Photovoltaic power optimization system and method based on power line carrier communication cooperative control

Abstract

The invention discloses a photovoltaic power optimization system and method based on power line carrier communication cooperative control. The system comprises a ripple characteristic detection module, an intelligent scheduling controller, a communication module and a distributed MPPT execution unit. Wherein the ripple characteristic detection module collects original current signals of a photovoltaic array in a photovoltaic system in real time and processes the original current signals to obtain detection data, and the intelligent scheduling controller dynamically distributes scanning priorities of MPPT nodes according to the detection data to obtain a scheduling instruction and sends the scheduling instruction to the distributed MPPT execution unit through the communication module. And then specific power point adjustment is executed at each MPPT node. The method comprises the steps of processing an acquired original current signal of a photovoltaic array to obtain detection data, and dynamically distributing scanning priorities of MPPT nodes according to the detection data to obtain a scheduling instruction to perform power adjustment on the nodes. According to the invention, the power regulation efficiency of the photovoltaic power generation system can be optimized, and the power generation efficiency of the photovoltaic power generation system in a dynamic shadow scene is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of photovoltaic power generation, and more particularly to a photovoltaic power optimization system and method based on power line carrier wave communication cooperative control. BACKGROUND

[0002] In a conventional photovoltaic power generation system, each power optimizer independently runs an MPPT (Maximum power point tracking) algorithm, which results in the following technical bottlenecks: 1. Multi-node scanning conflict: When multiple power optimizer nodes simultaneously perform MPPT algorithm scanning, the system presents a state of random scanning. Experimental data under the IEC 62894 test standard show that, in a 5-node system, the scanning conflict probability under a shadow blocking scenario is as high as 68%, causing the system power oscillation amplitude to exceed ±15%.

[0003] 2. Communication time delay impact: Existing PLC (Power Line Communication) protocols (such as IEEE 1901.1) transmit full-amount current waveform data (bandwidth requirement > 800 kbps when the sampling rate is 1 MSPS), resulting in a central controller response delay > 50 ms, which cannot achieve real-time coordination.

[0004] 3. Ripple feature loss: Independent MPPT controllers only focus on average power values and ignore key information such as system impedance and shadow distribution contained in current ripple. Shadow blocking causes a sudden increase in ripple coefficient K (ΔI / I_avg).

[0005] In view of the technical bottlenecks of the maximum power point tracking algorithm, existing technologies have adopted various solutions, but they still have the following limitations: 1. Time slicing scheduling method: The MPPT algorithm scanning time is divided into fixed time slots, but this method cannot adapt to dynamic shadow changes, resulting in significant system efficiency loss.

[0006] 2. Existing ripple suppression circuits suppress current ripple through LC filters, but this results in a decrease in the tracking speed of the MPPT algorithm and increases hardware costs.

[0007] 3. Centralized MPPT control: The central controller directly issues PWM instructions, but it needs to transmit full-amount sensor data (bandwidth > 1 Mbps), which has low communication reliability.

[0008] 4、Multi-node independent MPPT algorithm: easy to cause power oscillation problem, each optimizer in the traditional photovoltaic power generation system independently runs the MPPT algorithm, under the IEC62894 standard test data, the probability of multiple nodes starting MPPT scanning at the same time under the shadow blocking scene is high, which leads to large amplitude of system power output fluctuation, causing grid-connected power quality deterioration; 5、PLC communication delay caused global regulation lag, the prior art needs to transmit full-amount current waveform data such as 1MSPS sampling rate, which requires a bandwidth of >800kbps, causing a response delay of the central controller of >50ms, which cannot realize real-time coordinated control under dynamic shadow scene. 6、The traditional MPPT algorithm control only focuses on the average power value, so that the ripple characteristic information is not effectively utilized, and the system impedance distribution information (the ripple coefficient K has a positive correlation with the shadow blocking area, R2=0.93) and the key role of ripple mutation on the MPPT tracking direction judgment (when the K value is >5%, the system efficiency decreases by 12.7%) hidden in the current ripple spectrum are ignored. SUMMARY

[0009] To solve the problems in the prior art, the application provides a photovoltaic power optimization system and method based on power line carrier wave communication cooperative control, which can solve the problems of low algorithm scanning efficiency and high node scanning conflict probability in the prior art under the shadow blocking scene.

[0010] The application adopts the following technical solutions.

[0011] A photovoltaic power optimization system based on power line carrier wave communication cooperative control, which comprises: a ripple characteristic detection module, an intelligent scheduling controller, a communication module and a distributed MPPT execution unit; The ripple characteristic detection module is connected with the output end of the photovoltaic array, and is used for collecting the original current signal of the photovoltaic array in the photovoltaic power generation system in real time and detecting and processing the collected original current signal to obtain a ripple coefficient and send the ripple coefficient to the intelligent scheduling controller; the intelligent scheduling controller dynamically allocates the scanning priority of each MPPT node in the photovoltaic power generation system based on the ripple coefficient to obtain a scheduling instruction; the communication module is used for sending the scheduling instruction to the distributed MPPT execution unit, and the distributed MPPT execution unit is used for executing specific power point regulation at each MPPT node according to the received scheduling instruction.

[0012] Preferably, the ripple characteristic detection module comprises a Hall sensor group, a signal conditioning unit, a data processing unit and a spectrum analysis processor. The Hall sensor group is used to collect the original current signal of the photovoltaic array output in real time. The collected original current signal is filtered of high-frequency noise by the signal conditioning unit, and then is subjected to digital processing by the data processing unit to obtain a digitized signal. The frequency spectrum analysis processor performs frequency spectrum analysis on the digitized signal to obtain an energy integral value of a characteristic frequency band, and calculates a ripple coefficient according to the energy integral value.

