Photovoltaic MPPT strategy based on IPO-IWCA algorithm

Abstract

Based on IPamp; a photovoltaic MPPT strategy based on an O-IWCA algorithm comprises the following steps: carrying out mathematical modeling on a photovoltaic module, and analyzing the output characteristics of the photovoltaic module in different light irradiance modes; identifying partial shading conditions of the photovoltaic module; if the condition is judged to be a uniform irradiation condition, adopting an improved perturbation and observation method IPamp; tracking the maximum power point through an O algorithm; if the condition is judged to be the partial shading condition, tracking the maximum power point by adopting an improved water circulation optimization algorithm IWCA; and monitoring the power change condition, if power fluctuation occurs, judging whether load change occurs, and if so, tracking the maximum power point by adopting a load line strategy. According to the strategy, various complex working conditions are effectively identified, and technical foundation support is provided for algorithm tracking characteristics; then improving an optimization mechanism of a water circulation optimization algorithm WCA to enhance the global search capability, and combining IPamp; according to the local accurate tracking characteristic of the O algorithm, collaborative optimization of MPPT performance of the photovoltaic system under different environment working conditions is realized.

Description

Technical Field

[0001] This invention relates to the field of MPPT technology for photovoltaic systems, and specifically to a photovoltaic MPPT strategy based on the IP&O-IWCA algorithm. Background Technology

[0002] Photovoltaic power generation has become a core development direction in the new energy field due to its clean and renewable characteristics. However, the output characteristics of photovoltaic arrays are significantly affected by factors such as light intensity, ambient temperature, and shading, and their output PV curves exhibit nonlinear single-peak or multi-peak characteristics, achieving maximum energy utilization only at the maximum power point. Therefore, maximum power point tracking (MPPT) technology has become the key to the efficient operation of photovoltaic systems, directly determining the power generation efficiency and economic benefits of photovoltaic power plants.

[0003] Current mainstream MPPT algorithms can be divided into two categories: traditional algorithms and intelligent optimization algorithms. Traditional algorithms, such as the Perturbation and Observation (P&O) method and the Incremental Conductance (INC) method, have the advantages of simple structure and easy implementation, but they suffer from drawbacks such as low tracking accuracy, steady-state oscillation, and slow dynamic response. In particular, they are prone to getting trapped in local maximum power points under complex shading conditions, resulting in severe energy loss. Intelligent optimization algorithms (such as particle swarm optimization and genetic algorithms) can improve the global optimization capability in multi-peak environments, but they generally suffer from high computational complexity, slow convergence speed, and high hardware implementation costs, making it difficult to meet the real-time and economic requirements of photovoltaic systems.

[0004] In recent years, the Water Cycle Optimization (WCA) algorithm has shown application potential in the MPPT field due to its simple optimization mechanism and fast convergence speed. However, standard WCA suffers from insufficient convergence accuracy in the later stages and is prone to premature convergence. The improved Propagation and Omission (IP&O) algorithm improves steady-state performance by optimizing the perturbation step size, but it still relies on the initial optimization direction and lacks robustness in multi-peak scenarios. Based on this, existing MPPT technology still faces the technical bottleneck of balancing global optimization capability and real-time tracking efficiency under complex working conditions. Summary of the Invention

[0005] To address the limitations of existing technologies in effectively identifying solar irradiance patterns and the impact of load variations on photovoltaic (PV) module output power, this invention proposes a PV MPPT strategy based on the IP&O-IWCA algorithm. This strategy effectively identifies various complex operating conditions, providing a fundamental technical basis for the algorithm's tracking characteristics. Subsequently, the optimization mechanism of the WCA water cycle optimization algorithm is improved to enhance global search capabilities. Combined with the locally precise tracking characteristics of the IP&O algorithm, this achieves coordinated optimization of the PV system's MPPT performance under different environmental conditions.

[0006] The technical solution adopted in this invention is as follows: The photovoltaic MPPT strategy based on the IP&O-IWCA algorithm includes the following steps: Step 1: Perform mathematical modeling of photovoltaic modules and analyze the output characteristics of photovoltaic modules under different light irradiance modes; Step 2: Identify partial shading conditions for photovoltaic modules: If the condition is determined to be uniform irradiation, the improved perturbation and observation method (IP&O) algorithm is used to track the maximum power point. If the condition is determined to be partially shaded, the improved water cycle optimization algorithm IWCA is used to track the maximum power point. Step 3: Monitor power changes. If power fluctuations occur, determine if a load change has occurred. If so, a load line strategy is then used to track the maximum power point; Otherwise, return to step two.

[0007] In step one, the photovoltaic module is composed of solar cells connected in series and parallel. The solar cells can be equivalently represented by a single diode model, and its equivalent circuit includes a current source, a diode, and parallel and series resistors. The output current of the solar cell can be expressed by the following formula (1): (1); In equation (1), Photocurrent; This is the reverse saturation current of the diode; A The ideal factor of the diode; k Boltzmann's constant; T The surface temperature of the photovoltaic cell; V This refers to the output voltage. This is the equivalent series resistance; It is the equivalent parallel resistance.

[0008] The output characteristics of photovoltaic modules under different irradiance modes are as follows: If the solar irradiance is constant, the IV and PV curves of a photovoltaic module have a unique extreme point, i.e., the global maximum power point. Under shading conditions, the output power curves of a photovoltaic module have multiple local maximum power points, resulting in multi-peak characteristics in the IV and PV curves. Specifically: When photovoltaic (PV) modules are partially shaded by trees, buildings, dust, etc., the shaded PV cells transform from power generation units into load resistors, consuming the energy of other PV cells and generating hot spots. The power generation characteristics of PV cells in shaded and unshaded areas differ significantly, resulting in multiple local current and power peaks in the IV curve or PV curve.

