Photovoltaic array double-diode seven-parameter model parameter identification method

Abstract

The application discloses a photovoltaic array double-diode seven-parameter model parameter identification method. Based on the application of the combination of the improved particle swarm algorithm and the analytic method, two diode ideal factors and series equivalent resistance are solved by optimization, and then the photogenerated current, parallel equivalent resistance and two diode reverse saturation currents are solved by the analytic method. In order to improve the performance of the improved algorithm, a double fitness function is proposed, and an adaptive evolution learning and adaptive mutation operator are introduced. The application solves the problem of difficulty in solving the double-diode seven-parameter model parameters of the photovoltaic array and the problem of low solving precision of the single-diode model parameters.

Description

TECHNICAL FIELD

[0001] The application relates to a photovoltaic array double-diode seven-parameter model parameter identification method and belongs to the technical field of photovoltaic power generation systems. BACKGROUND

[0002] At present, models for describing the output characteristics of photovoltaic arrays mainly include a single-diode five-parameter model and a double-diode seven-parameter model. The single-diode model has become the main model used in engineering due to its few parameters and simple calculation. However, the double-diode seven-parameter model has relatively higher precision, especially under low light conditions. This is mainly because the single-diode model ignores the recombination saturation current of the PN junction depletion region.

[0003] The methods for identifying the parameters of the double-diode seven-parameter model of the photovoltaic array mainly include an analytical method, a numerical method and an intelligent algorithm. The analytical method needs to make certain assumptions or ignore some parameters, and is simple and fast to solve, but has relatively low precision. The numerical method generally needs to introduce new parameters, such as the voltage / current temperature coefficient, the series equivalent resistance at the open-circuit voltage and the parallel equivalent resistance at the short-circuit current, to construct a new equation, and is sensitive to the initial value when solving and is prone to fall into local optimization or even fail to solve the value. Intelligent algorithms have been widely used in photovoltaic array modeling due to their universal global search capability and effectiveness in processing nonlinear functions. The commonly used intelligent algorithms include the particle swarm algorithm, the genetic algorithm, the neural network and the simulated annealing algorithm. Such random algorithms are generally unstable and slow in convergence, and directly using the intelligent algorithm to extract the seven parameters of the double-diode seven-parameter model is relatively complex and has large calculation amount.

[0004] Therefore, how to simply and conveniently and accurately extract the seven parameters of the double-diode seven-parameter model of the photovoltaic array is of great significance to the research and development of photovoltaic power generation systems. SUMMARY

[0005] The application provides a photovoltaic array double-diode seven-parameter model parameter identification method, which combines the improved particle swarm algorithm and the analytical method, improves the solving precision and reduces the algorithm complexity.

[0006] To achieve the above object, the technical scheme adopted by the application is as follows:

[0007] The application provides a photovoltaic array double-diode seven-parameter model parameter identification method, and the seven parameters of the double-diode seven-parameter model include two diode ideal factors, a series equivalent resistance, a photo-generated current, a parallel equivalent resistance and two diode reverse saturation currents.

[0008] The method comprises the following steps:

[0009] 1) Establishing a seven-parameter model of photovoltaic array with dual diodes;

[0010] 2) Extracting two diode ideal factor and series equivalent resistance in the seven-parameter model of photovoltaic array with dual diodes by using improved particle swarm algorithm;

[0011] 3) Solving the photo-generated current, parallel equivalent resistance and two diode reverse saturation current in the seven-parameter model of photovoltaic array with dual diodes by using analytical method based on the extracted two diode ideal factor and series equivalent resistance.

[0012] Further, the seven-parameter model of photovoltaic array with dual diodes is established, including the following steps:

[0013] 1-1) The current output equation of the seven-parameter model of photovoltaic array with dual diodes is:

[0014]

[0015] In the formula, I is the output current of the photovoltaic array, V is the output voltage of the photovoltaic array, I ph is the photo-generated current, I o1 and I o2 are the reverse saturation currents of two diodes, a1 and a2 are the diode ideal factors, N s is the number of series cells, R s and R p are the series equivalent resistance and parallel equivalent resistance respectively, q is the electronic charge (1.6e -19 C), k is the Boltzmann constant (1.38e -23 J / K), and T is the absolute temperature of the photovoltaic array.

