A photovoltaic optimizer control system

Abstract

The application discloses a photovoltaic optimizer control system, comprising a data acquisition module, a multi-peak detection and modeling module, a control algorithm module, an execution module, a hardware coordination module and a system monitoring module. The application, by real-time monitoring and dynamic adjustment of the output impedance of the photovoltaic module, makes it always keep the best matching state with the load impedance, significantly improves the energy transmission efficiency of the photovoltaic system, reduces energy loss, and performs efficient energy transmission. Through the optimization algorithm, the application has fast response capability and can respond to the change of the output impedance of the photovoltaic module in real time, ensures that the system always runs in the best state under dynamic conditions, and at the same time, the application can be applied to different types of photovoltaic modules and complex operating environments, and can effectively respond to the influence of various factors such as light intensity change, temperature change, shadow shielding and the like on the output characteristics of the photovoltaic module.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic power generation technology, specifically to a photovoltaic optimizer control system. Background Technology

[0002] Photovoltaic intelligent optimizers are core components of photovoltaic (PV) power generation systems, primarily used in PV power plants and distributed PV systems. With the increasing global demand for clean energy, PV power generation, as an important renewable energy source, has received widespread attention and rapid development. In recent years, PV technology has continuously advanced, with PV module conversion efficiency gradually improving and costs decreasing, leading to a continuous increase in the proportion of PV power generation in the energy mix. In PV systems, the role of PV intelligent optimizers is increasingly important. They can precisely optimize the output power of PV modules, improving the overall power generation efficiency and stability of the PV system. Simultaneously, they enable intelligent management of the PV system, providing strong support for the efficient utilization of PV power generation. The development of related technologies mainly focuses on the optimization of maximum power point tracking (MPPT) algorithms, innovative hardware design, and coordinated control with inverters and energy storage systems to adapt to the diverse needs of different application scenarios.

[0003] In existing technologies, photovoltaic optimizers often emphasize the maximum power point of a single module, neglecting whether the entire string of modules operates at its maximum power point. This approach has the following problems in practical applications:

[0004] Limited power generation efficiency: When some components are shaded or otherwise mismatched, traditional optimizers can only guarantee the maximum power output of a single component, but cannot guarantee the overall maximum power output of the entire string of components, resulting in limited power generation efficiency;

[0005] Energy loss: Because maximum power point tracking of the entire string of modules cannot be achieved, the system will experience energy loss when some modules are shaded or mismatched, which reduces the power generation of the entire photovoltaic system.

[0006] Insufficient adaptability: The existing optimizer is not adaptable enough to complex environmental conditions and cannot quickly respond to changes in light intensity and temperature, causing the system to be unable to always operate in the optimal state. Summary of the Invention

[0007] The purpose of this invention is to provide a photovoltaic optimizer control system to solve the problems of photovoltaic optimizers in the background art that often emphasize the maximum power point of a single component and ignore whether the entire string of components is operating at its maximum power point. Such technical solutions have problems such as limited power generation efficiency, energy loss and insufficient adaptability in practical applications.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a photovoltaic optimizer control system, comprising a data acquisition module, a multi-peak detection and modeling module, a control algorithm module, an execution module, a hardware coordination module, and a system monitoring module. The data acquisition module is used to collect electrical parameters and environmental data of photovoltaic modules / arrays in real time, providing accurate input for the control algorithm. The multi-peak detection and modeling module is used to identify multi-peak output characteristics caused by shading and construct an accurate multi-peak model of the photovoltaic array. The control algorithm module is used to switch control strategies according to single-peak / multi-peak scenarios to achieve maximum power point tracking and dynamic impedance matching. The execution module is used to adjust hardware parameters according to control algorithm instructions to achieve power optimization and fault response. The hardware coordination module is used to enhance system reliability and handle module failures and extreme operating conditions. The system monitoring module is used to monitor the system status in real time and optimize algorithm parameters using historical data.

[0009] Preferably, the data acquisition module specifically includes a multi-dimensional sensor module, a hardware-accelerated acquisition module, and an anti-interference module;

[0010] The multi-dimensional sensor consists of a voltage sensor, a current sensor, and an environmental sensor. The environmental sensor integrates light intensity and temperature sensors and is deployed on the surface of the photovoltaic module.