[0009] In step two, to effectively distinguish between uniform irradiance (UIC) and partial shading (PSC) conditions, it is necessary to analyze the critical value for the transition between UIC and PSC. The PV curve of UIC exhibits a single power peak, and the voltage at the maximum power point... V M and current I M It can be calculated using formulas (2) and (3) respectively: (2); (3); In the above formula, , These are the maximum voltage and current values ​​under standard test conditions, respectively. This refers to the short-circuit current under standard test conditions. , These are the temperature coefficients of short-circuit current and open-circuit voltage, respectively. Measured temperature T Temperature under standard test conditions T STC The difference, T The temperature is kept constant at 25℃. This represents the short-circuit current.

[0010] Under uniform irradiation (UIC) conditions, the voltage at the maximum power point is directly obtained from the PV curve. V p and power P p The results obtained by formulas (2) and (3) V M and P M They will be approximately equal.

[0011] In partial shading (PSC) scenarios, the PV curve exhibits multiple power peaks. Obviously, due to the existence of local peaks, the maximum power point calculated according to formulas (2) and (3) is at a different value. V M and P M The peak voltage and power values ​​measured from the actual PV curve may not be close to those obtained directly from the curve. Therefore, under partial shading (PSC) conditions, the values ​​measured from the PV curve may differ. V p and P p The results obtained by formulas (2) and (3) V M and PM Value, and power difference and voltage difference All values ​​are significantly different from those under the uniform irradiance (UIC) condition. Therefore, when the condition of equation (4) is met, the mode is the uniform irradiance (UIC) condition; otherwise, it is the partial shading (PSC) condition. (4).

[0012] In step two, the traditional P&O algorithm is improved: the P&O algorithm compares the voltage and power values ​​of the photovoltaic module in two consecutive iterations, and adjusts the duty cycle with a perturbation step size. Adjustments can be made. The choice of perturbation step size is crucial; a larger one... It can speed up tracking, but can lead to steady-state oscillations; smaller While reducing oscillations, it also decreases tracking speed. Therefore, this invention proposes an adaptive adjustment... To improve the tracking performance of the algorithm. At the start of the tracking process, the initial duty cycle of the P&O algorithm is set to 0.8, and the measured short-circuit current is substituted into formula (2) and formula (3) to calculate. V M , I M as well as P M Then calculate according to formula (5) And thereby reduce the duty cycle, when As the voltage approaches zero, the algorithm records the power and voltage at this point as follows: P p and V p This indicates that the operating point has reached the local maximum power point or the global maximum power point.

[0013] (5); In equation (5), The minimum step size to add for each perturbation; This is a scaling factor, with a value range of 1%-2%. I pv and V pv These represent the current and voltage of the photovoltaic module during the current iteration; Next, the irradiation conditions are identified according to formula (4): If the conditions are partially shaded, IWCA is used to search for the global maximum power point; specifically, the following steps are included: Step 1: Set photovoltaic system parameters: Measure the open-circuit voltage of the photovoltaic modules. V oc and minimum operating voltageV min Set the initial parameters for the WCA algorithm, including population size. N Maximum number of iterations t max Evaporation coefficient α Generate the initial population: according to the formula V i(0) = V min + rand ·[ V oc - V min ]generate N candidate voltages V 1(0) , V 2(0) , …, V N(0) ,in, rand Use a random number between [0,1] to ensure that each voltage is within [...]. V min , V oc ]Inside.

[0014] Step 2: Sample each candidate voltage using a voltage / current sensor. V i(0) Corresponding output current I i(0) ; Calculate the fitness (output power) of each water droplet: P i(0) = V i(0) · I i(0) .

[0015] Step 3: Press P i(0) Sort the water droplets in descending order to determine the initial global optimal voltage. V gb(0) =argmax{ P i(0)}, where argmax() is the function to find the maximum value; according to the formula Number of rivers ,in rand () is the rounding function, representing the rounding value before and after the specified number of elements. S Each water droplet represents a local optimal voltage. V lb,1(0) , …, V lb,S(0) .

[0016] Step 4: For the remaining N -S Each ordinary water droplet is assigned a nearest guide (river or main stream); the position of each ordinary water droplet is updated according to the surface runoff formula, corresponding to the candidate voltage: V i(1) = V i(0) + rand ·[ V leader(0) - V i(0) ],in, rand A random number within the range [0,1] V leader(0) This represents the "high-quality reference voltage" (from the river or main channel) assigned to the "low-power ordinary water droplet" in iteration 0; the updated voltage is limited to [ V min , V oc ]Inside.

[0017] Step 5: Calculate the evaporation probability of each river. P evap,j(0) ,like P evap,j(0) > α The river is marked as "evaporating"; new water droplets are generated according to the rainfall formula to replenish the evaporated river and ensure that the population size remains at [a certain level]. N ; Sample the current corresponding to the new water droplet I new(1) Calculate fitness P new,(1) = V new,(1) · I new(1) .

[0018] Step 7: Compare the fitness values ​​of all water droplets (updated ordinary water droplets + new water droplets) in the first iteration; the fitness value refers to the output power of the photovoltaic module, which is consistent with the power = voltage × current in Step 2; Update global optimal voltage V gb(1) If the power of a new water droplet exceeds the global maximum power. P gb(0) ,but V gb(1) = V new,gb(1) Otherwise keep V gb(0) .

[0019] Step 7: Check if the termination condition is met; the number of iterations has been reached. t maxOr power convergence; if not, let the number of iterations... t = t +1, repeat Steps 2 through 6 to continue iterating; if satisfied, output the current globally optimal voltage. V gb(t) The photovoltaic output voltage is stabilized at [value] using a DC-DC converter. V gb(t) The corresponding power is the maximum power point of the photovoltaic module.