[0016] 1-2) Taking a1, a2 and R s in the seven-parameter model of photovoltaic array with dual diodes as the algorithm optimization parameters, and the remaining four parameters I ph , R p , I o1 and I o2 can be expressed as functions of a1, a2 and R s ;

[0017] In order to facilitate the subsequent expression of the current output equation when the short-circuit current, open-circuit voltage and maximum power point are substituted, let:

[0018]

[0019]

[0020]

[0021] In the formula, Isc Vshunt is short circuit current, V oc Vopen is open circuit voltage, V m Vmp is maximum power point voltage, V m Imax is maximum power point current, A

[0022] In short circuit state, the short circuit current point (0, I sc ) is substituted into formula (1) to obtain:

[0023]

[0024] In open circuit state, the open circuit voltage point (V oc , 0) is substituted into formula (1) to obtain:

[0025]

[0026] In maximum power point state, the maximum power point (V m , I m ) is substituted into formula (1) to obtain:

[0027]

[0028] In maximum power point state, the derivative of voltage is 0, that is:

[0029]

[0030] The derivative of formula (1) is carried out and substituted into formula (8) to obtain:

[0031]

[0032] The simultaneous equations of formula (5), formula (6), formula (7) and formula (9) can be solved to obtain:

[0033]

[0034]

[0035]

[0036]

[0037] In the formula,

[0038]

[0039]

[0040]

[0041] D = l m V oc -Vm I sc (14)

[0042] In the formula, A i B i C and D are variables in the substitution formula, I ph Io1 and Io2 are the photocurrents, Io1 and Io2 are the reverse saturation currents of the two diodes, and R is the photocurrent. p For parallel equivalent resistance;

[0043] 1-3) Short-circuit current I under different illumination and temperature sc Maximum power point current I m Open circuit voltage V oc Maximum power point voltage V m The parameter update formula is:

[0044]

[0045]

[0046]

[0047]

[0048] In the formula, I scref I mref V ocref V mref S ref and T ref These are the short-circuit current, maximum power point current, open-circuit voltage, maximum power point voltage, illuminance, and absolute temperature of the photovoltaic array under standard operating conditions, K. i K represents the temperature coefficient of current in a photovoltaic array. v Let α be the voltage temperature coefficient of the photovoltaic array, α be the voltage-illuminance correction coefficient of the photovoltaic array, and S be the actual illuminance of the photovoltaic array.

[0049] Furthermore, the parameters of the seven-parameter model of the photovoltaic array dual diodes all have certain constraints, which can be described as follows:

[0050] g j (x)≤0,j=1,2,...,J (19)

[0051] In the formula, x = [a1, a2, R] s ], where J is the number of inequality constraints.

[0052] Furthermore, a dual fitness value comparison method is used to extract the ideality factors and series equivalent resistances of the two diodes, separating the objective function and constraints; each constraint is standardized; the two fitness functions are as follows:

[0053]

[0054]

[0055] In the formula, x i Let i represent the i-th particle, the fit function correspond to the objective function value, represent the root mean square error (RMSE) between the model-calculated current value and the actual measured current value, the vio function correspond to the constraint conditions, and represent the degree to which the particle violates the constraints, n is the number of measurement data sets, and I cl Calculate the output current value I for the model corresponding to the l-th group of measured voltages. el The actual measured current value for group l. N is the number of particles in the population.

[0056] Furthermore, an improved particle swarm optimization algorithm is used to extract the ideality factors of the two diodes and the series equivalent resistance from the seven parameters of the photovoltaic array dual-diode seven-parameter model, including the following steps:

[0057] 2-1) Set the algorithm parameters, operating condition parameters, photovoltaic array parameters, and the value range of the seven parameters in the dual-diode seven-parameter model;

[0058] Randomly initialize the positions and velocities of N particles within the range of values; update the short-circuit current I under different operating conditions according to equations (15) to (18). sc Maximum power point current I m Open circuit voltage V oc and maximum power point voltage V m ;

[0059] Solve for the remaining four parameters of the seven-parameter model of the photovoltaic array dual diodes according to equations (10) to (13): Photocurrent I ph Parallel equivalent resistance R p The reverse saturation current I of the two diodes o1 and I o2 The initial population of particles discards infeasible solutions to ensure that all seven parameters of the seven-parameter model of the photovoltaic array dual diode are within the range of feasible solutions.

[0060] 2-2) Use Newton's iteration method to obtain the approximate solution of the model output current corresponding to the measured voltage, calculate the particle objective function value, and obtain the individual optimal position pbest and the global optimal position gbest. Set the particle violation degree to 0.