[0011] The hardware-accelerated acquisition module uses an FPGA parallel processing unit;

[0012] The anti-interference module uses sensor circuits equipped with electromagnetic shielding, and data transmission uses differential signals.

[0013] Preferably, the multi-peak detection and modeling module includes a five-parameter model extension module and a multi-peak feature detection module;

[0014] The five-parameter model extension module adopts a traditional photovoltaic cell five-parameter model, introducing an equivalent circuit of a bypass diode. The modeling formula is as follows:

[0015] ;

[0016] Where I is the output current of the photovoltaic array, and n is the number of photovoltaic cell sub-modules. Let i be the photogenerated current of the i-th submodule. Let be the reverse saturation current of the i-th submodule, and q be the electron charge. Let be the terminal voltage of the i-th submodule. Where is the series resistance, c is the diode ideality factor, k is the Boltzmann constant, T is the photovoltaic cell temperature, and m is the number of bypass diodes. Let be the leakage current of the j-th bypass diode;

[0017] Multi-peak feature detection includes DFT spectral analysis and a sliding window algorithm. DFT spectral analysis is used to identify multi-peak scenes with a spacing greater than 5V, and the sliding window algorithm uses the second derivative... Locate the peak point and determine the multi-peak region, where P is the output power of the photovoltaic array, U is the output voltage of the photovoltaic array, and d is the differential operator.

[0018] Preferably, the control algorithm module includes a single-peak module, a multi-peak module, and an impedance matching adjustment module;

[0019] The single-peak module employs a reinforcement learning strategy, specifically: Q-learning optimizes the duty cycle of the DC-DC converter, with the objective function being impedance matching degree. The tracking accuracy is ≥99.5%, where J is the impedance matching degree. The output impedance of the photovoltaic module. For load impedance, The rated impedance is then used to regulate the energy storage charging and discharging through the PPO algorithm.

[0020] Preferably, the multi-peak module employs enhanced particle swarm optimization and impedance matching adjustment. Enhanced particle swarm optimization includes a two-stage search and adaptive population adjustment. Specifically, the two-stage search involves dynamic inertia weighting for the first 50 iterations of full-area exploration. The contraction factor K = 0.729, and the particles cover the entire voltage range, where w is the inertial weight and t is the current iteration number. To maximize the number of iterations, after 50 iterations, local exploitation with gradient-guided terms is employed. ;

[0021] Impedance matching adjustment uses single-peak 50-particle and multi-peak 100-particle methods to achieve... .

[0022] Preferably, the execution module consists of a DC-DC converter and a bypass switch. The DC-DC converter uses closed-loop PWM modulation with an adaptive duty cycle adjustment step size of 0.1% to 5%. The bypass switch component exhibits low degradation. Or the light intensity is less than 200W / m 2 Automatic removal at that time.

[0023] Preferably, the hardware collaboration module includes a degradation assessment and impedance equalization circuit, and the degradation assessment formula is as follows:

[0024] ;

[0025] in, Let x be the degradation degree of the i-th component. i Let x0 be the real-time parameter of the i-th component, x0 be the parameter value of the component in its normal state, and x1 be the critical value of the component parameter. An early warning is triggered when the value is greater than 0.8, and automatic bypass is activated when the value is greater than 0.9.

[0026] The impedance circuit specifically involves deploying an impedance equalizer in a series array and adjusting the voltage division of each component through an adjustable resistor network.

[0027] Preferably, the system monitoring module consists of a real-time data storage module, a remote monitoring interface module, and a self-optimization module;

[0028] The real-time data storage module uses data such as storage power curves, particle swarm iteration trajectories, and component degradation to perform 24 / 7 operation analysis.

[0029] The remote monitoring interface module uses the Modbus / TCP protocol to upload status data;

[0030] The self-optimization module is used to periodically train EPSO parameters using historical data, and the algorithm performance improves with running time in multi-peak scenarios.

