A photovoltaic panel control management system

Abstract

The present application relates to a kind of photovoltaic panel control management systems, belong to photovoltaic power generation technical field.The system includes edge perception unit, group string optimization unit and central control unit.Edge perception unit is deployed in each photovoltaic panel, and real-time acquisition electrical parameter, environmental parameter and physical state;Group string optimization unit connects the edge perception unit in the same group string, and executes dynamic working point matching and fault isolation;Central control unit is connected group string optimization unit by communication network, and executes whole station monitoring and strategy scheduling.Group string optimization unit is divided into several communication areas, and each area is set up regional agent node to gather data, and random backoff is used to transmit data to central control unit to avoid channel conflict.When regional agent node fails, subnode is automatically switched to direct communication mode.The system improves the power generation efficiency and reliability of photovoltaic system through multi-level architecture and intelligent communication mechanism.

Description

TECHNICAL FIELD

[0001] The present application relates to a photovoltaic power generation system, in particular to a photovoltaic panel control management system. BACKGROUND

[0002] With the continuous development of photovoltaic power generation technology, the scale of photovoltaic power stations is increasing, and higher requirements are put forward for the monitoring and management of photovoltaic systems. The existing photovoltaic system usually adopts a centralized monitoring architecture, which realizes the monitoring and management of the photovoltaic system by collecting the current, voltage and other parameters of the photovoltaic panel.

[0003] At present, the photovoltaic panel control management system mainly includes data acquisition unit, data processing unit and monitoring platform and other components. Among them, the data acquisition unit is responsible for collecting the current, voltage and other parameters of the photovoltaic panel; the data processing unit is responsible for processing and analyzing the collected data; the monitoring platform is responsible for monitoring and managing the photovoltaic system.

[0004] However, the existing photovoltaic panel control management system still has the following problems in practical application:

[0005] Firstly, in a large-scale distributed system, the existing communication architecture usually adopts a centralized proxy mode, when the proxy node fails, it will cause the communication of the whole region to be interrupted, affecting the stability and reliability of the system.

[0006] Secondly, in terms of communication mode, when the number of nodes is large, the communication efficiency is low, the system load is large, and it is difficult to meet the real-time monitoring demand of large-scale photovoltaic power stations.

[0007] Therefore, it is urgent to develop a photovoltaic panel control management system that can solve the above problems and improve the communication reliability of the system. SUMMARY

[0008] In order to solve the above problems, the present application provides a photovoltaic panel control management system.

[0009] The technical scheme adopted by the present application to solve its technical problems is to provide a photovoltaic panel control management system, which comprises an edge perception unit, a string optimization unit and a central control unit, wherein the edge perception unit is deployed on each photovoltaic panel and is used to collect electrical parameters, environmental parameters and physical states in real time; the string optimization unit is connected to the edge perception units in the same string and is configured to perform dynamic working point matching and fault isolation; the central control unit is connected to the string optimization unit through a communication network and is used to perform whole station monitoring and strategy scheduling.

[0010] Preferably, the group string optimization unit is divided into a plurality of communication areas, each of which is provided with a regional agent node for gathering data of the sub-nodes in the area; the regional agent node adopts a random backoff method when transmitting data to the central control unit to avoid channel conflict; when the central control unit detects a failure of the regional agent node, the sub-nodes in the communication area switch to a direct communication mode with the central control unit and communicate with the central control unit using a random backoff method until a new regional agent node is assigned.

[0011] Further, the edge perception unit comprises a fault isolation module for triggering physical disconnection of the photovoltaic panel based on current anomaly detection; and a scanning module for generating current and voltage curves through an adjustable load module.

[0012] Optionally, the group string optimization unit comprises an analog calculation module for performing calculation processing of maximum power point tracking.

[0013] Preferably, the analog calculation module comprises a logarithmic transformation module configured to convert the group string voltage signal and the group string current signal into a logarithmic voltage signal and a logarithmic current signal, respectively; a power calculation unit receiving an output of the logarithmic transformation module, performing summation operation on the logarithmic voltage signal and the logarithmic current signal through an analog adder to generate a logarithmic power signal; and a gradient generation unit configured to inject a controllable disturbance signal into the logarithmic voltage signal and output a power gradient signal based on a response of the logarithmic power signal to the disturbance.