[0020] Otherwise, continue iterative calculations using the P&O algorithm near the maximum power point; specifically, this includes the following steps: The perturbation and observation (P&O) method gradually approaches the maximum power point through iterative cycles of "small perturbation of the photovoltaic operating point → observation of output power changes → adjustment of perturbation direction," and includes the following steps: Step 1: Set the initial operating voltage V 0; Set the initial perturbation step size Set the initial disturbance direction sign 1 = +1, where, sign 1 indicates the direction indicator of the first voltage disturbance; a value of +1 indicates a boost, and a value of -1 indicates a buck; sample the initial voltage. V 0 and current I 0, calculate initial power P 0= V 0· I 0, P 0 represents the initial power; Step 2: Calculate the voltage after the first disturbance based on the initial disturbance direction: V 1= V 0+ sign 1· , V 1 represents the voltage after the first disturbance; Step 3: After the voltage stabilizes (to avoid transient interference), sample the voltage after the first disturbance. V 1 and current I 1; Step 4: Calculate power and power change. P 1= V 1· I 1, = P 1- P 0; I 1 represents the current after the first disturbance; This represents the change in power after the first disturbance; Step 5: According to The sign of the second disturbance is used to determine its direction: if >0: Maintain direction, sign 2= sign 1 = +1; if <0: Reverse direction, sign 2 = -1; if =0: Keep V 1. Remain unchanged and do not disturb for the time being; sign 2 indicates the direction indicator of the second voltage disturbance; Step 6: Iterate continuously, and calculate the second disturbance voltage. V 2= V 1+ sign 2. Repeat steps 3 through 5 until the power convergence condition is met, i.e., the power level stabilizes near the maximum photovoltaic power point.

[0021] If the condition is uniform irradiation, the improved perturbation-observation (IP&O) algorithm will cause the operating point to oscillate between the left and right sides of the maximum power point within the region around the maximum power point. If this oscillation satisfies the steady-state oscillation condition shown in formula (6), the improved perturbation-observation (IP&O) algorithm will terminate, and the duty cycle at the maximum power point will be stored and output. D MPP and power P MPP Otherwise, iterative execution needs to continue until the termination condition is met.

[0022] (6); In equation (6), It is a symbolic function.

[0023] In step two, the water cycle optimization algorithm IWCA is improved to overcome the limitations of traditional algorithms. The specific steps are as follows: (1) Raindrop initialization: IWCA first initializes four raindrops uniformly within a large search space, representing the duty cycles of the photovoltaic inverter. After initialization, it calculates the output power corresponding to each duty cycle by measuring current and voltage, thereby evaluating the objective function for each duty cycle. Subsequently, the output power is sorted in descending order, and the highest power value is selected. P best The duty cycle is used as a "sea" representing the global optimal solution. D sea The two raindrops with the next lowest power values ​​were designated as "rivers". D riv And "Stream 1" D str1 To improve the algorithm's convergence speed, the remaining "Stream 2"... D str2 Determined by formula (7): (7); In equation (7), D weak The duty cycle corresponding to the lowest power value; r A random number between 0 and 1.

[0024] (2) Streams flow into rivers or the sea: In nature, rivers flow towards the sea, where the water level is lowest. "River" D riv And "Stream 1" D str1 The location will be determined according to formulas (8) and (9) towards the "sea". D sea Update, "Stream 2" D str2 The position is then updated to "river" according to formula (10): (8); (9); (10); In the formula, express t The position of the river at +1 iteration; express t The position of stream 1 at +1 iteration; express t The position of stream 2 at +1 iteration; express t The position of the river at the next iteration; express t The position of stream 1 in the next iteration; express t The position of stream 2 in the next iteration. C These are parameters used to balance the ability to explore and develop algorithms. C If the value is too large to enhance the exploration capability, the water cycle optimization algorithm WCA may experience severe output oscillations, reducing the convergence speed towards the global maximum power point; if C If the value is too small to improve development performance, it is easy to cause premature convergence of the algorithm. The proposed improved algorithm uses formula (11) to... C Adaptive adjustment is performed as the number of iterations increases. t The increase, C The value gradually decreases, thus achieving a good balance between exploration and development, improving the algorithm's convergence ability and tracking accuracy towards the global maximum power point. (11); In equation (11),t max This represents the maximum number of iterations.

[0025] (3) Evaporation and precipitation processes: The Water Cycle Optimization Algorithm (WCA) determines the distance between a stream or river and the sea based on the condition shown in formula (12). If this condition is met, it indicates that the stream or river is close to the sea, at which point a rainfall process will be initiated, generating new raindrops in the search space to increase particle diversity and avoid getting trapped in a local maximum power point. (12); In equation (12), d max The critical value for determining the evaporation process of streams and rivers is a small value that approaches 0; In the improved water cycle optimization algorithm IWCA, the new raindrops are determined by formula (13): (13); In equation (13), express t The updated position of the river at +1 iteration; express t The updated position of Stream 1 at +1 iteration; m It is a constant, usually set to m =0.1; This invention uses a chaotic function, replacing the random function in traditional WCA. The chaotic function possesses similar randomness to the random function, while exhibiting superior dynamic performance and statistical properties. This invention employs a sine map as the chaotic mapping function, and its expression is as follows: (14); To make the search process closer to the "ocean," d max The following attenuation is applied: (15); In equation (15), express t The critical value for determining the evaporation process of streams and rivers is determined at the +1 iteration; express t The critical values ​​for the evaporation process of streams and rivers are determined in the next iteration; (4) Termination conditions: Traditional WCA will continue searching until the maximum number of iterations is reached. t max .like t max If the value is too small, the algorithm may terminate before reaching the maximum power point, leading to premature convergence; if tmax If the value is too large, the algorithm will continue searching even after the maximum power point has been found, resulting in unnecessary waste of computing power. The proposed improved algorithm terminates the search process by using the raindrop convergence condition defined by formula (16). When this condition is met, the duty cycle, current, voltage, and power are assigned values ​​respectively. D MPP , I MPP , V MPP and P MPP Then, if the condition of formula (16) is not met, the updated parameters corresponding to the raindrops will be applied to the photovoltaic inverter, and the tracking process will continue until the termination condition is met: (16); In equation (15), and These are the global optimal power for the current iteration and the previous iteration, respectively.