[0031] Compared with the prior art, the beneficial effects of the present invention are:

[0032] In this invention, by real-time monitoring and dynamic adjustment of the output impedance of photovoltaic modules, the system maintains optimal matching with the load impedance, significantly improving the energy transmission efficiency of the photovoltaic system, reducing energy loss, and achieving efficient energy transmission. Through optimized algorithms, the system possesses rapid response capabilities, enabling it to respond in real time to changes in the output impedance of photovoltaic modules, ensuring that the system always operates in optimal condition under dynamic conditions. Furthermore, this invention is applicable to different types of photovoltaic modules and complex operating environments, effectively addressing the impact of various factors such as changes in light intensity, temperature, and shading on the output characteristics of photovoltaic modules. Attached Figure Description

[0033] Figure 1 This is a flowchart of a photovoltaic optimizer control system according to the present invention;

[0034] Figure 2 This is a system diagram of a photovoltaic optimizer control system according to the present invention.

[0035] In the picture:

[0036] 1. Data acquisition module; 2. Multi-peak detection and modeling module; 3. Control algorithm module; 4. Execution module; 5. Hardware coordination module; 6. System monitoring module. Detailed Implementation

[0037] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0038] Example: Refer to Figures 1-2 As shown: A photovoltaic optimizer control system includes a data acquisition module 1, a multi-peak detection and modeling module 2, a control algorithm module 3, an execution module 4, a hardware coordination module 5, and a system monitoring module 6.

[0039] The data acquisition module 1 is used to collect electrical parameters and environmental data of photovoltaic modules / arrays in real time, providing accurate input for the control algorithm. Specifically, it includes a multi-dimensional sensor module, a hardware-accelerated acquisition module, and an anti-interference module.

[0040] The multi-dimensional sensor consists of a voltage sensor, a current sensor, and an environmental sensor. The environmental sensor integrates light intensity and temperature sensors and is deployed on the surface of the photovoltaic module.

[0041] The hardware-accelerated acquisition module uses an FPGA parallel processing unit to acquire 16 channels of component data simultaneously, with a data synchronization error of <1μs, meeting the high-frequency detection requirements in multi-peak scenarios.

[0042] The anti-interference module uses sensor circuits equipped with electromagnetic shielding, and data transmission uses differential signals to reduce the impact of noise on acquisition accuracy.

[0043] The multi-peak detection and modeling module 2 is used to identify the multi-peak output characteristics caused by shading and to build an accurate multi-peak model of the photovoltaic array, including a five-parameter model extension module and a multi-peak feature detection module.

[0044] The five-parameter model extension module adopts a traditional photovoltaic cell five-parameter model, introducing an equivalent circuit of a bypass diode. The modeling formula is as follows:

[0045] ;

[0046] Where I is the output current of the photovoltaic array, and n is the number of photovoltaic cell sub-modules. Let i be the photogenerated current of the i-th submodule. Let be the reverse saturation current of the i-th submodule, and q be the electron charge. Let be the terminal voltage of the i-th submodule. Where is the series resistance, c is the diode ideality factor, k is the Boltzmann constant, T is the photovoltaic cell temperature, and m is the number of bypass diodes. Let be the leakage current of the j-th bypass diode;

[0047] Multi-peak feature detection includes DFT spectral analysis and a sliding window algorithm. DFT spectral analysis is used to identify multi-peak scenes with a spacing greater than 5V, and the sliding window algorithm uses the second derivative... Locate the peak point and determine the multi-peak region, where P is the output power of the photovoltaic array, U is the output voltage of the photovoltaic array, and d is the differential operator.