[0013] Preferably, the calculation formula of the ripple coefficient K value is as follows: K = (energy integral value / I_avg) x 100% Wherein, I_avg is the average current value.

[0014] Preferably, the intelligent scheduling controller dynamically allocates the MPPT scanning priority based on the ripple coefficient, specifically including: According to the ripple coefficient, a ripple coefficient change rate ΔK is calculated, and a system working mode is judged according to the ripple coefficient change rate and a set threshold value; When the ripple coefficient change rate ΔK < the set change threshold value, it indicates that the system is in a stable state, at this time, a regular polling mode is entered, and a fixed time slice is used to cyclically access each MPPT node; When the ripple coefficient change rate ΔK ≥ the set change threshold value, it indicates that the system is in a dynamic change state, at this time, an interrupt preemption mode is entered, and an emergency scan is triggered to respond to the change of the system state.

[0015] Preferably, the calculation method of the ripple coefficient change rate ΔK is specifically as follows: ΔK = |K n -K n-1 | / Δt Wherein, Kn and K n-1 are the ripple coefficients at the current time and the previous time respectively, and Δt is the calculation time interval, and the unit of Δt is ms.

[0016] Preferably, the intelligent scheduling controller dynamically allocates the MPPT scanning priority based on the ripple coefficient, and further includes: According to the historical ripple coefficient and the ripple coefficient change rate, a future ripple coefficient is predicted, a future ripple coefficient change rate is calculated according to the future ripple coefficient, and the system working mode is adjusted according to the future ripple coefficient change rate; When the predicted future ripple coefficient change rate exceeds a preset prediction change threshold value, the interrupt preemption mode is triggered in advance, a scheduling instruction of the emergency scan is generated, and the regular polling is restored after the emergency scan is completed.

[0017] Preferably, it further includes: constructing a mapping model of the ripple coefficient and the shadow blocking area, calculating the current shadow blocking area according to the ripple coefficient, and judging the node priority according to the current shadow blocking area. The mapping model of the ripple coefficient and the shadow blocking area is as follows: K=0.12*shadow blocking area+0.05 According to the mapping model and the ripple coefficient K value, the shadow blocking area of the current node is obtained, and whether the node is a high-priority node is judged according to the shadow blocking area: When the blocking area exceeds the preset blocking threshold, the MPPT node is judged as a high-priority node, and the ripple coefficient change rate is taken as the trigger basis of the emergency event.

[0018] Preferably, the scheduling instruction includes a mode identifier, a scanning priority and a time slice allocation. The mode identifier includes a regular polling mode identifier and an interrupt preemption mode identifier. The scanning priority and the time slice allocation are performed according to the system working mode, and the details are as follows: In the regular polling mode, the intelligent scheduling controller cyclically accesses each MPPT node according to a preset fixed time slice, so as to ensure the periodicity of the system operation. In the interrupt preemption mode, the intelligent scheduling controller pauses the current polling task according to the high-priority mark in the detection data, and preferentially allocates scanning resources to the high-priority node.

[0019] The application also provides a photovoltaic power optimization method based on power line carrier wave communication cooperative control, which is realized by using a photovoltaic power optimization system based on power line carrier wave communication cooperative control. Step 1: The ripple feature detection module is connected to the output end of the photovoltaic array in the photovoltaic system, real-time acquires the original current signal of the photovoltaic array and performs detection processing, obtains the ripple coefficient and sends the ripple coefficient to the intelligent scheduling controller. Step 2: The intelligent scheduling controller dynamically allocates the scanning priority of the MPPT node based on the ripple coefficient, and obtains the scheduling instruction. Step 3: The communication module sends the scheduling instruction to the distributed MPPT execution unit. Step 4: The distributed MPPT execution unit is used for executing specific power point adjustment at each MPPT node according to the received scheduling instruction.

[0020] Compared with the prior art, the application provides a photovoltaic power optimization system and method based on power line carrier wave communication cooperative control, which has at least the following beneficial effects: 1. By calculating the energy integral value of the 10Hz~100Hz characteristic frequency band in the ripple feature detection module, and normalizing it with the average current value, a ripple coefficient strongly positively correlated with the shadow blocking area is obtained, realizing the quantification of current ripple features into a calculable and transmissible collaborative control core parameter, breaking through the traditional MPPT algorithm which only relies on voltage / power judgment logic, and providing a new decision basis for system collaborative optimization.

[0021] 2. By selecting the conventional polling mode or the trigger interrupt preempt mode according to the ripple coefficient change rate and the set threshold in the intelligent scheduling controller, and combining with the historical ripple coefficient sequence for pre-judgment scheduling, the fast and active response to dynamic shadow changes is realized, the system response delay is shortened from 50ms of the traditional scheme to 8ms, and the adjustment speed of the system under dynamic working conditions is significantly improved.

[0022] 3. By quantizing, differential encoding and adaptive encoding the ripple coefficient data in the communication module, and cooperating with CRC-8 checksum and three-way handshake mechanism, the bandwidth occupation of control data is greatly reduced from more than 800kbps of the prior art to 400bps, a lightweight and highly reliable power line carrier communication architecture is built, and the global regulation lag problem caused by communication delay is solved.