[0061] 2-3) Update the particle's position and velocity using an adaptive evolutionary learning approach:

[0062]

[0063]

[0064]

[0065]

[0066] In the formula, ω is the inertia factor, ω max ω min These represent its maximum and minimum values, respectively. min and fit avg Representing the minimum and average objective function values ​​respectively, the h function corresponds to the particle's adaptive evolutionary learning factor, t is the current iteration number, c1 and c2 are the individual learning factor and the social learning factor respectively, r1 and r2 are random numbers in [0,1], v i ,x i pbest i Let represent the velocity, position, and optimal position of the i-th particle, respectively, and gbest be the global optimal position;

[0067] After the particle position and velocity are updated, unreasonable particles are processed to avoid exceeding the limit. If the position and velocity values ​​exceed the maximum value, the maximum value is taken; if they exceed the minimum value, the minimum value is taken.

[0068] 2-4) Solve for the remaining four parameter values ​​of the photovoltaic array dual-diode seven-parameter model according to equations (10) to (13): Photocurrent I ph Parallel equivalent resistance R p The reverse saturation current I of the two diodes o1 and I o2 ; Calculate the particle dual fitness function value according to equations (20) and (21);

[0069] 2-5) The formula for calculating the aggregation degree δ of the t-th generation particles is:

[0070]

[0071] In the formula, Let be the average, maximum, and minimum values ​​of the objective function for the t-th generation particle, respectively. For the t-th generation particle x i The objective function value;

[0072] The mutation probability of each generation of particles is dynamically adjusted based on the aggregation degree δ:

[0073]

[0074] In the formula, β is a constant used to adjust the rate of change of the mutation probability, and its value ranges from [2,4].

[0075] The random number r generated in the range [0,1] is less than the mutation probability. At this time, the optimal position of an individual particle is mutated using the Cauchy distribution. After the mutation operation, the double fitness function value of the optimal individual particle is recalculated.

[0076] pbest=pbest(1+0.5tan(π(rand-0.5))) (28)

[0077] 2-6) Update pbest according to the particle comparison principle, including:

[0078] a) When two particles x i and x j When both are feasible, particles with smaller fit function values ​​are preferred;

[0079] b) When two particles x i and x j When neither is feasible, the particle with the smaller vio function value is preferred;

[0080] c) When particle x i Feasible and x j If it is not feasible, if particle x j If the value of the vio function is less than a predetermined small positive number ε, then the particle with the smaller fit function value is preferred; otherwise, the particle x is preferred. i Excellent;

[0081] Based on the particle comparison principle, update gbest, including:

[0082] a) When two particles x i and x j When both are feasible, particles with smaller fit function values ​​are preferred;

[0083] b) When particle x i Feasible and x j When it is not feasible, particle x i It is excellent.

[0084] 2-7) Repeat steps 2-3 to 2-6 until the following relationship holds true for a predetermined number of times M or the algorithm reaches its maximum iteration count. Output the two diode ideality factors a1 and a2 and the series equivalent resistance R. s The optimal solution;

[0085] fit(gbest t-1 )-fit(gbest t ) < e min (29)

[0086] In the formula, e min For accuracy requirements, fit(gbest) t) represents the objective function value of the globally optimal particle gbest in generation t.

[0087] Furthermore, based on the extracted ideality factors and series equivalent resistance of the two diodes, the analytical method for solving the photogenerated current, parallel equivalent resistance, and reverse saturation current of the two diodes in the seven-parameter model of a photovoltaic array dual-diode includes:

[0088] The remaining four parameter values ​​I of the photovoltaic array dual-diode model are solved using the following formula. ph R p I o1 and I o2 ;

[0089]

[0090]

[0091]

[0092]

[0093] In the formula,

[0094]

[0095]

[0096]

[0097] D = I m V oc -V m I sc (14)

[0098] In the formula, A i B i C and D are variables in the substitution formula, I ph For photocurrent, I o1 and I o2 R represents the reverse saturation current of the two diodes. p It represents the parallel equivalent resistance.

[0099] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0100] This invention combines an improved particle swarm optimization algorithm with analytical methods to solve the problems of difficulty in solving the parameters of the seven-parameter model of a dual-diode photovoltaic array and low accuracy in solving the parameters of a single-diode model, providing a new approach for parameter identification of photovoltaic arrays.

