Wave power generation device approximate optimal reference speed generation method based on simulated annealing particle swarm optimization
Through the simulation of annealed particle swarm algorithm, the damping coefficient and reference speed of the wave power generation device are optimized, and the efficiency and stability problems of the wave energy power generation system in complex environments are solved, and the efficient approximate optimal reference speed generation is achieved, which improves the system's adaptability.
Patent Information
- Application Number
- CN202510508756.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-22
- Publication Date
- 2025-08-05
- Estimated Expiration
- 2045-04-22
AI Technical Summary
In the prior art, the optimal damping coefficient and reference speed generation strategy of wave energy power generation systems are difficult to optimize efficiently in complex wave environments, the convergence speed of simulated annealing algorithm is slow, and the particle swarm algorithm is prone to convergence early, resulting in limited system efficiency and stability.
Using simulated annealed particle swarm algorithm, by establishing a frequency domain model of point-absorbing direct drive wave power generation device, constructing an equivalent circuit model, optimizing the motor output power, and updating the particle velocity and position using Metropolis criterion and learning factors to generate an approximate optimal reference velocity.
It improves the overall performance of the algorithm, enhances the device's adaptability to complex sea conditions, overcomes the shortcomings of premature convergence and slow convergence speed, and improves the efficiency and stability of the wave energy power generation system.
Smart Images

Figure CN120430164A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of renewable energy power generation, and in particular relates to a method for generating an approximate optimal reference speed of a wave power generation device based on a simulated annealing particle swarm algorithm. Background Art
[0002] Wave power generation, a promising clean energy technology within the vast field of renewable energy generation, has recently attracted widespread attention from academia and industry worldwide. However, the efficiency and stability of wave power systems rely heavily on optimizing their control system parameters, particularly the strategy for generating the optimal damping coefficient and reference velocity. The complex, variable, and nonlinear nature of the wave environment makes this problem challenging to solve. Simulated annealing and particle swarm optimization (PSO) are two commonly used intelligent optimization algorithms, each applied independently to parameter optimization of wave power systems. Simulated annealing avoids falling into local optima by probabilistically accepting less favorable solutions, but its convergence rate is relatively slow, particularly when dealing with complex, high-dimensional problems. On the other hand, the PSO, through information sharing and collaboration among particles, seeks the global optimal solution, offering the advantages of fast convergence and ease of implementation. However, the PSO is prone to premature convergence—prematurely locking onto a local optimal solution and hindering further exploration of the solution space. This limits its application to complex optimization problems. Summary of the Invention
[0003] Purpose of the invention: The technical problem to be solved by the present invention is to address the deficiencies of the existing technology and provide a method for generating an approximate optimal reference speed for a wave power generation device based on a simulated annealing particle swarm algorithm, comprising the following steps:
[0004] Step 1: Based on the point absorption direct drive wave power generation device, a system frequency domain model is established according to the linear motion equation;
[0005] Step 2: Consider the force component in the system frequency domain model as the electric component in the circuit, and establish an equivalent circuit model through the force-electricity analogy method;
[0006] Step 3: Use the working condition optimization algorithm to construct the operation optimization objective function of the motor output power;
[0007] Step 4: According to the equivalent circuit model, configure the initial parameters of the simulated annealing particle swarm and randomly generate the initial particle population. The initial parameters include random control parameters r1 and r2, population size D, and maximum iteration step size dB. rmax , maximum number of iterations K, initial annealing temperature T0, minimum annealing temperature T end , annealing coefficient α, and learning factors c1 and c2; r1 and r2 are random numbers between 0 and 1;
[0008] Step 5: Determine whether to accept the current solution based on the Monte Carlo (Metropolis) criterion;
[0009] Step 6: Update the speed and position of the particles in the population according to the global optimal alternative solution, and update the individual output power of the particles in the population and the global optimal output power;
[0010] Step 7: When the annealing temperature is lower than the termination temperature and the maximum number of iterations is reached, the optimal output power of the objective function and the corresponding approximate optimal damping coefficient are output;
[0011] Step 8: Obtain the approximate optimal reference speed according to the reference speed expression and the approximate optimal damping coefficient.