[0048] The control algorithm module 3 is used to switch control strategies according to single-peak / multi-peak scenarios to achieve maximum power point tracking and dynamic impedance matching, including a single-peak module, a multi-peak module and an impedance matching adjustment module;

[0049] The single-peak module employs a reinforcement learning strategy, specifically: Q-learning optimizes the duty cycle of the DC-DC converter, with the objective function being impedance matching degree. The tracking accuracy is ≥99.5%, where J is the impedance matching degree. The output impedance of the photovoltaic module. For load impedance, The rated impedance is then used to adjust the energy storage charging and discharging through the PPO algorithm;

[0050] The multi-peak module employs enhanced particle swarm optimization and impedance matching adjustment. Enhanced particle swarm optimization includes a two-stage search and adaptive population adjustment. Specifically, the two-stage search involves dynamic inertia weighting for the first 50 iterations of the full-area exploration. The contraction factor K = 0.729, and the particles cover the entire voltage range, where w is the inertial weight and t is the current iteration number. To maximize the number of iterations, after 50 iterations, local exploitation with gradient-guided terms is employed. The convergence speed is improved by 40%;

[0051] Impedance matching adjustment uses single-peak 50-particle and multi-peak 100-particle methods to achieve... .

[0052] Execution module 4 is used to adjust hardware parameters according to control algorithm instructions to achieve power optimization and fault response. It consists of a DC-DC converter and a bypass switch. The DC-DC converter adopts closed-loop PWM modulation with an adaptive duty cycle adjustment step size of 0.1% to 5%. The bypass switch component deteriorates... Or the light intensity is less than 200W / m 2 Automatically cuts off in time, with a response time of less than 200ms.

[0053] Hardware coordinating module 5 is used to enhance system reliability and handle component failures and extreme operating conditions, including degradation assessment and impedance balancing circuits. The degradation assessment formula is as follows:

[0054] ;

[0055] in, Let x be the degradation degree of the i-th component. i Let x0 be the real-time parameter of the i-th component, x0 be the parameter value of the component in its normal state, and x1 be the critical value of the component parameter. An early warning is triggered when the value is greater than 0.8, and automatic bypass is activated when the value is greater than 0.9.

[0056] The impedance circuit specifically involves deploying an impedance equalizer in a series array and adjusting the voltage division of each component through an adjustable resistor network to ensure that the voltage fluctuation of the shaded component is ≤15%.

[0057] The system monitoring module 6 is used to monitor the system status in real time and optimize algorithm parameters using historical data. It consists of a real-time data storage module, a remote monitoring interface module, and a self-optimization module.

[0058] The real-time data storage module uses data such as storage power curves, particle swarm iteration trajectories, and component degradation to perform 24 / 7 operation analysis.

[0059] The remote monitoring interface module uses the Modbus / TCP protocol to upload status data, and the fault warning response time is less than 10 seconds.

[0060] The self-optimization module is used to periodically train EPSO parameters using historical data. In multi-peak scenarios, the algorithm performance improves by 10% to 15% over time.

[0061] In this invention, by real-time monitoring and dynamic adjustment of the output impedance of photovoltaic modules, the system maintains optimal matching with the load impedance, significantly improving the energy transmission efficiency of the photovoltaic system, reducing energy loss, and achieving efficient energy transmission. Through optimized algorithms, the system possesses rapid response capabilities, enabling it to respond in real time to changes in the output impedance of photovoltaic modules, ensuring that the system always operates in optimal condition under dynamic conditions. Furthermore, this invention is applicable to different types of photovoltaic modules and complex operating environments, effectively addressing the impact of various factors such as changes in light intensity, temperature, and shading on the output characteristics of photovoltaic modules.