[0101] Introducing adaptive evolutionary learning can not only improve the search capability of constraint boundaries, but also accelerate the convergence speed.

[0102] Introducing an adaptive Gaussian mutation operator can improve population diversity and prevent premature convergence. Attached Figure Description

[0103] Figure 1 This is a flowchart of the method of the present invention;

[0104] Figure 2 The equivalent circuit of the seven-parameter model of the dual diodes in a photovoltaic array;

[0105] Figure 3 This invention presents measured data and simulated IV output characteristic curves of the photovoltaic array under different illumination and temperature conditions. Detailed Implementation

[0106] The parameter identification method for the seven-parameter model of a photovoltaic array dual diode provided by the present invention will be described in detail below with reference to the accompanying drawings.

[0107] This embodiment provides a method for parameter identification of a seven-parameter model of a photovoltaic array dual diode, such as... Figure 1 As shown, it includes the following steps:

[0108] 1) Establish a seven-parameter model of a photovoltaic array with two diodes;

[0109] 2) An improved particle swarm optimization algorithm is used to extract the ideality factors of the two diodes and the series equivalent resistance from the seven parameters of the photovoltaic array dual-diode seven-parameter model;

[0110] 3) Based on the extracted ideal factors of the two diodes and the series equivalent resistance, the photogenerated current, parallel equivalent resistance and reverse saturation current of the two diodes in the seven-parameter model of the photovoltaic array dual diode are solved analytically.

[0111] Step 1) establishes a seven-parameter model of the photovoltaic array dual diodes, including the following steps:

[0112] 1-1) Figure 2 The equivalent circuit of the seven-parameter model of the photovoltaic array dual diodes can be obtained according to Holkieff's current law:

[0113]

[0114] In the formula, I is the output current of the photovoltaic array, V is the output voltage of the photovoltaic array, and I ph For photocurrent, I o1 and I o2 Let a1 and a2 be the reverse saturation currents of the two diodes, and N be the diode ideality factor. sR is the number of cells connected in series. s and R p These are the equivalent resistances in series and parallel, respectively, where q is the electron charge (1.6e). -19 C), k is the Boltzmann constant (1.38e -23 J / K), where T is the absolute temperature of the photovoltaic array;

[0115] 1-2) The seven parameters a1, a2, and R in the seven-parameter model of the photovoltaic array dual diode are... s As parameters to be optimized in the algorithm, the other four parameters I ph R p I o1 and I o2 It can be represented as a1, a2, and R. s The function;

[0116] For the sake of simplicity in expressing the current output equation when substituting the short-circuit current, open-circuit voltage, and maximum power point into the equation, let:

[0117]

[0118]

[0119]

[0120] In the formula, I sc For short-circuit current, V oc V is the open-circuit voltage. m I is the voltage at the maximum power point. m This is the current at the maximum power point;

[0121] Under short-circuit conditions, the short-circuit current point (0, I) sc Substituting into equation (1), we get:

[0122]

[0123] In the open-circuit state, the open-circuit voltage point (V) oc Substituting 0 into equation (1) yields:

[0124]

[0125] At the maximum power point, the maximum power point (V) m I m Substituting into equation (1), we get:

[0126]

[0127] At the maximum power point, the derivative with respect to voltage is 0, that is:

[0128]

[0129] Differentiating equation (1) and substituting it into equation (8), we get:

[0130]

[0131] By solving the system of equations (5), (6), (7), and (9), we can obtain the following:

[0132]

[0133]

[0134]

[0135]

[0136] In the formula,

[0137]

[0138]

[0139]

[0140] D = l m V oc -V m I sc (14)

[0141] In the formula, A i B i C and D are variables in the substitution formula, I ph For photocurrent, I o1 and I o2 R represents the reverse saturation current of the two diodes. p For parallel equivalent resistance;

[0142] 1-3) Short-circuit current I under different illumination and temperature sc Maximum power point current I m Open circuit voltage V oc Maximum power point voltage V m The parameter update formula is:

[0143]

[0144]

[0145]

[0146]

[0147] In the formula, I scref I mref V ocref V mref S ref and T ref These are the short-circuit current, maximum power point current, open-circuit voltage, maximum power point voltage, illuminance, and absolute temperature of the photovoltaic array under standard operating conditions, K. i K represents the temperature coefficient of current in a photovoltaic array. v Let α be the voltage temperature coefficient of the photovoltaic array, α be the voltage-illuminance correction coefficient of the photovoltaic array, and S be the actual illuminance of the photovoltaic array.