[0026] In step three, power fluctuations are monitored to determine if load changes have occurred. Changes in irradiance or load can cause multi-peak phenomena in the IV and PV curves of photovoltaic modules. A load line strategy is needed to distinguish between load changes and irradiance changes, avoiding the waste of time and computing power caused by MPPT algorithm restarts, thereby improving the overall system performance. In the IV coordinate system, the angle between the load line and the voltage axis... Calculated using the following formula: (17); To illustrate the load line strategy under varying load conditions: Assuming the photovoltaic system operates at its maximum power point under uniform irradiance (UIC) conditions, when the load suddenly decreases, the operating point will shift to the left of the maximum power point. Increase and exceed the angle corresponding to the maximum power point Conversely, when the load increases, the operating point shifts to the right of the maximum power point, causing... Less than ; Similarly, in Partial Shading Condition (PSC) mode, when the load changes, the operating point will shift to different load lines on the IV curve, resulting in... Increase or decrease. Therefore, if the rate of change of angle... A value less than 0.1 indicates that changes in irradiance require the algorithm to reinitialize the tracking process to locate the new maximum power point. When the angle change rate... A value exceeding 0.1 indicates power fluctuation caused by load changes. (Angle change rate) The calculation formula is: (18); The multi-peak phenomenon in the IV and PV curves caused by load variations requires adjusting the duty cycle using an adaptive disturbance step size until... convergence to When the load changes, Keep it unchanged, only adjust the duty cycle to make Towards To move closer. Therefore, with For reference, by following Direction based on perturbation step size right By adjusting the increments and decrements, the maximum power point can be effectively tracked. The direction of the disturbance is determined based on the following conditions: (19); However, larger This may cause the operating point to oscillate around the maximum power point, with smaller... This will reduce the convergence speed towards the maximum power point. In this invention, the following formula is used to update... This ensures that while rapidly converging to the global maximum power point, the impact of steady-state oscillations is minimized.

[0027] (20); In equation (20), P ( t )and P ( t -1) V ( t )and V ( t -1) represents the power and voltage of the photovoltaic module during two consecutive iterations; This is the minimum step size used to update the duty cycle during each perturbation, set in the range of 1%-2%. The modulation factor is crucial for achieving optimal tracking performance. When the load decreases, the operating point shifts to the right of the maximum power point, and the slope d of the PV curve changes. P / d V The value is relatively small, requiring a larger modulation factor; as the load increases, d P / d V Larger, smaller That can make Keep within a reasonable range The calculation formula is as follows: (twenty one).

[0028] The photovoltaic MPPT process based on the IP&O-IWCA algorithm includes the following steps: (1): Initialize and set four duty cycles: D 1 = 0.2 D 2 = 0.4 D 3 = 0.6 D 4 = 0.8, and define d max =1×10 -16 ; (2): Measure the short-circuit current I sc Substitute into formulas (2) and (3) to calculate. V M and I M Then calculate the maximum power. P M = V M I M ; (3): Determine the irradiation conditions according to formula (4). If it is a uniform irradiation condition, then use the IP&O algorithm and set the initial duty cycle to 0.8. During the algorithm iteration process, the duty cycle is adaptively adjusted according to formula (5). D， The peak power point is tracked, and then the termination condition of the IP&O algorithm is determined according to formula (6). If partial shading is detected, IWCA is used to track the global maximum power point; (4): Use formula (15) to determine whether the multi-peak situation of the PV curve is caused by irradiation change or load change. If it is determined to be irradiation change, re-initialize the tracking process to accurately locate the new maximum power point. Otherwise, adjust the duty cycle according to formula (19) until the operating point converges to the maximum power point.

[0029] This invention discloses a photovoltaic MPPT strategy based on the IP&O-IWCA algorithm, with the following beneficial effects: 1) The strategy proposed in this invention can effectively identify the impact of different light irradiance and load changes on the output power, providing basic technical support for algorithm optimization and iteration; 2) The improved algorithm proposed in this invention enhances the global search capability for multi-peak PV curves. In complex environments such as cloud cover and partial component shadows, it can quickly traverse the output characteristic curves of the photovoltaic array and accurately locate the global maximum power point. 3) The overall algorithm of this invention requires no additional hardware cost and can be directly deployed on a conventional DSP platform, balancing performance advantages with the economics of engineering applications. Attached Figure Description

[0030] The present invention will be further described below with reference to the accompanying drawings and examples; Figure 1 This is the photovoltaic MPPT process of the IP&O-IWCA algorithm of this invention; Figure 2(a) shows different light irradiance patterns (UIC-1 mode) of the embodiments of the present invention. Figure 2(b) shows different light irradiance patterns (UIC-2 mode) of the embodiments of the present invention. Figure 2(c) shows different light irradiance pattern curves (PSC-1 mode) of the embodiments of the present invention. Figure 2(d) shows the light irradiance pattern curves (PSC-2 mode) of different embodiments of the present invention.

[0031] Figure 3 The above are the simulation results corresponding to Example 1 in the embodiments of the present invention.

[0032] Figure 4 The simulation results are for example 2 in this embodiment of the invention.

[0033] Figure 5 The simulation results are for example 3 in this embodiment of the invention.

[0034] Figure 6(a) shows the simulation results (UIC-1 mode) corresponding to example 4 in the embodiment of the present invention. Figure 6(b) shows the simulation results (PSC-1 mode) corresponding to example 4 in the embodiment of the present invention. Detailed Implementation

[0035] The photovoltaic MPPT strategy based on the IP&O-IWCA algorithm firstly models the photovoltaic module mathematically and analyzes its output characteristics under different irradiance modes. Secondly, it identifies partial shading conditions of the photovoltaic module. If it is determined to be a uniform irradiance condition, the improved Perturbation and Observation method (IP&O) is used to track the maximum power point. If it is determined to be a partial shading condition, the improved water cycle algorithm (IWCA) is used to track the maximum power point. Then, it monitors power changes. If power fluctuations occur, it determines whether a load change has occurred and uses a load line strategy to track the maximum power point. Otherwise, it still determines the irradiance condition and uses the algorithm for iterative calculation. Finally, a hardware-in-the-loop simulation platform is established to verify the effectiveness of the proposed strategy.