[0023] 4. By establishing a mapping model of the ripple coefficient and the shadow blocking area, and using the distributed MPPT execution unit to perform priority scanning on high-priority nodes according to the instructions of the HSS intelligent scheduling controller, the collaborative control among multiple nodes is realized, and the power oscillation caused by independent scanning of each unit in the traditional distributed MPPT system is avoided, reducing the multi-node scanning conflict probability. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 is the structural block diagram of the photovoltaic power optimization system based on power line carrier communication collaborative control in the present application; Figure 2 is the flow chart of the photovoltaic power optimization method based on power line carrier communication collaborative control in the present application. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical scheme and advantages of the present application clearer, the technical scheme of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. The embodiments described in the present application are only a part of the embodiments of the present application, not all the embodiments. Based on the spirit of the present application, other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0026] In the traditional distributed MPPT system, the power point tracking process of the MPPT execution unit is as follows: each MPPT unit independently collects the voltage and current data of the photovoltaic array of the node, and a fixed algorithm is used: the traditional perturb and observe method or conduct increment method, the perturb and observe method determines the maximum power point by comparing the power change direction through a small perturbation voltage, periodic scanning is performed according to a fixed time interval, the duty cycle is adjusted each time, the power change is measured, there is no cooperation between units, and no communication with other units.

[0027] In the dynamic shadow scenario, multiple units scan at the same time, resulting in a high probability of scanning conflict, only focusing on the average power and ignoring the ripple characteristics, and being unable to respond to shadow changes in advance.

[0028] As Figure 1 shown, the application provides a photovoltaic power optimization system based on power line carrier communication cooperative control, which comprises a ripple characteristic detection module, an intelligent scheduling controller (Hierarchical Smart Scheduler, HSS), a communication module and a distributed MPPT execution unit. The ripple characteristic detection module is connected with the output end of the photovoltaic array, is used for collecting the original current signal of the photovoltaic array in the photovoltaic power generation system in real time, and detects and processes the collected original current signal to obtain a ripple coefficient and send it to the intelligent scheduling controller; the intelligent scheduling controller dynamically allocates the scanning priority of the MPPT node according to the ripple coefficient to obtain a scheduling instruction; the communication module is used for sending the scheduling instruction to the distributed MPPT execution unit, and the distributed MPPT execution unit is used for executing specific power point adjustment at each MPPT node according to the received scheduling instruction. The MPPT node in the application is each power optimizer in the photovoltaic power generation system.

[0029] The ripple characteristic detection module comprises a Hall sensor group, a signal conditioning unit, a data processing unit and a spectrum analysis processor. The Hall sensor group is used for non-contact collection of the original current signal output by the photovoltaic array, and the original current signal comprises a direct current component and a ripple component. The detection and processing of the original current signal specifically comprises: The high-speed ADC (Analog to Digital Converter) of the data processing unit is used for digital processing of the original current signal; the resolution of the high-speed ADC is 12 bits, and the sampling rate is 100 ksps. The collection time window is set to 100 ms, so as to ensure that enough ripple period information is contained.

[0030] The original current signal collected is filtered of high-frequency noise by a signal conditioning unit, and is subjected to digital processing by a data processing unit to obtain a digitized signal; The spectrum analysis processor performs spectrum analysis on the digitized signal using a fast Fourier transform algorithm, and the analysis frequency band is limited to 10Hz-100Hz. This frequency band is highly correlated with the shadow blocking characteristics of the photovoltaic system. The energy integral value of the characteristic frequency band is calculated through spectrum analysis: All frequency components in the range of 10Hz-100Hz are screened from the spectrum. This frequency band is a key frequency band that is highly correlated with the shadow blocking characteristics of the photovoltaic system, and can effectively reflect the influence of shadow blocking on current ripple.

[0031] For each frequency point screened, the square value of the spectrum amplitude (i.e. the power spectral density) is calculated, which represents the energy contribution of the frequency point.

[0032] The energy integral value is obtained by integrating the energy. In digital signal processing, the integration process is approximated by summation because the frequency is discrete. The power spectral density values of all frequency points in the frequency band of 10Hz-100Hz are added up to obtain the total energy of the frequency band as the energy integral value.

[0033] The energy integral value is normalized to obtain the ripple coefficient K value, and the calculation formula is as follows: K=(energy integral value / I_avg)×100% Where I_avg is the average current value.

[0034] Through the above process, the energy characteristics of the current ripple in the frequency band of 10Hz-100Hz can be accurately quantified. This characteristic is highly correlated with the degree of shadow blocking. After the ripple coefficient K value is obtained through the ripple characteristic detection module, it is transmitted to the intelligent scheduling controller for further control.

[0035] Further preferably, when the spectrum analysis processor performs spectrum analysis on the digitized signal, wavelet transform can be used instead of the fast Fourier transform algorithm to enhance the analysis capability of non-stationary ripples.

[0036] The fast Fourier transform algorithm is suitable for stationary signals and resource-limited scenarios, and is a low-cost and efficient solution. The wavelet transform algorithm is suitable for non-stationary signals and high-precision demand scenarios, and can more comprehensively capture the instantaneous changes of ripple characteristics, thereby improving the analysis capability of dynamic shadows and other non-stationary ripples In practical applications, the present application can also use the fast Fourier transform algorithm to quickly analyze the main frequency component of the signal in order to combine the advantages of the two algorithms, and determine whether it is necessary to further process the transient event by wavelet transform. For example, in a dynamic shadow scene, the sudden trend of the ripple coefficient K value is detected by the fast Fourier transform algorithm first, and if an anomaly is detected, wavelet transform is called for local detail analysis, as follows: The ripple coefficient change rate ΔK is the absolute difference between the K values of the adjacent two times divided by the time interval: ΔK = |K n -K n-1 | / Δt Wherein, K n and K n-1 are the ripple coefficients at the current time and the previous time, and Δt is the calculation time interval (unit: ms).

[0037] Preferably, the K value is calculated every 10 ms, so Δt = 10 ms, and the unit of the ripple coefficient change rate ΔK is % / ms.