[0012] In step 1, the point absorption direct-drive wave power generation device includes a buoy and a permanent magnet synchronous generator, wherein the permanent magnet synchronous generator includes a coil and a permanent magnet; the permanent magnet synchronous generator is used as a power output shaft, and the buoy and the power output shaft are coupled in a vertical direction to form a motion device;
[0013] Considering only the vertical force, the permanent magnet synchronous generator reciprocates under the drive of the buoy. The coil cuts the magnetic lines of force, generating an induced current. The induced current is regulated to adjust the amplitude of the generator force. Based on Newton's second law and Cummins equation, the linear motion equation is:
[0014]
[0015] Among them, M and m ∞ are the mass of the moving device and the additional mass at infinite frequency, K r (t) is the kernel function convolved with the velocity at time t, K h is the sum of the hydrostatic stiffness coefficient and the spring stiffness coefficient, F e (t) is the excitation force generated by the incident wave at time t, F g (t) is the back electromagnetic force of the permanent magnet synchronous generator at time t, x(t) and v(t) are the vertical displacement of the moving device at time t and the velocity of the moving device at time t, respectively.
[0016] In step 1, by performing frequency domain transformation on formula (1), the following system frequency domain model is obtained:
[0017]
[0018] Where j is an imaginary unit, ω is the angular frequency, V(ω) is the velocity of the converter and float at the angular frequency ω, and F e (ω) is the excitation force generated by the incident wave at angular frequency ω, F g (ω) is the counter-electromagnetic force of the generator at angular frequency ω, Kr (ω) is the kernel function convolved with velocity at angular frequency ω, K r (ω)=B r (ω)+jωA r (ω), A r (ω) and B r (ω) are the so-called radiative added mass and radiation damping at the angular frequency ω, respectively.
[0019] In step 2, the velocity V(ω) in formula (2) is used as the current component in the circuit; the excitation force is used as the voltage source; the counter-electromagnetic force is used as the impedance, which can be adjusted as the output load; the coefficient K r (ω),ω(M+m ∞ )and As the inherent resistive and inductive loads in the circuit, the equivalent circuit model is obtained; according to the basic Ohm's law of the circuit, the expression of the speed V(ω) is obtained as follows:
[0020]
[0021] Among them, Z w (ω) is the system inherent equivalent impedance of the power output shaft at the angular frequency ω, When the impedance of the counter-electromagnetic force is adjusted to the resonance of the system, the generator output power is maximum and the speed is in phase with the exciting force.
[0022] In step 3, the objective function of the operation optimization problem of the motor output power is the average output power model derived from the point absorption direct-drive wave power generation device model, which is expressed as:
[0023]
[0024] in is the average output power of the point absorption direct-drive wave power generation device at the angular frequency ω, Re[] means the real part of the formula in the square brackets, V * (ω) is the conjugate of the power take-off shaft system speed at angular frequency ω;
[0025] The operating condition optimization algorithm makes the speed and the excitation force in phase, so that the equivalent circuit presents a resistance characteristic. This requires that the control force applied by the power output shaft conforms to the following equation:
[0026]
[0027] in is Z at angular frequency ω w The conjugate of (ω), Z g (ω) is the output impedance of the motor at the angular frequency ω;
[0028] With the goal of maximizing output power, an operation optimization objective function is established to satisfy:
[0029] max[f(x)]=max[P(B r )] (6)
[0030] in For the equivalent circuit with a damping coefficient of B r The maximum output power expression when B r is the damping coefficient in the equivalent circuit; R g is the equivalent output resistance, is the conjugate of the damping coefficient in the equivalent circuit.
[0031] In step 5, the constraints for accepting the current solution are:
[0032] P(B r(k+1) )>P(B r(k) ) (7)
[0033] Among them B r(k+1) is the damping coefficient selected at the (k+1)th iteration; if the constraints are met, the current solution is accepted; otherwise, the current solution is accepted with a specific probability. The specific probability expression is:
[0034]
[0035] Where T k is the annealing temperature after k annealing operations, and exp refers to the natural exponential function.