[0036] Its process is as follows Figure 1 As shown, the main steps include the following: Step 1: Mathematical modeling of photovoltaic modules and analysis of output characteristics under different irradiance modes; Step 2: Identify partial shading conditions of photovoltaic modules; Step 3: Improve the traditional P&O algorithm; Step 4: Improve the traditional WCA; Step 5: Monitor power fluctuations and determine if load changes have occurred; Step 6: Photovoltaic MPPT process based on IP&O-IWCA algorithm.

[0037] Based on steps 1-6, a hardware-in-the-loop simulation experimental platform was built using MATLAB software. This platform uses a photovoltaic simulator to generate uniform and partial shading conditions for the photovoltaic modules, and a DSP microcontroller to achieve maximum power point tracking. In the experimental platform, the photovoltaic system uses a SEPIC converter, with the following parameter settings: input capacitance... C in =1000μF, coupling capacitor C c =330μF, output capacitor C out =1000μF, inductance L 1= L 2 = 470μH, switching frequency f s =20kHz, sampling time 0.05s, load resistance R L =30Ω. An LV-25P voltage sensor and an LA-25NP current sensor were used to measure the photovoltaic voltage and current, respectively. This invention employs four different photovoltaic irradiance modes, each generated by a photovoltaic simulator, including two different uniform irradiance modes, UIC-1 and UIC-2, and two different partial shading modes, PSC-1 and PSC-2. Their PV curves are shown in Figures 2(a) to 2(d). The effectiveness of the proposed method is verified using the following four calculation examples.

[0038] Example 1: First, the proposed method was tested under UIC-1 conditions, and then switched to UIC-2 conditions. The simulation results are attached. Figure 3 As shown.

[0039] For UIC-1, after the P&O algorithm started with an initial duty cycle of 0.8, it tracked the peak power of UIC-1 mode to be 121.9W within 0.25 seconds, corresponding to a voltage of 32.45V. The measured short-circuit current... I sc Substituting 2.65A into formulas (2) and (3), the result is... V M =32.11V,P M =122.15W, which satisfies formula (4), therefore it is determined to be a uniform irradiation condition. Continue to iterate the IP&O algorithm. When the proposed method detects steady-state oscillation, the maximum power operating point corresponds to i MPP =6.5°.

[0040] During the transition from UIC-1 to UIC-2, the calculation is obtained. i PV =tan -1 (2.67 / 23.3) = 6.51°, and i MPP They are approximately equal, according to formula (18). i Since 1 < 0.1, the power change was determined to be caused by irradiation variation. Therefore, the IP&O algorithm was restarted, and the first peak power of 73.1 W in UIC-2 mode was successfully tracked within 0.3 seconds, corresponding to a voltage of 29.6 V. The short-circuit current at this time was then... I sc Substituting 2.65A into formulas (2) and (3), we obtain the result. V M =28.9V P M =73.5W, which satisfies formula (4), so it is determined to be a uniform irradiation condition. Continue to iterate the IP&O algorithm, and terminate the operation when the algorithm detects steady-state oscillation.

[0041] Example 2: To evaluate the performance of the proposed method when switching from UIC-1 to PSC-1 mode, simulations were performed, and the results are as follows. Figure 4 As shown. When the irradiation condition is UIC-1, similar to Example 1, the IP&O algorithm successfully tracked the maximum power point of 121.9W in the UIC-1 mode within 0.3 seconds. i MPP =6.5°.

[0042] When the irradiation mode is switched from UIC-1 to PSC-1, due to the calculated photovoltaic angle i PV =tan -1 (2.5 / 21.5) = 6.63°, and i MPP They are approximately equal, according to formula (18). i 1 < 0.1, indicating the output power change is caused by irradiation variation; therefore, the tracking process is reinitialized. The IP&O algorithm identifies the first peak after three perturbations, because | / V| and exceeds 0.1, which does not satisfy formula (4), so it is determined to be a partial shading condition. Then, IWCA is used for global search to ensure accurate tracking of the global maximum power point. The algorithm effectively tracks the global maximum power point within 0.3 seconds, corresponding to a power of 72W.

[0043] Example 3: First, the effectiveness of the proposed method was verified in PSC-2 mode. Then, the method was switched to PSC-1 mode, and the simulation results were obtained as follows. Figure 5 As shown. The IP&O algorithm initially runs with a duty cycle of 0.8, and the first peak value it tracks corresponds to a power of 21W and a voltage of 6V. Since the tracked voltage and power are much lower than the calculated values... V M and P M The condition was determined to be partially shaded. Therefore, IWCA was used to perform a global search on the PV curve, and the global maximum power point was successfully located within 0.3 seconds. The photovoltaic angle at maximum power operation was determined. i MPP =5.58°.

[0044] When switching to PSC-1 mode i PV =tan -1 (2.58 / 26) = 5.6°, and i MPP They are approximately equal, according to formula (18). i 1 < 0.1, indicating the power change is caused by irradiation variation; therefore, the IP&O algorithm is restarted. In this scenario, due to the tracked... V M , P M The value differed significantly from the calculated value, so IWCA was activated, and the power corresponding to the global maximum power point of 72W was successfully tracked within 0.3 seconds.

[0045] Example 4: To evaluate the impact of the proposed load line strategy on the stability and efficiency of the photovoltaic system under different load conditions, simulation experiments were conducted on the proposed method in both UIC-1 and PSC-1 modes. In each test, the initial load resistance of the photovoltaic system was 30Ω. After successfully tracking the maximum power point, the load was reduced from 30Ω to 15Ω, and then increased back to 30Ω. The average value of the 10 test results was taken as the final result. The simulation results are shown in Figure 6(a) and Figure 6(b).