[0038] Determine whether the ripple coefficient change rate reaches the preset threshold value to determine whether to start wavelet transform: When ΔK ≥ 0.5% / ms, it is determined that the dynamic shadow changes, and wavelet transform is automatically called for transient analysis; when ΔK < 0.5% / ms, the fast Fourier transform algorithm is continued to be used for regular analysis.

[0039] Preferably, since the K value change rate under the dynamic shadow scene is usually greater than 0.5% / ms, and the system is in a stable state, it is less than 0.5% / ms, so the threshold value is set to 0.5% / ms, thereby effectively distinguishing the system state.

[0040] Preferably, the ripple feature detection module is directly connected with the intelligent scheduling controller, and is used for transmitting the ripple coefficient.

[0041] ‌After receiving the ripple coefficient sent by the ripple feature detection module, the intelligent scheduling controller dynamically allocates the MPPT scanning priority based on the ripple coefficient. In the present application, the intelligent scheduling controller is a hierarchical intelligent scheduler, which specifically refers to an intelligent controller with dynamic priority allocation capability in a photovoltaic system. The intelligent scheduling controller can dynamically adjust the MPPT scanning strategy based on the ripple coefficient change rate.

[0042] The core mechanism includes two working modes: a regular polling mode and an interrupt preemption mode.

[0043] Among them, the regular polling mode adopts a fixed time slice to cyclically access each MPPT node, ensuring the periodicity of system operation; and the interrupt preemption mode is used to cope with sudden situations, and when the ripple coefficient change rate exceeds the set threshold value, an emergency scan is triggered to quickly respond to the change of the system state.

[0044] In terms of hardware composition, the intelligent scheduling controller adopts a SoC chip with integrated multi-task processing capability as a main control unit, is equipped with a priority register group for storing state identifiers of each node, and guarantees the accuracy of scheduling timing through a clock synchronization unit.

[0045] Further preferably, the MPPT scanning priority is dynamically allocated based on the ripple coefficient, specifically including: The ripple coefficient change rate ΔK is calculated according to the ripple coefficient, and the system working mode is determined according to the ripple coefficient change rate and a set threshold value; When the ripple coefficient change rate ΔK < the set change threshold value, it indicates that the system is in a stable state, at which time the regular polling mode is entered, and the regular polling mode adopts a fixed time slice to cyclically access each MPPT node; When the ripple coefficient change rate ΔK ≥ the set change threshold value, it indicates that the system is in a dynamic change state, at which time the interrupt preemption mode is entered, and an emergency scan is triggered to respond to the change of the system state.

[0046] The ripple coefficient change rate ΔK is the absolute difference between the K values of two adjacent times divided by the time interval: ΔK = |K n -K n-1 | / Δt Wherein, K n and K n-1 are the ripple coefficients at the current time and the previous time, respectively, and Δt is the calculation time interval (unit: ms).

[0047] Preferably, the K value is calculated every 10 ms, so Δt = 10 ms, and the unit of the ripple coefficient change rate ΔK is % / ms.

[0048] Preferably, the set change threshold value is 0.5% / ms.

[0049] Further, the ripple coefficient K value is positively correlated with the shadowed area of the photovoltaic array, and the mapping model of the ripple coefficient and the shadowed area is as follows: K = 0.12 × shaded area + 0.05 According to the mapping model and the ripple coefficient K value, the current shadowed area can be calculated; According to the shadowed area, it is determined whether the node is a high-priority node: When the shaded area exceeds 20%, the node is marked as a high-priority node, and the ripple coefficient change rate ΔK is used as the trigger basis for an emergency event; Further, the ripple coefficient K value in the application as a static index of system state can reflect the current photovoltaic system shadow blocking degree, the ripple coefficient and the shadow blocking area are strongly positively correlated (K=0.12*blocking area+0.05, R²=0.93). The ripple coefficient is used for priority allocation of the conventional MPPT scanning and system state evaluation, and is a core parameter for realizing the ripple feature driven MPPT decision of the application.

[0050] The ripple coefficient change rate DK as a dynamic index of system state is used for representing the change rate of the K value in a unit time, and is used for judging whether the system is in a dynamic shadow change state.

[0051] When DK is greater than or equal to 0.5% / ms, the system is determined to be in a dynamic shadow change state, the wavelet transform is automatically started for transient feature analysis, the MPPT emergency scanning is triggered in advance, the system response delay is shortened from 15 ms to 8 ms, and the adaptability of the system to the dynamic shadow is significantly improved.

[0052] The communication module is used for sending scheduling instructions to the distributed MPPT execution unit, and guarantees the reliable transmission of the scheduling instructions, the communication module transmits the instructions based on LCPO (Lightweight Communication Protocol for Power Line, lightweight power line communication protocol); Preferably, the distributed MPPT execution unit is used for executing specific power point adjustment on each MPPT node according to the scheduling instruction, the ripple feature detection module collects the adjusted photovoltaic current ripple signal after each MPPT node executes the adjustment instruction, generates new detection data and repeats the above optimization process; The data flow feature in the application is: a closed-loop control process of "feature detection→decision generation→instruction issuing→execution feedback".

[0053] The feature of the scheduling algorithm is the judgment based on the K value change rate DK, the ripple feature detection module collects and extracts the frequency spectrum feature of the current ripple at a certain frequency, and calculates the change rate DK according to the sent adjacent two K values; the change threshold is set, and the setting optimization of the change threshold is positively correlated with the power.

[0054] When DK is detected to be greater than or equal to the set threshold, the system will pause the current polling task, and preferentially start the scanning of the high-priority node, and restore the regular polling after completing the emergency scanning; if the threshold is not reached, the scanning resource is allocated according to the preset time slice.