[0148] 1-4) In the seven-parameter model of a photovoltaic array with two diodes, the ideality factors a1 and a2 of the two diodes and the series equivalent resistance R are included. s The model remains within the feasible solution range, meaning it only needs to handle constraints on the remaining four parameters. These constraints can be described as follows:

[0149] g j (x)≤0,j=1,2,...,J (19)

[0150] In the formula, x = [a1, a2, R] s ], where J is the number of inequality constraints;

[0151] 1-5) When extracting the ideality factors and series equivalent resistances of the two diodes, a dual fitness value comparison method is used to separate the objective function and constraints; in addition, to mitigate the differences between the constraints, each constraint is standardized; the two fitness functions are as follows:

[0152]

[0153]

[0154] In the formula, x i Let i represent the i-th particle, the fit function correspond to the objective function value, represent the root mean square error (RMSE) between the model-calculated current value and the actual measured current value, the vio function correspond to the constraint conditions, and represent the degree to which the particle violates the constraints, n is the number of measurement data sets, and I cl Calculate the output current value I for the model corresponding to the l-th group of measured voltages. el The actual measured current value for group l. N is the number of particles in the population.

[0155] In step 2), an improved particle swarm optimization algorithm is used to extract the ideality factors of the two diodes and the series equivalent resistance from the seven parameters of the photovoltaic array dual-diode seven-parameter model, including the following steps:

[0156] 2-1) Set the algorithm parameters: learning factors c1 and c2 are 2, population size is 20, maximum number of iterations is 1000, maximum and minimum inertia factors are 0.9 and 0.4 respectively, and parameters a1, a2 and R... s The maximum speeds are 0.1, 0.2, and 0.1 respectively, and the algorithm accuracy is e. min The value is 1e-5, and the number of iterations M is 50; set the operating parameters and photovoltaic array parameters; set the value ranges of a1 and a2 to [1,2] and [1,4], respectively. ph The value range is [0.95I]. sc 1.05I sc ], I o1 The value range is [0, I o2 ], I o2 The value range is [0, 0.1]. sc ], R s and R p The value range is [0.01, 3] and [50, 5000];

[0157] Randomly initialize the positions and velocities of N particles within the range of values; update the short-circuit current I under different operating conditions according to equations (15) to (18). sc Maximum power point current I m Open circuit voltage V oc and maximum power point voltage V m ;

[0158] Solve for the remaining four parameters of the seven-parameter model of the photovoltaic array dual diodes according to equations (10) to (13): Photocurrent I ph Parallel equivalent resistance R p The reverse saturation current I of the two diodes o1 and I o2 The initial population of particles discards infeasible solutions, thus ensuring that all seven parameters of the seven-parameter model of the photovoltaic array dual diode are within the range of feasible solutions.

[0159] 2-2) Use Newton's iteration method to obtain the approximate solution of the model output current corresponding to the measured voltage, calculate the particle objective function value, and obtain the individual optimal position pbest and the global optimal position gbest. Set the particle violation degree to 0.

[0160] 2-3) Update the particle's position and velocity using an adaptive evolutionary learning approach:

[0161]

[0162]

[0163]

[0164]

[0165] In the formula, ω is the inertia factor, ω max ω min These represent its maximum and minimum values, respectively. min and fit avg Representing the minimum and average objective function values ​​respectively, the h function corresponds to the particle's adaptive evolutionary learning factor, t is the current iteration number, c1 and c2 are the individual learning factor and the social learning factor respectively, r1 and r2 are random numbers in [0,1], v i ,x i pbest i Let represent the velocity, position, and optimal position of the i-th particle, respectively, and gbest be the global optimal position. Introducing adaptive evolutionary learning can not only improve the search capability of the constraint boundary, but also accelerate the convergence speed.

[0166] After the particle position and velocity are updated, unreasonable particles are processed to avoid exceeding the limit. If the position and velocity values ​​exceed the maximum value, the maximum value is taken; if they exceed the minimum value, the minimum value is taken.