[0036] In step 6, the speed and position of the particles in the population are updated according to the global optimal alternative solution, satisfying:
[0037] dB rpq (k+1)=dB rpq (k)+c1r1(k)[P pq (k)-B rpq (k)]+c2r2(k)[P gq (k)-B rpq (k)] (9)
[0038] B rpq (k+1)=B rpq (k)+dB rpq (k+1) (10)
[0039] Where p, q = 1, 2, ..., D; D is the selected particle dimension; dB rpq (k) is the iteration step size of each particle at the kth iteration, dB rpq (k)∈[-dB rmax , dB rmax ], Brpq (k) represents the approximate maximum damping coefficient of each particle at the kth iteration, dB rmax Indicates dB rpq (k) Upper limit of value, P pq (k) is the time when B rpq Substitute the value of into the formula The obtained individual optimal position of the particle itself is the component value in the qth dimension, P gq (k) is the time when B rpq Substitute the value of into the formula The component value of the global optimal position of the group in the qth dimension is obtained, Pointer B rpq The conjugate of , c1 and c2 are two learning factors with different values, r1(k) and r2(k) are two uniform random numbers in the range of [0,1] with different values at the kth iteration; the individual optimal output power and the global optimal output power of each particle are determined again according to the global optimal position update result of the particles in the population.
[0040] In step 7, it is determined whether the annealing temperature satisfies the following constraint:
[0041] T k <T end (11)
[0042] Among them, T end is the annealing termination temperature;
[0043] If the constraint (11) is not satisfied, the temperature annealing operation is performed and the updated temperature rule expression is:
[0044] T k+1 =α·T k (12)
[0045] Where α is the annealing coefficient;
[0046] If the constraint (11) is satisfied, then determine whether the number of iterations satisfies the following constraint:
[0047] k <K (13)
[0048] Where k is the number of annealing operations performed, and K is the maximum number of annealing operations set. If constraint (13) is satisfied, the probability calculation is repeated using formula (8) until the maximum number of iterations is reached, and the optimal output power of the objective function and the corresponding approximate optimal damping are output.
[0049] In step 8, when the equivalent circuit exhibits resistance characteristics, the reference speed expression of the power output shaft system is:
[0050]
[0051] Among them, V ref (ω) is the reference speed at the angular frequency ω, and the corresponding reference speed at this time is the approximate optimal reference speed.
[0052] The present invention also provides an electronic device, comprising a processor and a memory, wherein the memory stores program code, and when the program code is executed by the processor, the processor executes the steps of the method.
[0053] The present invention converts the damping coefficient in a point-absorption direct-drive wave power generation device into a group of particles to be optimized. Taking the device output power as the optimization objective function, each particle undergoes a simulated annealing process in parallel. During the annealing process, the Metropolis criterion is used to selectively accept the new state of each particle, and its jumping characteristics are used to jump out of the local optimum, finally converging to the global optimal solution.
[0054] The present invention has the following beneficial effects: the method of the present invention adopts the simulated annealing particle swarm algorithm to realize the generation of the optimal reference speed. Compared with the standard particle swarm algorithm or the simulated annealing algorithm, the method of the present invention overcomes the disadvantage of premature convergence of the particle swarm algorithm, and improves the disadvantage of slow convergence speed of the simulated annealing algorithm, thereby improving the overall performance of the algorithm and enhancing the adaptability of the device to complex sea conditions. BRIEF DESCRIPTION OF THE DRAWINGS
[0055] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, and the above and / or other advantages of the present invention will become more apparent.
[0056] Figure 1 This is a flow chart of the simulated annealing algorithm of the present invention.
[0057] Figure 2 Schematic diagram of damping coefficient optimization in an embodiment of the present invention.
[0058] Figure 3 is the output power during the parameter adjustment process in the embodiment of the present invention.