[0055] Further preferably, the intelligent scheduling controller can adopt the way of introducing a historical data prediction model to realize the predictive scheduling; the time slice length is dynamically adjusted through fuzzy logic, and the flexibility and adaptability of the scheduling strategy are further optimized.

[0056] The historical data prediction model includes input and output. (1) Input Historical ripple coefficient sequence: The system collects ripple coefficient K values (such as K1, K2, K3, …, Kn) at multiple time points to capture the dynamic changes of ripple characteristics.

[0057] The ripple coefficient change rate ΔK sequence is calculated based on the historical ripple coefficient sequence: The difference between adjacent two ripple coefficients (such as ΔK1=K2 K1, ΔK2=K3 K2), reflecting the mutation trend of ripple characteristics.

[0058] (2) Output Future ripple coefficient prediction value: The future ripple coefficient is predicted by the model within a certain period of time, which is used to determine whether the shadow blocking is likely to be intensified.

[0059] Future ripple coefficient change rate prediction value: The future ripple coefficient change rate is obtained based on the future ripple coefficient prediction value within a certain period of time, which is used to determine whether the adjustment of the dispatching strategy is likely to be triggered.

[0060] Further, if the predicted ΔK≥set threshold, immediately start the fuzzy logic adjustment time slice, including activating the fuzzy logic system, dynamically calculating and adjusting the time slice length of MPPT scanning according to the predicted ΔK; the system enters the interrupt preemption mode in advance, and preferentially scans the high-risk nodes; after completing the emergency scanning, according to the actual ΔK value, restore the normal polling or continue to predict the scheduling.

[0061] For example: set the threshold to 0.5% / ms, when the predicted ΔK=0.6% / ms, the time slice adjustment coefficient is 0.3 (i.e. 100ms×0.3=30ms); when the predicted ΔK=0.3% / ms, the time slice adjustment coefficient is 1.0 (i.e. 100ms×1.0=100ms). The system enters the interrupt preemption mode in advance according to the calculated time slice length, and preferentially scans the high-risk nodes.

[0062] Further, the prediction of shadow blocking intensity: Based on the positive correlation between K value and shadow blocking area, output the prediction value of future shadow blocking degree (such as blocking area ratio, blocking duration).

[0063] By collecting current ripple data under different blocking conditions, the corresponding relationship between K value and blocking area is calculated: K=0.12×shadow blocking area+0.05 Wherein, goodness of fit R 2 =0.93. The formula is verified by experiment, and can be directly used for real-time estimation of shadow blocking degree, avoiding the neglect of traditional MPPT to ripple characteristics.

[0064] The formula is the relationship between K value and blocking area, and the data measured under different blocking conditions are expressed by the relationship, and the goodness of fit R 2 =0.93.

[0065] The control instruction obtained by the intelligent scheduling controller includes an instruction of cyclically accessing each MPPT node by using a fixed time slice, or an instruction of triggering an emergency scan.

[0066] Further, the protocol stack design of the communication module is divided into three layers: the physical layer adopts a differential signal transmission circuit, has an anti-electromagnetic interference design, and ensures the stability of signal transmission; the data link layer defines a frame structure, including a synchronization header, a data segment and a check code, and guarantees data integrity through a three-way handshake confirmation mechanism; and the application layer defines a ripple coefficient compression encoding rule, for optimizing data transmission efficiency.

[0067] The ripple coefficient compression encoding rule specifically includes: The application dynamically encodes according to the dynamic characteristics of the ripple coefficient, to reduce the bandwidth when the K value is stable (such as no shadow), and to retain the accuracy when the ripple coefficient mutates (such as dynamic shadow); the K value compression encoding rule is specifically as follows: (1) Quantize the ripple coefficient, to map continuous ripple coefficient values (such as 2%~8%) into a limited discrete value, to reduce the transmission data amount, and specifically as follows: Fixed interval quantization: divide the range of the ripple coefficient into multiple intervals (such as 0~10% into 10 intervals), and represent each interval by 4-bit binary (0~15).

[0068] (2) Calculate the difference between the current ripple coefficient and the ripple coefficient at the last moment according to the quantized ripple coefficient. Specifically, differential encoding is performed on the quantized ripple coefficient, to reduce the redundant transmission of continuous K values, and only the changed part is transmitted. The differential encoding specifically includes: calculating the K value difference between the current K value and the K value at the last moment, instead of directly transmitting the K value.

[0069] For example: K1=5%→binary 0101 (4 bits) K2=6%→K value difference +1%→binary 0001 (4 bits) Sign bit: 1 bit is used to represent the positive and negative of the K value difference (+1% or -1%), and the remaining 3 bits represent the absolute value.

[0070] By differential coding, the data amount can be reduced, especially for the case where the ripple coefficient changes little; at the same time, the dynamic characteristics of K value (such as the judgment basis for shadow occlusion mutation) are reserved.

[0071] (3) According to the difference between the current ripple coefficient and the ripple coefficient at the last moment, adaptive coding is carried out, when the difference is less than the preset threshold, differential coding is used, otherwise fixed quantization coding is used Among them, the target of adaptive coding is to dynamically adjust the coding mode according to the volatility of K value, balance the compression rate and accuracy, and the implementation is as follows: Threshold judgment is carried out: If ΔK < preset threshold, differential coding is used, and only ΔK is transmitted.

[0072] If ΔK ≥ preset threshold, fixed quantization coding is used, and K value is directly transmitted.

[0073] Further, according to the volatility (such as standard deviation) of historical ΔK, the quantization step or coding mode is dynamically adjusted. When the K value is stable (such as no shadow), the bandwidth is reduced, and when the K value is mutated (such as dynamic shadow), the accuracy is reserved.