[0046] As can be seen from Figures 6(a) and 6(b), in both modes, when the load decreases from 30Ω to 15Ω, i PV Greater than i MPPAccording to formula (18). i If 1 exceeds 0.1, activate the load line strategy to adjust the perturbation step size ( ) to the duty cycle i MPP The direction decreases. The adaptive direction is determined according to formula (20). This allows the maximum power point tracking method to extract the maximum power with only one perturbation step. Similarly, when the load increases back to 30Ω, i 1 still exceeds 0.1, and the proposed method identifies that the power change is caused by the load change, therefore, through adaptive... To the duty cycle i MPP As the direction increases, the maximum power point can still be tracked with only one perturbation step. This verifies the effectiveness of the proposed method.

[0047] In summary, through simulations of four examples, the results show that the proposed algorithm can effectively identify different solar irradiance modes and power fluctuations caused by load changes. Furthermore, the proposed algorithm can effectively track the maximum power point, which matches the peak value of the curve given by the photovoltaic simulator, thus verifying the effectiveness of the proposed strategy.

Claims

1. Photovoltaic MPPT strategy based on IP&O-IWCA algorithm, characterized in that The method comprises the following steps: Step 1: mathematical modeling of the photovoltaic module and analysis of the output characteristics of the photovoltaic module under different light irradiance modes; Step 2: identification of the partial shading condition of the photovoltaic module: If it is judged to be a uniform irradiation condition, the improved perturb and observe (IP&O) algorithm is used to track the maximum power point; If it is judged to be a partial shading condition, the improved water cycle algorithm (IWCA) is used to track the maximum power point; Step 3: monitoring the power change, if the power fluctuates, it is judged whether the load changes: If yes, the load line strategy is used to track the maximum power point; Otherwise, return to step 2.

2. The photovoltaic MPPT strategy based on IP&O-IWCA algorithm according to claim 1, characterized in that: In step 1, the photovoltaic module is composed of solar cells connected in series and parallel, and the solar cells are equivalent to a single diode model, which includes a current source, a diode, and a parallel resistor and a series resistor; the output current of the solar cell is represented by the following formula (1): （1）； In formula (1), is the photo-generated current; is the reverse saturation current of the diode; A is the ideality factor of the diode; k is the Boltzmann constant; T is the surface temperature of the photovoltaic cell; V is the output voltage; is the equivalent series resistance; is the equivalent parallel resistance.

3. The photovoltaic MPPT strategy based on IP&O-IWCA algorithm according to claim 2, characterized in that: The output characteristics of the photovoltaic module under different light irradiance modes are as follows: if the light irradiance is constant, the I-V curve and the P-V curve of the photovoltaic module have a unique extreme point, i.e. the global maximum power point; under the shading condition, the output power curve of the photovoltaic module has multiple local maximum power points, which causes the I-V curve and the P-V curve of the photovoltaic module to have multiple peaks.

4. The photovoltaic MPPT strategy based on IP&O-IWCA algorithm according to claim 3, characterized in that: In the step two, in order to effectively distinguish the uniform irradiation condition (UIC) and the partial shading condition (PSC), it is necessary to analyze the critical value of the conversion between the uniform irradiation condition (UIC) and the partial shading condition (PSC), the P-V curve of the uniform irradiation condition (UIC) presents a single power peak value, the voltage at the maximum power point V M and the current I M The voltage and the current can be calculated by formula (2) and formula (3) respectively: （2）； （3）； In the above formulae, , are the maximum voltage and current values under standard test conditions, respectively; is the short circuit current under standard test conditions; , are the temperature coefficients of the short circuit current and open circuit voltage, respectively; is the measured temperature T the difference between the measured temperature and the standard test condition temperature T STC T is the constant value 25°C; denotes the short circuit current;​ In the uniform irradiance condition (UIC) mode, the voltage at the maximum power point directly derived from the P-V curve V p and power P p will be approximately equal to the voltage at the maximum power point calculated by equations (2) and (3) V M and P M will be approximately equal to the voltage at the maximum power point calculated by equations (2) and (3) In the Partial Shading Condition (PSC) mode, the values measured from the P-V curve V p and P p , calculated by equations (2) and (3) V M and P M , and the power difference and the voltage difference are significantly different from the values in the Uniform Irradiation Condition (UIC) mode, so when the following equation (4) is satisfied, the mode is the Uniform Irradiation Condition (UIC) mode, otherwise it is the Partial Shading Condition (PSC) mode: （4）。 5. The photovoltaic MPPT strategy based on IP&O-IWCA algorithm according to claim 4, characterized in that: The step two is improved to the traditional P&O algorithm: the tracking performance of the algorithm is improved by self-adaptive adjustment At the beginning of the tracking process, the initial duty ratio of the P&O algorithm is set to 0.8, and the measured short-circuit current is substituted into formula (2) and formula (3) to calculate V M , I M and P M , then the duty ratio is reduced according to formula (5) , and when tends to zero, the algorithm records the power and voltage at this time as P p and V p , indicating that the working point has reached the local maximum power point or the global maximum power point; （5）； In formula (5), is the minimum step added at each perturbation; is a proportional factor; I pv and V pv is the current and voltage of the photovoltaic module in the current iteration process, respectively. Then, the irradiation condition is identified according to formula (4): If it is a partial shading condition, the IWCA is used to search for the global maximum power point; otherwise, the P&O algorithm is used to iteratively operate near the maximum power point; If the irradiation condition is judged to be uniform, the improved perturb and observe (IP&O) algorithm will make the working point oscillate between the maximum power point and its vicinity. If the oscillation satisfies the steady-state oscillation condition shown in equation (6), the IP&O algorithm is terminated, and the duty cycle at the maximum power point is stored and output D MPP and power P MPP Otherwise, the iteration needs to be continued until the termination condition is reached. （6）； In formula (6), is a sign function.