[0059] Figure 4 This is a comparison chart of control performance under inconsistent parameters in an embodiment of the present invention.
[0060] Figure 5 This is the output power objective function convergence curve when the method is used in the embodiment of the present invention.
[0061] Figure 6 This is a comparison chart of the generator output energy obtained by using this method in an embodiment of the present invention compared with the case of fixed parameters. DETAILED DESCRIPTION
[0062] In this embodiment, a point absorption direct-drive wave power generation device is first provided, including a buoy and a permanent magnet synchronous generator, the permanent magnet synchronous generator including a coil and a permanent magnet; the permanent magnet synchronous generator is used as a power output shaft, and the buoy and the power output shaft are coupled in the vertical direction to form a motion device; only considering the vertical force, the permanent magnet synchronous generator reciprocates under the drive of the buoy, the coil cuts the magnetic lines of force, generates induced current, and the induced current is regulated to adjust the amplitude of the generator force.
[0063] On this basis, the embodiment of the present invention provides a method for generating an approximate optimal reference speed of a wave power generation device based on a simulated annealing particle swarm algorithm, such as Figure 1 As shown, the method comprises the following steps:
[0064] Step 1: Based on the point absorption direct-drive wave power generation device, a system frequency domain model is established according to its linear motion equation;
[0065] Step 2: Consider the force component in the system model as the electric component in the circuit and establish an equivalent circuit model through the force-electricity analogy method;
[0066] Step 3: Use the working condition optimization algorithm to construct the operation optimization objective function of the motor output power;
[0067] Step 4: According to the equivalent circuit model, configure the initial parameters of the simulated annealing particle swarm and randomly generate the initial particle population, including random control parameters r1 and r2, population size D, and maximum iteration step size dB. rmax , maximum number of iterations K, initial annealing temperature T0, minimum annealing temperature T end , annealing coefficient α and learning factors c1 and c2; Among them, the initial parameters set in this embodiment are population size D = 600, maximum number of iterations K = 100, initial annealing temperature T0 = 1000, minimum annealing temperature T end =0.01, annealing coefficient α=0.6, learning factors c1=2.05, c2=2.05, and the initial population is randomly generated.
[0068] Step 5: Determine whether to accept the current solution based on the Monte Carlo (Metropolis) criterion;
[0069] Step 6: updating the speed and position of the particles in the population according to the global optimal alternative solution, and updating the individual output power and the global optimal output power of the particles in the population;
[0070] Step 7: When the annealing temperature is lower than the termination temperature and the maximum number of iterations is reached, the optimal output power of the objective function and the corresponding approximate optimal damping coefficient are output.
[0071] Step 8: Obtain the approximate optimal reference speed according to the reference speed expression and the above approximate optimal damping coefficient.
[0072] Specifically, in step 1, the linear motion relationship is:
[0073]
[0074] Among them, M and m ∞ They are the mass of the motion device and the additional mass at infinite frequency, which are M = 773254 kg, m ∞ =400073kg, K r (t) is the kernel function convolved with the velocity at time t, F e (t) is the excitation force generated by the incident wave at time t, F g (t) is the back electromagnetic force of the permanent magnet synchronous generator at time t, x(t) and v(t) are the vertical displacement of the moving device at time t and the speed of the moving device at time t, respectively, K h is the sum of the hydrostatic stiffness coefficient and the spring stiffness coefficient, K h Satisfies the expression:
[0075] K h =ρ e Sg (2)
[0076] Wherein, the density of seawater ρ e =1025kg / m 3 , cross-sectional area S = 16πm 2 , the acceleration due to gravity g is 9.81m / s 2 .
[0077] By performing frequency domain transformation on equation (1), the frequency domain expression is obtained as follows:
[0078]
[0079] Where j is the imaginary unit, ω is the angular frequency, V(ω) is the velocity of the converter and the buoy at the angular frequency ω, and F e (ω) is the excitation force generated by the incident wave at angular frequency ω, F g (ω) is the counter-electromagnetic force of the generator at angular frequency ω, K r (ω) is the kernel function convolved with velocity at angular frequency ω, K r (ω)=B r (ω)+jωA r (ω), A r (ω) and B r (ω) are the so-called radiation added mass and radiation damping at different frequencies at the angular frequency ω, respectively.