[0167] 2-4) Solve for the remaining four parameter values ​​of the photovoltaic array dual-diode seven-parameter model according to equations (10) to (13): Photocurrent I ph Parallel equivalent resistance R p The reverse saturation current I of the two diodes o1 and I o2 ; Calculate the particle dual fitness function value according to equations (20) and (21);

[0168] 2-5) The formula for calculating the aggregation degree δ of the t-th generation particles is:

[0169]

[0170] In the formula, Let be the average, maximum, and minimum values ​​of the objective function for the t-th generation particle, respectively. For the t-th generation particle x i The objective function value;

[0171] The mutation probability of each generation of particles is dynamically adjusted based on the aggregation degree δ:

[0172]

[0173] In the formula, β is a constant used to adjust the rate of change of the mutation probability, and its value ranges from [2,4].

[0174] The random number r generated in the range [0,1] is less than the mutation probability. At this time, the optimal position of an individual particle is mutated using the Cauchy distribution. After the mutation operation, the double fitness function value of the optimal individual particle is recalculated.

[0175] pbest=pbest(1+0.5tan(π(rand-0.5))) (28)

[0176] 2-6) Update pbest according to the particle comparison principle, specifically:

[0177] a) When two particles x i and x j When both are feasible, particles with smaller fit function values ​​are preferred;

[0178] b) When two particles x i and x j When neither is feasible, the particle with the smaller vio function value is preferred;

[0179] c) When particle x i Feasible and x j If it is not feasible, if particle x j If the value of the vio function is less than a predetermined small positive number ε, then the particle with the smaller fit function value is preferred; otherwise, the particle x is preferred. i Excellent;

[0180] Based on the particle comparison principle, update gbest as follows:

[0181] a) When two particles x i and x j When both are feasible, particles with smaller fit function values ​​are preferred;

[0182] b) When particle x i Feasible and x j When it is not feasible, particle x i Excellent;

[0183] 2-7) Repeat steps 2-3 to 2-6 until the following relation is true M times consecutively or the algorithm reaches its maximum number of iterations. Output a1, a2, and R. s The optimal solution;

[0184] fit(gbest t-1 )-fit(gbest t ) < e min (29)

[0185] In the formula, e min For accuracy requirements, fit(gbest) t ) represents the objective function value of the globally optimal particle gbest in generation t.

[0186] In step 3), based on the two extracted diode ideality factors a1 and a2 and the series equivalent resistance R... s The optimal solution is obtained by solving the remaining four parameter values ​​of the photovoltaic array dual diode seven-parameter model according to equations (10) to (13): photocurrent I ph Parallel equivalent resistance R p The reverse saturation current I of the two diodes o1 and I o2 At this point, all seven parameters in the dual-diode seven-parameter model have been extracted through the steps described above.

[0187] This embodiment estimates the IV output characteristic curve of the photovoltaic array (CSUN340-72M). The characteristic parameters of the photovoltaic array under standard operating conditions are shown in Table 1:

[0188] Table 1 Characteristic parameters of CSUN340-72M photovoltaic array

[0189] Parameter Parameter value Parameter Parameter value I sc (A) 9.62 K i (°C) (% / °C) 0.039 U oc (V)]]> 47.6 K v (°C) (% / °C) -0.307 I m (A) 8.89 <![CDATA[N s ]]> 72 U m (V)]]> 37.9 α 0.005

[0190] Table 2. Root mean square error of measured and simulated currents of photovoltaic arrays.

[0191] Number of groups irradiance (W / m 2 )]]> Backplane temperature (°C) RMSE (A) 1 872 38.5 0.0741 2 677 30.5 0.1234 3 758 34.8 0.0976 4 826 40.3 0.0872 5 492 27.5 0.1108

[0192] Figure 3 Table 2 shows the measured data and simulated IV output characteristic curves of the photovoltaic array under different illumination and temperature conditions. Figure 3 The root mean square error between the measured current and the simulated current under different illumination and temperature conditions.

[0193] As shown in Table 2, the RMSE between the measured current and the simulated current of the photovoltaic array does not exceed 0.13A. This invention has successfully simulated the IV output characteristics of the CSUN340-72M photovoltaic array.