[0014] Further, the analog calculation module further comprises a multi-peak scanning unit configured to perform the following operations in response to a power drop or an environmental mutation trigger signal: control the group string operating voltage to scan within a preset range; capture the maximum power value in the scanning process through an analog peak detection circuit; and when the real-time power reaches a set threshold of the maximum power value, lock the corresponding operating voltage and switch to run at the voltage point.

[0015] Optionally, the trigger signal is generated by differentiating a smoothed power signal and detecting that the negative duration thereof exceeds a preset threshold, or is generated according to a mutation detection signal of an illumination sensor and a temperature sensor.

[0016] Preferably, the data generated by the sub-nodes in the communication area comprises key data and periodic data, the key data being transmitted directly from the sub-nodes to the central control unit by skipping the regional agent node; and the key data being sent to the central control unit after being encrypted in an exclusive or manner using a communication key.

[0017] Further, the communication key obtaining step comprises: obtaining a unique hardware identifier of the group string optimization unit and taking it as an input, combining a digest value of the periodic data transmitted in the last communication process, and generating the communication key through a key derivation function.

[0018] Optionally, the key derivation function is a function using an HMAC algorithm, and the combination of the last periodic data digest and the time factor is subjected to a hash operation with the hardware identifier as the key.

[0019] The present application has the beneficial effect that: through the fusion communication method, the technical effect of considering the efficiency of large-scale deployment and fault tolerance ability is realized. The system adopts a layered proxy architecture, and in the normal state, the data is aggregated through the proxy node to improve the communication efficiency, and when the proxy node fails, the child node can automatically switch to a random communication avoidance mode to ensure the reliability of the system. Compared with the prior art, the present application can dynamically adapt to the network topology changes, such as the increase and decrease of proxy nodes, and by distinguishing the transmission modes of regular data and emergency data, the transmission efficiency of different types of data is optimized, so that regular data can be processed in batches, and emergency data can be responded preferentially, which is particularly suitable for large-scale photovoltaic power station and communication network mixed deployment scenarios. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 FIG. 1 is a structural schematic diagram of a photovoltaic panel control management system in an embodiment of the present application;

[0021] Figure 2 FIG. 2 is a structural schematic diagram of a simulation calculation module in an embodiment of the present application;

[0022] Figure 3 FIG. 3 is a communication method flow chart of a photovoltaic panel control management system in an embodiment of the present application;

[0023] Figure 4 FIG. 4 is a key synchronization method flow chart of a center control unit and a child node in an embodiment of the present application. DETAILED DESCRIPTION

[0024] The technical solutions of the present application will be described below in conjunction with the embodiments and the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0025] Embodiment 1: see Figures 1-2 A photovoltaic panel control management system, comprising an edge perception unit, a group string optimization unit and a center control unit.

[0026] The edge perception unit is arranged on each photovoltaic panel to collect electrical parameters, environmental parameters and physical states in real time. In a specific scheme, the edge perception unit can be directly installed on the back of the photovoltaic panel to monitor the running state of the photovoltaic panel through the built-in sensor. The electrical parameters include key electrical indicators such as photovoltaic panel output voltage, output current and power; the environmental parameters include external factors affecting the power generation efficiency such as light intensity, environmental temperature and photovoltaic panel surface temperature; and the physical state includes physical characteristics such as photovoltaic panel inclination, vibration condition and surface dust degree.

[0027] In an embodiment, the edge perception unit includes a fault isolation module and a scanning module. The fault isolation module triggers the physical disconnection of the photovoltaic panel based on current anomaly detection. When the photovoltaic panel current anomaly is detected, the fault isolation module will immediately trigger the physical disconnection mechanism to isolate the faulty photovoltaic panel from the string, preventing the spread of faults and safety hazards. The current anomaly detection adopts an adaptive threshold algorithm to dynamically adjust the abnormality judgment standard according to the light conditions and historical operation data to reduce the false alarm rate.

[0028] The scanning module generates current and voltage curves through the adjustable load module. The adjustable load module can quickly scan the characteristics of the photovoltaic panel without affecting normal power generation, and generate accurate I-V curves. During the scanning process, the adjustable load module samples 20 points at intervals of 50 ms in the range of 0V to the open circuit voltage of the photovoltaic panel, forming a complete current-voltage characteristic curve. These curve data are used to evaluate the health status of the photovoltaic panel, predict potential faults and optimize the working point. The scanning module adopts a low-power design and is activated only when the system needs it, performing a complete scan once every hour on average, and triggering additional scans in special cases.