[0080] In step 2, the velocity component in formula (3) is regarded as the current component in the circuit; the excitation force is regarded as a voltage source; the counter-electromagnetic force is regarded as an output load with adjustable impedance; the coefficient K is r (ω),ω(M+m ∞ )and As the inherent resistive and inductive loads in the circuit, the equivalent circuit model is obtained. According to the basic Ohm's law of the circuit, the speed expression can be obtained as:
[0081]
[0082] Where Z w (ω) is the system inherent equivalent impedance of the power output shaft at the angular frequency ω, When the impedance of the counter-electromagnetic force is adjusted to the resonance of the system, the generator output power is maximum and the speed is in phase with the exciting force.
[0083] In step 3, the objective function of the operation optimization problem of the motor output power is the average output power model derived from the point absorption direct-drive wave power generation device model:
[0084]
[0085] in, is the average output power of the point absorption direct-drive wave power generation device at the angular frequency ω, Re[] means the real part of the formula in the square brackets, V * is the conjugate of the PTO system speed.
[0086] This method is inspired by photovoltaic energy capture strategies. The operating point of a photovoltaic module's characteristics is primarily determined by the impedance matching of the module with respect to the connected load. The output characteristics of the DDWEC system are also related to the system impedance and the load impedance. The algorithm ensures that the speed and excitation force are in phase, resulting in a resistive equivalent circuit. This requires that the control force applied by the power output shaft conform to the following equation:
[0087]
[0088] in, is Z at angular frequency ω w The conjugate of Z g (ω) is the output impedance of the motor at the angular frequency ω.
[0089] With the goal of maximizing output power, an operation optimization objective function is established to satisfy:
[0090] max[f(x)]=max[P(B r )] (7)
[0091] in, For the equivalent circuit with a damping coefficient of B r The maximum output power expression when B r is the damping coefficient in the equivalent circuit.
[0092] Among them, R g is the equivalent output resistance, is the conjugate of the damping coefficient in the equivalent circuit. The proposed method makes R g =B r , where B r It does not represent the true value of the system damping coefficient. g The actual damping coefficient When the two are equal, the system output power reaches the optimum. It is still an unknown parameter in the system, so it is necessary to find its true value.
[0093] In step 5, the simulated annealing algorithm uses a thermodynamic system to gradually cool the system to the lowest energy state to present the optimization process. By borrowing the mechanism of the simulated annealing algorithm, in this embodiment, the expression of the constraint condition accepting the current solution is:
[0094] P(B r(k+1) )>P(B r(k) ) (8)
[0095] Among them B r(k+1) is the damping coefficient selected at the (k+1)th iteration. If the constraints are met, the current solution is accepted.
[0096] If the constraints are not met, the Metropolis criterion is used to determine whether to accept the current solution. Metropolis is an effective key sampling method. Its algorithm is: when the system changes from one energy state to another, the corresponding energy changes from P(B r(k) ) changes to P(B r(k+1) ), its probability expression is:
[0097]
[0098] Where T k is the annealing temperature after k annealing operations, and exp refers to the natural exponential function.
[0099] In step 6, the speed and position of the particles in the population are updated according to the global optimal alternative solution to satisfy:
[0100] dB rpq (l+1)=dB rpq (k)+c1r1(k)[P pq (k)-B rpq(k)]+c2r2(k)[P gq (k)-B rpq (k)] (10)
[0101] B rpq (k+1)=B rpq (k)+dB rpq (k+1) (11)
[0102] Where p, q = 1, 2, ..., D; D is the selected particle dimension; dB rpq (k) is the iteration step size of each particle at the kth iteration, dB rpq (k)∈[-dB rmax , dB rmax ], B rpq (k) represents the approximate maximum damping coefficient of each particle at the kth iteration, dB rmax Indicates dB rpq (k) Upper limit of value, P pq (k) is the time when B rpq Substitute the value of into the formula The obtained individual optimal position of the particle itself is the component value in the qth dimension, P gq (k) is the time when B rpq Substitute the value of into the formula The component value of the global optimal position of the group in the qth dimension is obtained, Pointer B rpq The conjugate of , c1 and c2 are two learning factors with different values, r1(k) and r2(k) are two uniform random numbers in the range of [0,1] with different values at the kth iteration; the individual optimal output power and the global optimal output power of each particle are determined again according to the global optimal position update result of the particles in the population.
