the smallest DC power fluctuation scenario.

[0073] 2. The main function of SMES - FCL is to limit the fault current. Therefore, set ω1 = 0.4, ω2 = 1, ω3 = 0.004 to maximize the suppression of the DC fault current and keep the system running stably.

[0074] 3. With the minimization of the initial energy of the superconducting magnetic energy storage coil as the objective function and the maximization of suppressing the DC fault current as the purpose, different weights are assigned to the three objective functions, and the optimal values of each parameter of the superconducting magnetic energy storage coil are found according to the steps of the particle swarm algorithm.

[0075] Steps of the particle swarm algorithm:

[0076] Assume the search space is D - dimensional, the total number of particles is n, and the position of the i - th particle is represented as the vector x i =(x i1 ,x i2 ....x iD ), the best position searched by the i - th particle so far is pbest i =(P i1 ,P i2 ....P iD ), the best position searched by the entire particle swarm so far is gbest=(g1,g2....g D ), and the position change rate (velocity) of the i - th particle is the vector v i =(v i1 ,v i2 ....v iD ). The velocity and position of each dimension of the particle change according to the following formulas:

[0077] v id (t + 1)=v id (t)+c1*r1*(P id (t)-x id (t))+c2*r2*(P gd (t)-x id (t)) (9)

[0078] x id (t + 1) = x id (t) + v id (t + 1) 1 ≤ i ≤ n 1 ≤ d ≤ D (10)

[0079] Among them, c1 and c2 are normal constants, called acceleration factors. c1 is the step size for adjusting the direction of the particle flying towards its own best position; c2 adjusts the step size for the particle flying towards the global best position; r1 and r2 are random numbers between (0, 1). The initial positions and velocities of the particle swarm are randomly generated and then iterated according to the formula until a satisfactory solution is found.

[0080] The process of the basic particle swarm optimization algorithm is as follows:

[0081] 1. According to the initialization process, initialize the random positions and velocities of the particle swarm.

[0082] 2. Calculate the fitness value of each particle.

[0083] 3. Compare the fitness value of each particle with the fitness value of the best position pbest i it has experienced. If it is better, then use it as the current best position.

[0084] 4. Compare the fitness value of each particle with the fitness value of the global best position gbest i it has experienced. If it is better, then use it as the current global best position.

[0085] 5. Evolve the velocities and positions of the particles according to Equation (9) and Equation (10).

[0086] 6. If the end condition is not reached (usually a good enough fitness value or reaching a preset maximum number of generations), then return to Step 2.

[0087] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non - restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be encompassed within the present invention. Any reference signs in the claims should not be regarded as limiting the claimed rights.

[0088] In addition, it should be understood that although this specification is described according to embodiments, not every embodiment only contains an independent technical solution. This narrative way of the specification is only for clarity. Those skilled in the art should regard the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A parameter optimization method for a wind farm integrated with VSC-HVDC based on a superconducting magnetic energy storage coil, the superconducting magnetic energy storage coil comprising: Initial inductance L SMES K Pi and initial current I SMES0 as well as four PI control parameters (and K Ii , i = 1, 2), characterized in that: the initial inductance L SMES and the initial current I SMES0 as well as four PI control parameters (K Pi and K Ii , i = 1, 2) are optimized by a particle swarm algorithm, and the optimization steps are as follows: S1. Based on the minimization of the cumulative error of DC power, the minimization of the cumulative deviation of DC voltage, and the minimization of the initial energy of the superconducting magnetic energy storage coil as the objective function, the system exhibits the smallest DC power fluctuation situation; S2. Different weights are assigned to the minimization of the cumulative error of DC power, the minimization of the cumulative deviation of DC voltage, and the minimization of the initial energy of the superconducting magnetic energy storage coil; S3. Taking the minimization of the initial energy of the superconducting magnetic energy storage coil as the objective function and aiming at maximizing the suppression of DC fault current, different weights are assigned according to the three objective functions, and the optimal values of the parameters of the superconducting magnetic energy storage coil are determined according to the particle swarm algorithm; The DC power cumulative error minimization LAE p The expression is as follows: Among them, t0 and t sim are the initial simulation time and the final simulation time respectively, and ΔP DC is the DC power P DC_ref minus the actual DC power P DC ; The DC voltage cumulative deviation minimization LAE V The expression is: Among them, t0 and t sim are the initial simulation time and the final simulation time respectively, and ΔV DC is the DC voltage V DC_ref minus the actual DC power V DC ; The initial energy minimization E of the superconducting magnetic energy storage coil SMES The expression is as follows: Among them, L SMES is the inductance of the superconducting magnetic energy storage coil, and I SMES0 is the instantaneous coil current.

2. The parameter optimization method for a wind farm integrated into VSC-HVDC based on a superconducting magnetic energy storage coil according to claim 1, characterized in that: The weight assignment expression is: ω1·LAE P +ω2·LAE V +ω3E SMES (1) Among them: LAE p For minimizing the cumulative error of DC power, LAE V For minimizing the cumulative deviation of DC voltage, E SMES For minimizing the initial energy of the superconducting magnetic energy storage coil, ω1, ω2, and ω3 are the weights of the first, second, and third terms respectively.

3. The parameter optimization method for a wind farm integrated into VSC-HVDC based on a superconducting magnetic energy storage coil according to claim 1, characterized in that: The steps of the particle swarm algorithm are as follows: S301. According to the initialization process, the random positions and velocities of the particle swarm are initially set; S302. Calculate the fitness value of each particle; S303. For each particle, compare its fitness value with the fitness value of the best position it has experienced to determine the current best position; S304. For each particle, compare its fitness value with the fitness value of the best position globally experienced to determine the current global best position; S305. Evolve the velocity and position of the particle according to equations (5) and (6), v id (t + 1) = v id (t) + c1 * r1 * (P id (t) - x id (t)) + c2 * r2 * (P gd (t) - x id (t)) (5) x id (t + 1) = x id (t) + v id (t + 1) 1 ≤ i ≤ n 1 ≤ d ≤ D (6) where c1 and c2 are positive constants, called acceleration factors. c1 is the step size for adjusting the direction of the particle flying towards its own best position; c2 is the step size for adjusting the particle flying towards the global best position; r1 and r2 are random numbers between (0, 1); S306. If the fitness value or the preset maximum number of generations is not reached, return to step S302.

Images