[0074] (4) Anti-interference mechanism: through CRC check code to ensure data reliability; Set the check code: use CRC-8 check code or exclusive or check to ensure data integrity.

[0075] When the receiving end fails to check, the retransmission mechanism (such as retransmitting the last valid data) is triggered.

[0076] Its core logic includes: Quantization: map K value to discrete value, reduce transmission data amount.

[0077] Differential coding: only transmit the change part of K value, reduce redundancy.

[0078] Adaptive adjustment: dynamically select coding mode according to K value volatility, balance compression rate and accuracy.

[0079] In view of the characteristics of limited bandwidth, large noise and high real-time requirement of power line carrier communication, the ripple coefficient is compressed and coded by a lightweight compression algorithm, the efficient and reliable transmission of the ripple coefficient is realized, and the system can realize real-time cooperative control of MPPT in the power line communication environment. The technical scheme not only solves the bandwidth bottleneck problem in the prior art, but also significantly improves the reliability and response speed of the system in the actual power line environment.

[0080] In addition, the communication module further improves the communication quality through performance optimization measures, for example, using forward error correction coding (FEC) to reduce the bit error rate, and using data compression technology to reduce transmission delay and improve communication efficiency.

[0081] Further preferably, the communication module improves communication security through the integration of an AES encryption engine, and the alternative solution can support LoRa spread spectrum technology to enhance long-distance transmission capability, further expanding the application scenarios of the module.

[0082] The distributed MPPT execution unit module executes power point tracking according to the scheduling instructions of the intelligent scheduling controller module, and feeds back real-time data to form a closed-loop control.

[0083] As shown in Figure 2 The application further provides a photovoltaic power optimization method based on power line carrier wave communication cooperative control, which can be implemented based on the photovoltaic power optimization system, and the method comprises the following steps: Step 1, the ripple feature detection module is connected to the output end of the photovoltaic array, real-time acquisition of the original current signal of the photovoltaic array is performed, and detection processing is performed to obtain detection data and send the detection data to the intelligent scheduling controller; Specifically, the method comprises the following steps: Real-time acquisition of the original current signal of the photovoltaic array is performed, and the original current signal is processed and calculated to obtain a ripple coefficient; Specifically, the calculation of the ripple coefficient comprises the following steps: The original current signal is acquired by the sensor unit, and after high-frequency noise is filtered by the signal conditioning unit, the digital processing unit performs digital processing; The digital signal is subjected to frequency spectrum analysis by using an FFT algorithm, and the energy integral value of the 10Hz-100Hz characteristic frequency band is calculated; The ripple coefficient K is calculated according to the energy integral value, and the calculation formula is as follows: K=(energy integral value / I_avg)×100% Wherein, I_avg is the average current value of the photovoltaic array, Preferably, the calculated ripple coefficient is compressed and encoded to obtain a compressed ripple coefficient; The calculated ripple coefficient is compressed and encoded, specifically comprising the following steps: The continuous ripple coefficient K value is quantized into a discrete value to reduce the data amount; the difference between the current ripple coefficient and the ripple coefficient at the previous moment is calculated; When |difference|<0.5%, differential encoding is used to transmit only the difference value, and when |difference|≥0.5%, fixed quantization encoding is used to directly transmit the ripple coefficient K value; The CRC-8 check code is used to ensure data integrity, and the compressed ripple coefficient data is generated; Step 2, the intelligent scheduling controller dynamically allocates the scanning priority of the MPPT node based on the detection data to obtain a scheduling instruction; Step 2 specifically includes: Step 2-1, judging the working mode of the photovoltaic power generation system according to the ripple coefficient change rate ΔK in the detection data; When ΔK < 0.5% / ms, the system is in a stable state, and enters a regular polling mode; When ΔK ≥ 0.5% / ms, the system is in a dynamic change state, and enters an interrupt preemption mode; Wherein, the ripple coefficient change rate ΔK is the absolute difference between the K values of adjacent two times divided by the time interval: ΔK = |K n -K n-1 | / Δt In the formula, K n and K n-1 are the ripple coefficients at the current time and the previous time, and Δt is the calculation time interval (unit: ms).

[0084] Preferably, the K value is calculated every 10 ms, so Δt = 10 ms, and the unit of the ripple coefficient change rate ΔK is % / ms.

[0085] Step 2-2, priority allocation according to the system working mode; In the regular polling mode, the intelligent scheduling controller cyclically accesses each MPPT node according to a preset fixed time slice, so as to ensure the periodicity of the system operation; In the interrupt preemption mode, the intelligent scheduling controller suspends the current polling task according to the high priority mark in the detection data, and preferentially allocates scanning resources for the high priority node, so as to realize the rapid response to the sudden shadow change; As a preferred embodiment, the intelligent scheduling controller estimates the future K value and the future K value change rate through a prediction model according to the historical ripple data and the ripple data change rate, and adjusts the system working mode in advance according to the future K value change rate; When the predicted future K value change rate ≥ 0.5% / ms, the interrupt preemption mode is triggered in advance, the scheduling instruction for emergency scanning is generated, and the pre-judgment scheduling is realized; after the emergency scanning is completed, the regular polling or the pre-judgment scheduling is resumed according to the actual ripple coefficient change rate ΔK value.

[0086] Preferably, the prediction model can use an ARIMA-based time series prediction model.