[0103] In this embodiment, when the damping coefficient is lower than the optimized value, even a slight change in it will cause significant fluctuations in the output power; on the contrary, when the damping coefficient exceeds the optimized threshold, its impact on the output power is significantly reduced, such as Figure 2 As shown, the vertical coordinate P g (pu) means the per-unit value of output power, and the horizontal axis R g (pu) means the per-unit value of the equivalent output resistance in the equivalent circuit. Therefore, the damping coefficient is logarithmically transformed to improve the control effect. The new lnBr-P curve is as follows Figure 3 As shown, the vertical coordinate P g (pu) means the per-unit value of output power, and the horizontal axis is lnB r (pu) means the per-unit value of the equivalent circuit damping coefficient after ln logarithmic transformation.
[0104] Specifically, dB rpq The expression is:
[0105]
[0106] Where ΔP is the output power change, ΔlnB rpq is the ln logarithmic transformation change of the damping coefficient of each particle, and λ is an adjustable parameter.
[0107] The right side of equation (10) consists of three parts: the first part is the "inertia" or "momentum" part, which reflects the particle's "habit" of movement and represents the particle's tendency to maintain its previous speed; the second part is the "cognition" part, which reflects the particle's memory or recollection of its own historical experience and represents the particle's tendency to approach its own historical optimal position; the third part is the "social" part, which reflects the group's historical experience of collaboration and knowledge sharing among particles and represents the particle's tendency to approach the group or neighborhood's historical optimal position. Based on the global optimal position update result of the particles in the population, the individual optimal output power and the global optimal output power of each particle are re-determined.
[0108] In step 7, it is determined whether the annealing temperature satisfies the constraint formula:
[0109] T k <T end (13)
[0110] Among them, T k is the annealing temperature after k annealing operations, T end is the annealing termination temperature. If the constraint (13) is not satisfied, the temperature annealing operation is performed, where the update temperature rule expression is:
[0111] T k+1 =α·T k (14)
[0112] Where α is the annealing coefficient. If the constraint (13) is satisfied, then determine whether the number of iterations satisfies the constraint:
[0113] k <K (15)
[0114] Where k is the number of annealing operations performed, and K is the maximum number of annealing operations set. If constraint (15) is satisfied, the probability calculation is repeated until the maximum number of iterations is reached, and the optimal output power of the objective function and the corresponding approximate optimal damping are output.
[0115] In step 8, when the equivalent circuit exhibits resistance characteristics, the reference speed expression of the power output shaft system is:
[0116]
[0117] Among them, V ref (ω) is the reference speed at the angular frequency ω. The corresponding reference speed at this time is the approximate optimal reference speed.
[0118] Based on the proposed simulated annealing particle swarm optimization algorithm (SAPSO algorithm), the convergence results of the approximate optimal damping coefficient are obtained, such as Figure 4 As shown, the vertical coordinate B r (pu) means the per-unit value of the equivalent circuit damping coefficient; the output power objective function convergence curve is as follows Figure 5 As shown, for the point absorption direct drive wave power generation device and its equivalent circuit, the generator output energy lateral comparison diagram obtained by this method is as follows Figure 6 shown.
[0119] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0120] The present invention provides a method for generating an approximate optimal reference speed for a wave power generation device based on a simulated annealing particle swarm algorithm. While there are numerous methods and approaches for implementing this technical solution, the foregoing description represents only a preferred embodiment of the present invention. It should be noted that those skilled in the art may make various improvements and modifications without departing from the principles of the present invention, and such improvements and modifications are also within the scope of protection of the present invention. Any components not specified in this embodiment may be implemented using existing technologies.