6. The photovoltaic MPPT strategy based on IP&O-IWCA algorithm according to claim 5, characterized in that: The IWCA is used to search for the global maximum power point, which comprises the following steps: Step 1: Set the parameters of the photovoltaic system: measure the open-circuit voltage of the photovoltaic module V oc and the minimum operating voltage V min ; Set the initial parameters of the WCA algorithm, including the population size N , the maximum number of iterations t max , the evaporation coefficient α ; Generate the initial population: according to the formula V i(0) = V min + rand ·[ V oc - V min ] generate N candidate voltages V 1(0) , V 2(0) , …, V N(0) , where rand is a random number between [0, 1], ensuring that each voltage is within [ V min , V oc ] Step 2: Sample each candidate voltage by voltage / current sensor V i(0) Corresponding output current I i(0) ; Calculate the fitness of each water droplet, i.e. output power P i(0) = V i(0) · I i(0) ; Step 3: According to the formula P i(0) Sort the water drops in descending order, and determine the initial global optimal voltage V gb(0) = argmax{ V (i)}, where argmax() is the maximum value function; according to the formula P i(0)}, where argmax() is the maximum value function; according to the formula Divide the number of rivers , where rand () is the rounding function, and the first S water drop is the local optimal voltage V lb,1(0) , …, V lb,S(0) ; Step 4: For the remaining N - S Each ordinary water droplet selects its nearest guide; the position of the ordinary water droplet is updated according to the surface runoff formula, corresponding to the candidate voltage: V i(1) = V i(0) + rand ·[ V leader(0) - V i(0) ],in, rand A random number within the range [0,1] V leader(0) This represents the "high-quality reference voltage" assigned to the "low-power ordinary water droplet" in iteration 0; the updated voltage is limited to [ V min , V oc ]Inside; Step 5: Calculate the evaporation probability of each river P evap,j(0) , if P evap,j(0) α , mark the river as "evaporated"; generate new water droplets according to the rainfall formula to replenish the evaporated river, ensuring that the population size is still N ; sample the current corresponding to the new water droplets I new(1) , calculate the fitness P new,(1) = V new,(1) · I new(1) ;​ Step 7: Compare the fitness value of all water drops including the updated common water drops and new water drops in the first iteration; update the global optimal voltage V gb(1) If the power of the new water drop is greater than the global maximum power P gb(0) , then V gb(1) = the power of the new water drop V new,gb(1) , otherwise keep V gb(0) ; Step 7: Check if the termination condition is met, the iteration number reaches t max or power convergence; if not, let the iteration number t = t +1, repeat Step 2~ Step 6, continue iteration; if yes, output the current global optimal voltage V gb(t) , the photovoltaic output voltage is stabilized at V gb(t) , the corresponding power is the maximum power point of the photovoltaic module.

7. The photovoltaic MPPT strategy based on IP&O-IWCA algorithm according to claim 6, characterized in that: The P&O algorithm is used to iteratively operate near the maximum power point, which comprises the following steps: step1: set initial operating voltage V 0; set initial perturbation step size ; set initial perturbation direction sign 1 = +1, where, sign 1 represents the direction indicator of the first voltage perturbation, only takes +1 to represent the voltage rise, and takes -1 to represent the voltage drop; sample the initial voltage V 0 and current I 0, calculate the initial power P 0= V 0· I 0, P 0 represents the initial power; Step 2: Calculate the voltage after the first perturbation according to the initial perturbation direction: V 1= V 0+ sign 1· , V 1 represents the voltage after the first perturbation; step3: sampling the voltage after the 1st disturbance V 1 and current I 1 ; Step 4: Calculate power and power change P 1= V 1· I 1, = P 1- P 0; I 1 represents the current after the first disturbance; represents the power change after the first disturbance; Step 5: According to The sign of the second disturbance is used to determine its direction: if >0: Maintain direction, sign 2= sign 1 = +1; if <0: Reverse direction, sign 2 = -1; if =0: Keep V 1. Remain unchanged and do not disturb for the time being; sign 2 indicates the direction indicator of the second voltage disturbance; Step 6: Loop iteration, continue to track the calculation of the 2nd disturbance voltage V 2= V 1+ sign 2· ; repeat step 3~ step 5 until the power convergence condition is met, that is, stable near the maximum power point of photovoltaic.