[0194] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for parameter identification of a seven-parameter model of a photovoltaic array dual diode, characterized in that, The seven parameters of the dual-diode seven-parameter model include: ideality factor of the two diodes, series equivalent resistance, photocurrent, parallel equivalent resistance, and reverse saturation current of the two diodes. The method includes the following steps: Establish a seven-parameter model of a photovoltaic array with two diodes; An improved particle swarm optimization algorithm was used to extract the ideality factors and series equivalent resistances of the two diodes in the seven-parameter model of the photovoltaic array dual diodes. Based on the extracted ideal factors and series equivalent resistance of the two diodes, the photogenerated current, parallel equivalent resistance and reverse saturation current of the two diodes in the seven-parameter model of the photovoltaic array dual diodes are solved analytically. The method for establishing the seven-parameter model of the photovoltaic array dual diodes includes: using the ideality factors of two diodes and the series equivalent resistance of the seven parameters in the photovoltaic array dual diode seven-parameter model as the parameters to be optimized in the algorithm, and the remaining four parameters: photogenerated current, parallel equivalent resistance, and reverse saturation current of the two diodes as functions of the parameters to be optimized, including the following steps: 1-1) The current output equation of the seven-parameter model of the photovoltaic array dual diodes is: In the formula, I is the output current of the photovoltaic array, V is the output voltage of the photovoltaic array, and I ph For photocurrent, I o1 and I o2 Let a1 and a2 be the reverse saturation currents of the two diodes, and N be the diode ideality factor. s R is the number of cells connected in series. s and R p These are the series equivalent resistance and the parallel equivalent resistance, respectively; q is the electron charge; k is the Boltzmann constant; and T is the absolute temperature of the photovoltaic array. 1-2) Order: In the formula, I sc For short-circuit current, V oc V is the open-circuit voltage. m I is the voltage at the maximum power point. m This is the maximum power point current; Under short-circuit conditions, the short-circuit current point (0, I) sc Substituting into equation (1), we get: In the open-circuit state, the open-circuit voltage point (V) oc Substituting 0 into equation (1) yields: At the maximum power point, the maximum power point (V) m I m Substituting into equation (1), we get: At the maximum power point, the derivative with respect to voltage is 0, that is: Differentiating equation (1) and substituting it into equation (8), we get: By solving the system of equations (5), (6), (7), and (9), we can obtain the following: In the formula, D=I m V oc -V m I sc (14) In the formula, A i B i C and D are variables in the substitution formula, I ph For photocurrent, I o1 and I o2 R represents the reverse saturation current of the two diodes. p For parallel equivalent resistance; 1-3) Short-circuit current I under different illumination and temperature sc Maximum power point current I m Open circuit voltage V oc Maximum power point voltage V m The parameter update formula is: In the formula, I scref I mref V ocref V mref S ref and T ref These are the short-circuit current, maximum power point current, open-circuit voltage, maximum power point voltage, illuminance, and absolute temperature of the photovoltaic array under standard operating conditions, K. i K represents the temperature coefficient of current in a photovoltaic array. v Let α be the voltage temperature coefficient of the photovoltaic array, α be the voltage-illuminance correction coefficient of the photovoltaic array, and S be the actual illuminance of the photovoltaic array. The method for establishing the seven-parameter model of the photovoltaic array dual diodes further includes: using a dual fitness value comparison method to separate the objective function and constraints when extracting the ideal factors and series equivalent resistance of the two diodes; standardizing each constraint; and the two fitness functions are as follows: In the formula, x i Let i represent the i-th particle, the fit function correspond to the objective function value, represent the root mean square error between the model-calculated current value and the actual measured current value, the vio function correspond to the constraint conditions, and represent the degree to which the particle violates the constraints, n is the number of measurement data sets, and I cl Calculate the output current value I for the model corresponding to the l-th group of measured voltages. el The actual measured current value for group l. N is the number of particles in the population; g j (x) represents the constraint conditions of the seven-parameter model of the photovoltaic array dual diode, and J represents the number of inequality constraints.

2. The parameter identification method for a seven-parameter model of a photovoltaic array dual diode according to claim 1, characterized in that, The parameters of the seven-parameter model of a photovoltaic array dual diode all have certain constraints, described as follows: g j (x)≤0,j=1,2,…,J (19) In the formula, x = [a1, a2, R] s ], J is the number of inequality constraints, a1 and a2 are diode ideality factors, R s It is the equivalent resistance in series.