Claims
1. A method for generating an approximate optimal reference speed for a wave power generation device based on a simulated annealing particle swarm algorithm, characterized in that: The following steps are involved: Step 1: Based on the point absorption direct drive wave power generation device, a system frequency domain model is established according to the linear motion equation; Step 2: Consider the force component in the system frequency domain model as the electric component in the circuit, and establish an equivalent circuit model through the force-electricity analogy method; Step 3: Use the working condition optimization algorithm to construct the operation optimization objective function of the motor output power; Step 4: According to the equivalent circuit model, configure the initial parameters of the simulated annealing particle swarm and randomly generate the initial particle population. The initial parameters include random control parameters r1 and r2, population size D, and maximum iteration step size dB. rmax , maximum number of iterations K, initial annealing temperature T0, minimum annealing temperature T end , annealing coefficient α, and learning factors c1 and c2; r1 and r2 are random numbers between 0 and 1; Step 5: Determine whether to accept the current solution based on the Monte Carlo criterion; Step 6: Update the speed and position of the particles in the population according to the global optimal alternative solution, and update the individual output power of the particles in the population and the global optimal output power; Step 7: When the annealing temperature is lower than the termination temperature and the maximum number of iterations is reached, the optimal output power of the objective function and the corresponding approximate optimal damping coefficient are output; Step 8: Obtain the approximate optimal reference speed according to the reference speed expression and the approximate optimal damping coefficient.
2. The method according to claim 1, characterized in that In step 1, the point absorption direct-drive wave power generation device includes a buoy and a permanent magnet synchronous generator, wherein the permanent magnet synchronous generator includes a coil and a permanent magnet; the permanent magnet synchronous generator is used as a power output shaft, and the buoy and the power output shaft are coupled in a vertical direction to form a motion device; Considering only the vertical force, the permanent magnet synchronous generator reciprocates under the drive of the buoy. The coil cuts the magnetic lines of force, generating an induced current. The induced current is regulated to adjust the amplitude of the generator force. Based on Newton's second law and Cummins equation, the linear motion equation is: Among them, M and m ∞ are the mass of the moving device and the additional mass at infinite frequency, K r (t) is the kernel function convolved with the velocity at time t, K h is the sum of the hydrostatic stiffness coefficient and the spring stiffness coefficient, F e (t) is the excitation force generated by the incident wave at time t, F g (t) is the back electromagnetic force of the permanent magnet synchronous generator at time t, x(t) and v(t) are the vertical displacement of the moving device at time t and the velocity of the moving device at time t, respectively.
3. The method according to claim 2, characterized in that In step 1, by performing frequency domain transformation on formula (1), the following system frequency domain model is obtained: Where j is an imaginary unit, ω is the angular frequency, V(ω) is the velocity of the converter and float at the angular frequency ω, and F e (ω) is the excitation force generated by the incident wave at angular frequency ω, F g (ω) is the counter-electromagnetic force of the generator at angular frequency ω, K r (ω) is the kernel function convolved with velocity at angular frequency ω, K r (ω)=B r (ω)+jωA r (ω), A r (ω) and B r (ω) are the so-called radiative added mass and radiation damping at the angular frequency ω, respectively.