[0062] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A photovoltaic optimizer control system, characterized in that: The system includes a data acquisition module (1), a multi-peak detection and modeling module (2), a control algorithm module (3), an execution module (4), a hardware coordination module (5), and a system monitoring module (6). The data acquisition module (1) is used to collect electrical parameters and environmental data of photovoltaic modules / arrays in real time, providing accurate input for the control algorithm. The multi-peak detection and modeling module (2) is used to identify the multi-peak output characteristics caused by shading and to build an accurate multi-peak model of the photovoltaic array. The control algorithm module (3) is used to switch control strategies according to single-peak / multi-peak scenarios to achieve maximum power point tracking and impedance dynamic matching. The execution module (4) is used to adjust hardware parameters according to control algorithm instructions to achieve power optimization and fault response. The hardware coordination module (5) is used to enhance system reliability and handle component failures and extreme conditions. The system monitoring module (6) is used to monitor the system status in real time and optimize algorithm parameters using historical data. The data acquisition module (1) specifically includes a multi-dimensional sensor module, a hardware-accelerated acquisition module, and an anti-interference module; The multi-dimensional sensor consists of a voltage sensor, a current sensor, and an environmental sensor. The environmental sensor integrates light intensity and temperature sensors and is deployed on the surface of the photovoltaic module. The hardware-accelerated acquisition module uses an FPGA parallel processing unit; The anti-interference module uses sensor circuits equipped with electromagnetic shielding, and data transmission uses differential signals; The multi-peak detection and modeling module (2) includes a five-parameter model extension module and a multi-peak feature detection module; The five-parameter model extension module adopts a traditional photovoltaic cell five-parameter model, introducing an equivalent circuit of a bypass diode. The modeling formula is as follows: ； Where I is the output current of the photovoltaic array, and n is the number of photovoltaic cell sub-modules. Let i be the photogenerated current of the i-th submodule. Let be the reverse saturation current of the i-th submodule, and q be the electron charge. Let be the terminal voltage of the i-th submodule. Where is the series resistance, c is the diode ideality factor, k is the Boltzmann constant, T is the photovoltaic cell temperature, and m is the number of bypass diodes. Let be the leakage current of the j-th bypass diode; Multi-peak feature detection includes DFT spectral analysis and a sliding window algorithm. DFT spectral analysis is used to identify multi-peak scenes with a spacing greater than 5V, and the sliding window algorithm uses the second derivative... Locate the peak point and determine the multi-peak region, where P is the output power of the photovoltaic array, U is the output voltage of the photovoltaic array, and d is the differential operator; The control algorithm module (3) includes a single-peak module, a multi-peak module, and an impedance matching adjustment module; The single-peak module employs a reinforcement learning strategy, specifically: Q-learning optimizes the duty cycle of the DC-DC converter, with the objective function being impedance matching degree. The tracking accuracy is ≥99.5%, where J is the impedance matching degree. The output impedance of the photovoltaic module. For load impedance, The rated impedance is then used to adjust the energy storage charging and discharging through the PPO algorithm; The multi-peak module employs enhanced particle swarm optimization and impedance matching adjustment. Enhanced particle swarm optimization includes a two-stage search and adaptive population adjustment. Specifically, the two-stage search involves dynamic inertia weighting for the first 50 iterations of the full-area exploration. The contraction factor K = 0.729, and the particles cover the entire voltage range, where w is the inertial weight and t is the current iteration number. To maximize the number of iterations, after 50 iterations, local exploitation with gradient-guided terms is employed. ; Impedance matching adjustment uses single-peak 50-particle and multi-peak 100-particle methods to achieve... ; The execution module (4) consists of a DC-DC converter and a bypass switch. The DC-DC converter adopts closed-loop PWM modulation, and the duty cycle adjustment step size is adaptive from 0.1% to 5%. The bypass switch component deteriorates. Or the light intensity is less than 200W / m 2 Automatic removal at that time; The hardware coordinating module (5) includes a degradation assessment and impedance equalization circuit. The degradation assessment formula is as follows: ； in, Let x be the degradation degree of the i-th component. i Let x0 be the real-time parameter of the i-th component, x0 be the parameter value of the component in its normal state, and x1 be the critical value of the component parameter. An early warning is triggered when the value is greater than 0.8, and automatic bypass is activated when the value is greater than 0.

9. The impedance circuit specifically involves deploying an impedance equalizer in a series array and adjusting the voltage division of each component through an adjustable resistor network.

2. The photovoltaic optimizer control system according to claim 1, characterized in that: The system monitoring module (6) consists of a real-time data storage module, a remote monitoring interface module, and a self-optimization module; The real-time data storage module uses storage power curves, particle swarm iteration trajectories, and component degradation data to perform 24 / 7 operation analysis. The remote monitoring interface module uses the Modbus / TCP protocol to upload status data; The self-optimization module is used to periodically train EPSO parameters using historical data, and the algorithm performance improves with running time in multi-peak scenarios.

Images