3. The parameter identification method for a seven-parameter model of a photovoltaic array dual diode according to claim 1, characterized in that, An improved particle swarm optimization algorithm is used to extract the ideality factors of two diodes and the series equivalent resistance from the seven parameters of a photovoltaic array dual-diode seven-parameter model. The steps include: 2-1) Set the value ranges of the algorithm parameters, operating condition parameters, photovoltaic array parameters, and the seven parameters in the dual-diode seven-parameter model; Randomly initialize the positions and velocities of N particles within the range of values; update the short-circuit current I under different operating conditions according to equations (15) to (18). sc Maximum power point current I m Open circuit voltage V oc and maximum power point voltage V m ; Solve for the remaining four parameters of the seven-parameter model of the photovoltaic array dual diodes according to equations (10) to (13): Photocurrent I ph Parallel equivalent resistance R p The reverse saturation current I of the two diodes o1 and I o2 The initial population of particles discards infeasible solutions to ensure that all seven parameters of the seven-parameter model of the photovoltaic array dual diode are within the range of feasible solutions. 2-2) Use Newton's iteration method to obtain the approximate solution of the model output current corresponding to the measured voltage, calculate the particle objective function value, and obtain the individual optimal position pbest and the global optimal position gbest. Set the particle violation degree to 0. 2-3) Update the particle's position and velocity using an adaptive evolutionary learning approach: In the formula, ω is the inertia factor, ω max ω min These represent its maximum and minimum values, respectively. min and fit avg Representing the minimum and average objective function values ​​respectively, the h function corresponds to the particle's adaptive evolutionary learning factor, t is the current iteration number, c1 and c2 are the individual learning factor and the social learning factor respectively, r1 and r2 are random numbers in [0,1], v i ,x i pbest i Let represent the velocity, position, and optimal position of the i-th particle, respectively, and gbest be the global optimal position; After the particle position and velocity are updated, unreasonable particles are processed to avoid exceeding the limit. If the position and velocity values ​​exceed the maximum value, the maximum value is taken; if they exceed the minimum value, the minimum value is taken. 2-4) Solve for the remaining four parameter values ​​of the photovoltaic array dual-diode seven-parameter model according to equations (10) to (13): Photocurrent I ph Parallel equivalent resistance R p The reverse saturation current I of the two diodes o1 and I o2 ; Calculate the particle dual fitness function value according to equations (20) and (21); 2-5) The formula for calculating the aggregation degree δ of the t-th generation particles is: In the formula, Let be the average, maximum, and minimum values ​​of the objective function for the t-th generation particle, respectively. For the t-th generation particle x i The objective function value; The mutation probability of each generation of particles is dynamically adjusted based on the aggregation degree δ: In the formula, β is a constant used to adjust the rate of change of the mutation probability, and its value ranges from [2,4]. The random number r generated in the range [0,1] is less than the mutation probability. At this time, the optimal position of an individual particle is mutated using the Cauchy distribution. After the mutation operation, the double fitness function value of the optimal individual particle is recalculated. pbest=pbest(1+0.5tan(π(rand-0.5))) (28) 2-6) Update pbest according to the particle comparison principle, including: a) When two particles x i and x j When both are feasible, particles with smaller fit function values ​​are preferred; b) When two particles x i and x j When neither is feasible, the particle with the smaller vio function value is preferred; c) When particle x i Feasible and x j If it is not feasible, if particle x j If the value of the vio function is less than a predetermined small positive number ε, then the particle with the smaller fit function value is preferred; otherwise, the particle x is preferred. i Excellent; Based on the particle comparison principle, update gbest, including: a) When two particles x i and x j When both are feasible, particles with smaller fit function values ​​are preferred; b) When particle x i Feasible and x j When it is not feasible, particle x i Excellent; 2-7) Repeat steps 2-3 to 2-6 until the following relationship holds true for a predetermined number of times M or the algorithm reaches its maximum iteration count. Output the two diode ideality factors a1 and a2 and the series equivalent resistance R. s The optimal solution; fit(gbest t-1 )-fit(gbest t )<e min (29) In the formula, e min For accuracy requirements, fit(gbest) t ) represents the objective function value of the globally optimal particle gbest in generation t.

4. The parameter identification method for a seven-parameter model of a photovoltaic array dual diode according to claim 1, characterized in that, Based on the extracted ideality factors and series equivalent resistance of the two diodes, the analytical method for solving the photogenerated current, parallel equivalent resistance, and reverse saturation current of the two diodes in the seven-parameter model of a photovoltaic array dual-diode includes: The remaining four parameter values ​​I of the photovoltaic array dual-diode model are solved using the following formula. ph R p I o1 and I o2 ; In the formula, D=I m V oc -V m I sc (14) In the formula, A i B i C and D are variables in the substitution formula, I ph For photocurrent, I o1 and I o2 R represents the reverse saturation current of the two diodes. p It represents the parallel equivalent resistance.

Images