4. The method according to claim 3, characterized in that In step 2, the velocity V(ω) in formula (2) is used as the current component in the circuit; the excitation force is used as the voltage source; and the counter-electromagnetic force is used as the impedance. Coefficient K r (ω),ω(M+m ∞ )and As the inherent resistive and inductive loads in the circuit, the equivalent circuit model is obtained; according to the basic Ohm's law of the circuit, the expression of the speed V(ω) is obtained as follows: Among them, Z w (ω) is the system inherent equivalent impedance of the power output shaft at the angular frequency ω, 5. The method according to claim 4, characterized in that In step 3, the objective function of the operation optimization problem of the motor output power is the average output power model derived from the point absorption direct-drive wave power generation device model, which is expressed as: in is the average output power of the point absorption direct-drive wave power generation device at the angular frequency ω, Re[] means the real part of the formula in the square brackets, V * (ω) is the conjugate of the power take-off shaft system speed at angular frequency ω; The operating condition optimization algorithm makes the speed and the excitation force in phase, so that the equivalent circuit presents a resistance characteristic. It requires that the control force applied by the power output shaft conforms to the following equation: in is Z at angular frequency ω w The conjugate of (ω), Z g (ω) is the output impedance of the motor at the angular frequency ω; With the goal of maximizing output power, an operation optimization objective function is established to satisfy: max[f(x)]=max[P(B r )] (6) in For the equivalent circuit with a damping coefficient of B r The maximum output power expression when B r is the damping coefficient in the equivalent circuit; R g is the equivalent output resistance, is the conjugate of the damping coefficient in the equivalent circuit.
6. The method according to claim 5, characterized in that In step 5, the constraints for accepting the current solution are: P(B r(k+1) )>P(B r(k) ) (7) Among them B r(k+1) is the damping coefficient selected at the (k+1)th iteration; if the constraints are met, the current solution is accepted; otherwise, the current solution is accepted with a specific probability. The specific probability expression is: Where T k is the annealing temperature after k annealing operations, and exp refers to the natural exponential function.
7. The method according to claim 6, characterized in that In step 6, the speed and position of the particles in the population are updated according to the global optimal alternative solution, satisfying: dB rpq (k+1)=dB rpq (k)+c1r1(k)[P pq (k)-B rpq (k)]+c2r2(k)[P gq (k)-B rpq (k)] (9) B rpq (k+1)=B rpq (k)+dB rpq (k+1) (10) Where p, q = 1, 2, ..., D; D is the selected particle dimension; dB rpq (k) is the iteration step size of each particle at the kth iteration, dB rpq (k)∈[-dB rmax , dB rmax ], B rpq (k) represents the approximate maximum damping coefficient of each particle at the kth iteration, dB rmax Indicates dB rpq (k) Upper limit of value, P pq (k) is the time when B rpq Substitute the value of into the formula The obtained individual optimal position of the particle itself is the component value in the qth dimension, P gq (k) is the time when B rpq Substitute the value of into the formula The component value of the global optimal position of the group in the qth dimension is obtained. Pointer B rpq The conjugate of , c1 and c2 are two learning factors with different values, r1(k) and r2(k) are two uniform random numbers in the range of [0,1] with different values at the kth iteration; the individual optimal output power and the global optimal output power of each particle are determined again according to the global optimal position update result of the particles in the population.
8. The method according to claim 7, characterized in that In step 7, it is determined whether the annealing temperature satisfies the following constraint: T k <T end (11) Among them, T end is the annealing termination temperature; If the constraint (11) is not satisfied, the temperature annealing operation is performed and the updated temperature rule expression is: T k+1 =α·T k (12) Where α is the annealing coefficient; If the constraint (11) is satisfied, then determine whether the number of iterations satisfies the following constraint: k <K (13) Where k is the number of annealing operations performed, and K is the maximum number of annealing operations set. If constraint (13) is satisfied, the probability calculation is repeated using formula (8) until the maximum number of iterations is reached, and the optimal output power of the objective function and the corresponding approximate optimal damping are output.
9. The method according to claim 8, characterized in that In step 8, when the equivalent circuit exhibits resistance characteristics, the reference speed expression of the power output shaft system is: Among them, V ref (ω) is the reference velocity at angular frequency ω.
10. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores program codes, and when the program codes are executed by the processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 9.
Citation Information
Patent Citations
Wave power generation method based on simulated annealing particle swarm algorithm
CN108899936A
Microgrid optimization method based on simulated annealing particle swarm algorithm
CN111178642A
Optimization method of SVC power oscillation damping controller
CN115933391A
Power spring control method based on improved simulated annealing particle swarm optimization QPR control
CN118573039A
AU2020103709A4