[0087] Further, a mapping model of the ripple coefficient and the shadow blocking area is constructed, the current shadow blocking area is calculated according to the ripple coefficient, and the node priority is judged according to the current shadow blocking area; Specifically, the ripple coefficient K value is positively correlated with the shadow blocking area of the photovoltaic array, and a mapping model of the ripple coefficient and the shadow blocking area is as follows: K = 0.12 * blocking area + 0.05 According to the mapping model and the ripple coefficient K value, the current shadow blocking area can be calculated. According to the shadow blocking area, whether the node is a high priority node is determined: When the proportion of the blocking area to the total area exceeds 20%, the node is marked as a high priority node, and the change rate ΔK of the ripple coefficient is taken as the triggering basis of the emergency event; Specifically, when the intelligent scheduling controller detects that the shadow blocking area of a node exceeds 20% and the K value change rate ΔK is greater than or equal to 0.5% / ms, the system triggers an emergency event, and at this time the intelligent scheduling controller generates a scheduling instruction immediately, marks the node as a high priority node, shortens the MPPT scanning time slice of the node (from 100 ms to 30 ms), and arranges the scanning order of the node preferentially.

[0088] The communication module transmits the scheduling instruction to the distributed MPPT execution unit through power line carrier communication, and after the execution unit receives the instruction, the MPPT scanning of the high priority node is performed preferentially, so that power optimization is performed at the initial stage of shadow change, and system power oscillation is avoided.

[0089] Step 2-3, generating a scheduling instruction: integrating the above decision results into a scheduling instruction containing mode identification (regular polling / interrupt preemption), node priority, scanning order and time slice allocation.

[0090] Step 3, the communication module sends the scheduling instruction to the distributed MPPT execution unit; The communication module in the application is a communication module based on a lightweight power line communication protocol, and step 3 specifically includes: Step 3-1, protocol encapsulation: the protocol stack of the communication module encapsulates the scheduling instruction to obtain an encapsulated scheduling instruction; The physical layer adopts a differential signal transmission circuit to enhance the anti-electromagnetic interference capability; the data link layer organizes the instruction data into a communication frame containing a synchronization header, a data segment and a CRC check code, and guarantees the integrity of data transmission through a three-way handshake confirmation mechanism; Step 3-2, the application layer adopts a lightweight communication protocol to compress the data, reduces the bandwidth occupation to 400 bps, and ensures efficient and reliable transmission of the scheduling instruction under the limited bandwidth of the power line carrier communication; Step 3-3, instruction delivery: power line carrier communication is performed through the direct current bus of the photovoltaic system, and the encapsulated scheduling instruction is broadcast to all distributed MPPT execution units.

[0091] Step 4, the distributed MPPT execution unit is used for executing specific power point adjustment at each MPPT node according to the received scheduling instruction.

[0092] Step 4 specifically includes: Step 4-1, instruction analysis: each distributed MPPT execution unit receives and analyzes the scheduling instruction sent by the communication module, extracts mode identification, scanning order and time slice allocation information, and identifies whether the node is a high priority node that needs to be scanned at present; Step 4-2, power point adjustment according to the system working mode: In the conventional polling mode, each execution unit uses the perturb and observe (P&O) or incremental conductance (IncCond) method to perform MPPT scanning at the node according to the time slice allocated by the instruction; In the interrupt preemption mode, the execution unit marked as high priority immediately starts MPPT scanning and completes power point tracking in priority.

[0093] Step 4-3, closed-loop feedback: after each MPPT node executes the adjustment instruction, the corresponding ripple feature detection module reacquires the adjusted photovoltaic current ripple signal, generates new detection data and feeds back to the HSS intelligent scheduling controller, forming a closed-loop control of “detection-decision-execution-feedback”, and continuously optimizing the output power of the photovoltaic system.

[0094] The beneficial effects of the present application are that, compared with the prior art, the photovoltaic power optimization system based on power line carrier communication cooperative control proposed in the present application improves the power generation efficiency of the photovoltaic power generation system in the shadow scenario, reduces the scanning conflict rate, and improves the optimization efficiency of the MPPT algorithm.

[0095] The present disclosure can be a system, a method, and / or a computer program product. The computer program product can include a computer readable storage medium having computer readable program instructions embodied therewith, which cause a processor to implement various aspects of the present disclosure.

[0096] Computer readable storage media can be tangible storage media which can retain and store instructions for use by an instruction execution device. Computer readable storage media can be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer readable storage media include the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded thereon, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.

[0097] Computer readable program instructions described herein can be downloaded to respective computing / processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and / or a wireless network. The network can comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and / or edge servers. A network adapter card or network interface in each computing / processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing / processing device.

[0098] Computer readable program instructions for carrying out operations of the present disclosure can be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C++ or the like, and conventional procedural programming languages such as the "C" programming language or similar programming languages. The computer readable program instructions can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate array (FPGA), or programmable logic array (PLA) can execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present disclosure.

[0099] Finally, it should be noted that the above-mentioned embodiments are merely used to illustrate the technical solutions of the present application, but not to limit it. Although the present application has been described in detail with reference to the above-mentioned embodiments, those skilled in the art should understand that the specific embodiments of the present application can be modified or replaced, and any modification or replacement without departing from the spirit and scope of the present application should be covered in the protection scope of the claims of the present application.

Claims

1. A photovoltaic power optimization system based on power line carrier communication cooperative control, characterized in that, The system includes: Ripple feature detection module, intelligent scheduling controller, communication module, and distributed MPPT execution unit; The ripple feature detection module is connected to the output terminal of the photovoltaic array, which collects the raw current signal of the photovoltaic array in the photovoltaic power generation system in real time, detects and processes the collected raw current signal to obtain the ripple coefficient, and sends it to the intelligent scheduling controller. The intelligent scheduling controller dynamically allocates the scanning priority of each MPPT node in the photovoltaic power generation system based on the ripple coefficient to obtain scheduling instructions. The communication module is used to send the scheduling instructions to the distributed MPPT execution unit, which is used to perform specific power point adjustments at each MPPT node according to the received scheduling instructions.