8. The photovoltaic MPPT strategy based on IP&O-IWCA algorithm according to claim 7, characterized in that: In step 2, the improved water cycle optimization algorithm (IWCA) is used to solve the limitations of the traditional algorithm, and the specific steps are as follows: (1) Raindrop initialization: IWCA first initializes 4 raindrops in a large search space, and these raindrops represent the duty cycle of the photovoltaic inverter; after initialization, the output power corresponding to each duty cycle is calculated by measuring the current and voltage, so as to evaluate the objective function of each duty cycle; Subsequently, the output power is arranged in descending order, and the highest power value P best is taken as the "ocean" representing the global optimal solution D sea , and the two raindrops with the second highest power value are taken as "river" and "stream 1" D riv respectively D str1 ; in order to improve the convergence speed of the algorithm, the remaining "stream 2" D str2 is determined by formula (7): （7）； In formula (7), D weak is the duty cycle corresponding to the lowest power value; r is a random number between 0 and 1. (2) Stream flows to river or sea: "River" D riv and "Stream1" D str1 The position of "Stream2" D sea will be updated according to equations (8) and (9) towards "Ocean" D str2 and the position of "Stream1" will be updated according to equation (10) towards "River" （8）； （9）； （10）； In the formula, represents t the position of the river at the +1th iteration; represents t the position of the stream 1 at the +1th iteration; represents t the position of the stream 2 at the +1th iteration; represents t the position of the river at the +2th iteration; represents t the position of the stream 1 at the +2th iteration; represents t the position of the stream 2 at the +2th iteration; C is a parameter for balancing the exploration and exploitation ability of the algorithm, if C is too large to enhance the exploration ability, the water cycle optimization algorithm WCA may appear severe output oscillation, and the convergence speed to the global maximum power point is reduced. like C If the value is too small to improve development performance, it is easy to cause premature convergence of the algorithm; the proposed improved algorithm uses formula (11) to... C Adaptive adjustment is performed as the number of iterations increases. t The increase, C The value gradually decreases, thus achieving a good balance between exploration and development: （11）； In formula (11), t max is the maximum number of iterations. (3) Evaporation and rainfall process: The water cycle optimization algorithm (WCA) judges the distance between the stream or river and the sea through the condition shown in formula (12); if the condition is met, it means that the stream or river is close to the sea, and at this time, the rainfall process is started to generate new raindrops in the search space, so as to increase the particle diversity and avoid falling into the local maximum power point: （12）； In formula (12), d max To judge the critical value of the evaporation process of streams and rivers, a small value tends to 0; In the improved water cycle optimization algorithm (IWCA), the new raindrop is determined by formula (13): （13）； In formula (13), represents t updated position of river at the 1th iteration; represents t updated position of stream 1 at the 1th iteration; μ is a constant; is a chaotic function, which replaces the random function in the traditional WCA; a sine mapping is used as the chaotic mapping function, and its expression is as follows: （14）； To make the search process closer to "the sea", to d max the following attenuation is made: （15）； In formula (15), denotes t Critical value for judging the evaporation process of stream and river at the first iteration; denotes t Critical value for judging the evaporation process of stream and river at the second iteration; (4) Termination condition: The search process is terminated by the raindrop convergence condition defined by equation (16), when this condition is satisfied, the duty cycle, current, voltage and power at this time are assigned as D MPP , I MPP , V MPP and P MPP After that, if the condition of equation (16) is not satisfied, the updated parameters corresponding to the raindrop are applied to the photovoltaic inverter and the tracking process continues until the termination condition is met: （16）； In formula (15), and Goptand Gopt-1are the global optimal powers of the current iteration and the previous iteration, respectively.

9. The photovoltaic MPPT strategy based on IP&O-IWCA algorithm according to claim 8, characterized in that: In the third step, the power fluctuation is monitored and it is determined whether a load change occurs; both light irradiance change and load change can cause the photovoltaic module I-V curve and P-V curve to appear multi-peak phenomenon, a load line strategy is proposed to distinguish the load change and light irradiance change, in the I-V coordinate system, the included angle between the load line and the voltage axis is calculated by the following formula: （17）； In order to illustrate the load line strategy under the load change condition: Assuming the photovoltaic system works at the maximum power point in the uniform irradiance condition (UIC) mode, when the load suddenly decreases, the working point will move to the left side of the maximum power point, at this time is greater than the angle corresponding to the maximum power point , on the contrary, when the load increases, the working point moves to the right side of the maximum power point, so that is less than ; Similarly, when the load changes in the partial shade condition (PSC) mode, the operating point will shift to a different load line on the I-V curve, resulting in an increase or decrease; therefore, if the angle change rate is less than 0.1, it indicates that the light irradiance changes such that the algorithm needs to reinitialize the tracking process to locate the new maximum power point, and when the angle change rate exceeds 0.1, it can be judged as a power fluctuation caused by a load change; the calculation formula of the angle change rate is: （18）； The multi-peak phenomenon of I-V curve and P-V curve caused by load change needs to adjust the duty ratio by adaptive perturbation step until converges to ; When the load changes Keep it unchanged, only adjust the duty cycle to make Towards To get closer; therefore, with For reference, by following Direction based on perturbation step size right By adjusting the increments and decrements, the maximum power point can be effectively tracked; the direction of the disturbance is determined based on the following conditions: （19）； However, the larger may cause the operating point to oscillate around the maximum power point, the smaller will reduce the convergence speed to the maximum power point; by updating , it ensures that the influence of steady-state oscillation is minimized while quickly converging to the global maximum power point; （20）； In formula (20), P ( t ) and P ( t -1) are the power and voltage of the photovoltaic module in the two successive iteration processes, respectively; V ( t ) and V ( t -1) are the power and voltage of the photovoltaic module in the two successive iteration processes, respectively; is the minimum step length for updating the duty ratio at each perturbation, and is set in the range of 1%-2%; is the modulation factor, and reasonable setting of the modulation factor is the key to achieve optimal tracking performance; when the load decreases, the working point moves to the right side of the maximum power point, and the slope d P / d V of the P-V curve is relatively small, and a larger modulation factor is required at this time; when the load increases, d P / d V is larger, and a smaller can keep in a reasonable range, and the calculation formula is as follows: （21）。 10. Photovoltaic MPPT procedure based on IP&O-IWCA algorithm, characterized in that: The process comprises the following steps: (1): initialization set four duty cycles: D 1 = 0.2, D 2 = 0.4, D 3 = 0.6, D 4 = 0.8, and define d max = 1 x 10 -16 ; (2): measuring short circuit current I sc and substituting into equations (2) and (3) V M and I M and further calculating maximum power P M = V M I M ; (3): According to formula (4), the irradiation condition is judged. If it is uniform irradiation condition, the IP&O algorithm is adopted, the initial duty cycle is set to 0.8, and in the iteration process of the algorithm, the duty cycle is adjusted according to formula (5) D， The power peak point is tracked, and then the termination condition of the IP&O algorithm is judged according to formula (6); If the partial shading condition is detected, the IWCA is used to track the global maximum power point; (4): Determine whether the irradiance or the load is the cause of the multi-peak P-V curve by formula (15). If the irradiance is the cause, re-initialize the tracking process to locate the new maximum power point. Otherwise, adjust the duty cycle according to formula (19) until the operating point converges to the maximum power point.

Images