2. The photovoltaic power optimization system based on power line carrier communication cooperative control according to claim 1, characterized in that, The ripple feature detection module includes a Hall sensor group, a signal conditioning unit, a data processing unit, and a spectrum analysis processor; The Hall sensor group is used to acquire the raw current signal output by the photovoltaic array in real time. After the acquired raw current signal is filtered out by the signal conditioning unit to remove high-frequency noise, it is digitized by the data processing unit to obtain a digital signal. The spectrum analysis processor performs spectrum analysis on the digital signal to obtain the energy integral value of the characteristic frequency band, and calculates the ripple coefficient based on the energy integral value.

3. A photovoltaic power optimization system based on power line carrier communication cooperative control according to claim 2, characterized in that, The formula for calculating the ripple coefficient K is as follows: K = (Energy integral value / I_avg) × 100% Where I_avg is the average current value.

4. A photovoltaic power optimization system based on power line carrier communication cooperative control according to claim 1, characterized in that, The intelligent scheduling controller dynamically allocates MPPT scan priorities based on the ripple coefficient, specifically including: The ripple coefficient change rate ΔK is calculated based on the ripple coefficient, and the system operating mode is determined based on the ripple coefficient change rate and the set threshold. When the rate of change of the ripple coefficient ΔK is less than the set threshold, it indicates that the system is in a stable state. At this time, it enters the normal polling mode, which uses a fixed time slice to cyclically access each MPPT node. When the rate of change of the ripple coefficient ΔK is greater than or equal to the set change threshold, it indicates that the system is in a dynamic state. At this time, the system enters the interrupt preemption mode and triggers an emergency scan to respond to the change in system state.

5. A photovoltaic power optimization system based on power line carrier communication cooperative control according to claim 4, characterized in that, The specific method for calculating the ripple coefficient change rate ΔK is as follows: ΔK=|K n -K n-1 | / Δt Among them, Kn and K n-1 These are the ripple coefficients for the current time and the previous time, respectively, and Δt is the calculation time interval, with the unit of Δt being ms.

6. A photovoltaic power optimization system based on power line carrier communication cooperative control according to claim 4, characterized in that, The intelligent scheduling controller dynamically allocates MPPT scan priorities based on the ripple coefficient, and also includes: Based on historical ripple coefficients and ripple coefficient change rates, predict future ripple coefficients, calculate future ripple coefficient change rates based on future ripple coefficients, and adjust the system operating mode based on future ripple coefficient change rates. When the predicted rate of change of the future ripple coefficient exceeds the preset prediction change threshold, the interruption preemption mode is triggered in advance, an emergency scan scheduling instruction is generated, and normal polling is resumed after the emergency scan is completed.

7. A photovoltaic power optimization system based on power line carrier communication cooperative control according to claim 4, characterized in that, It also includes: constructing a mapping model between ripple coefficient and shadow occlusion area, calculating the current shadow occlusion area based on the ripple coefficient, and determining node priority based on the current shadow occlusion area; The mapping model between the ripple coefficient and the shadow occlusion area is as follows: K = 0.12 × shadow occlusion area + 0.05 The shadow occlusion area of ​​the current node is obtained based on the mapping model and the ripple coefficient K, and the node is determined to be a high-priority node based on the shadow occlusion area: When the occlusion area exceeds the preset occlusion threshold, the MPPT node is determined to be a high-priority node, and the ripple coefficient change rate is used as the triggering basis for the emergency event.

8. A photovoltaic power optimization system based on power line carrier communication cooperative control according to claim 7, characterized in that, The scheduling instructions include: mode identifier, scan priority, and time slice allocation; Among them, the mode identifiers include the regular polling mode identifier and the interruption preemption mode identifier; Scanning priority and time slice allocation are determined based on the system's operating mode, as follows: In the conventional polling mode, the intelligent scheduling controller visits each MPPT node in a loop according to a preset fixed time slice to ensure the periodicity of system operation; In interrupt preemptive mode, the intelligent scheduling controller pauses the current polling task based on the high-priority markers in the detection data and prioritizes the allocation of scanning resources to high-priority nodes.

9. A photovoltaic power optimization system based on power line carrier communication cooperative control according to claim 1, characterized in that, The protocol stack of the communication module includes: physical layer, data link layer and application layer; The physical layer employs differential signal transmission circuitry and features electromagnetic interference resistance to ensure signal transmission stability. The data link layer defines the frame structure, which includes a synchronization header, data segments, and a checksum, and ensures data integrity through a handshake confirmation mechanism. The application layer compresses the ripple coefficient according to the ripple coefficient compression coding rules.

10. A photovoltaic power optimization method based on power line carrier communication cooperative control, implemented using the photovoltaic power optimization system based on power line carrier communication cooperative control as described in any one of claims 1-9, characterized in that, The method includes the following steps: Step 1: The ripple feature detection module is connected to the output terminal of the photovoltaic array in the photovoltaic system, collects the original current signal of the photovoltaic array in real time and performs detection processing to obtain the ripple coefficient and send the ripple coefficient to the intelligent scheduling controller. Step 2: The intelligent scheduling controller dynamically allocates the scanning priority of MPPT nodes based on the ripple coefficient to obtain scheduling instructions; Step 3: The communication module sends the scheduling instructions to the distributed MPPT execution unit; Step 4: The distributed MPPT execution unit is used to perform specific power point adjustments on each MPPT node according to the received scheduling instructions.

Images