[0029] The string optimization unit is connected to the edge perception units in the same string and is configured to perform dynamic working point matching and fault isolation. The string optimization unit adopts a distributed architecture, each string is equipped with an independent optimization unit and a string fault isolation module, which is responsible for managing the cooperative work of all photovoltaic panels in the string. The string optimization unit establishes a communication connection with the edge perception unit through the RS-485 bus to realize data exchange and control instruction issuing. The dynamic working point matching is used to automatically adjust the working point of the photovoltaic panels in the string according to the real-time environmental conditions and their respective characteristics, to maximize the overall power generation efficiency of the string. The string fault isolation module is used to isolate the photovoltaic panel in the string when a fault is detected, to ensure the normal work of other photovoltaic panels.

[0030] The string optimization unit also includes an analog calculation module for performing a calculation process of maximum power point tracking. The analog calculation module uses an analog circuit to realize complex calculation functions, which has lower power consumption and faster response speed compared with traditional digital processing methods. The analog calculation module can process string voltage and current signals in real time, find the best working point through a preset algorithm, and make the photovoltaic string work near the maximum power point at all times.

[0031] In an embodiment, the analog calculation module includes a logarithmic transformation module, a power calculation unit, and a gradient generation unit. The logarithmic transformation module is configured to convert the string voltage signal and the string current signal into a logarithmic voltage signal and a logarithmic current signal, respectively. The logarithmic transformation module uses a high-precision logarithmic amplifier circuit to convert linear voltage and current signals into logarithmic domain signals, realizing wide dynamic range signal processing. The logarithmically transformed signals have better processing characteristics and are suitable for subsequent power calculation and gradient analysis.

[0032] The power calculation unit receives the output of the logarithmic transformation module, and performs summation operation on the logarithmic voltage signal and the logarithmic current signal through an analog adder to generate a logarithmic power signal. In the logarithmic domain, multiplication operation is converted into addition operation, greatly simplifying circuit design and power consumption. The analog adder uses a high-precision operational amplifier, which has low offset voltage and temperature drift, ensuring stable calculation accuracy under various working conditions. The logarithmic power signal directly reflects the real-time output power of the photovoltaic string, providing basic data for subsequent maximum power point tracking.

[0033] The gradient generation unit is configured to inject a controllable disturbance signal into the logarithmic voltage signal and output a power gradient signal based on the response of the logarithmic power signal to the disturbance. In a specific embodiment, the gradient generation unit uses a small signal disturbance method to inject a sinusoidal disturbance signal with an amplitude of 0.5% of the working voltage and a frequency of 200 Hz into the system. By detecting the response of the logarithmic power signal to the disturbance, the gradient generation unit calculates the slope of the power-voltage curve in real time, i.e. the power gradient. The power gradient signal directly reflects the position relationship of the current working point relative to the maximum power point, providing the basis for adjusting direction and amplitude for the control system.

[0034] Because the traditional V*I multiplier has limited accuracy in a wide dynamic range (especially in low light), and calculating the power gradient (dP / dV) requires additional differential circuits, it is easily disturbed by noise. The analog calculation module used in this embodiment includes the following processing processes:

[0035] Logarithmic transformation module signal processing: The string voltage V and current I signals are input into high-precision logarithmic amplifiers (using matched bipolar transistors to realize logarithmic characteristics), and logarithmic voltage Log(V) and logarithmic current Log(I) signals are output.

[0036] The power calculation unit generates a logarithmic power:

[0037] The Log(V) and Log(I) signal inputs a high-precision analog adder (using a low offset voltage, low temperature drift operational amplifier), which directly outputs a logarithmic power signal Log(P), satisfying:

[0038]

[0039] The gradient generation unit injects a controllable perturbation ΔLog(V) into the Log(V) signal:

[0040]

[0041] The perturbation type is a 200Hz sine wave consistent with the above.

[0042] The perturbation amplitude is 0.5% of the operating point voltage (i.e. 0.5%·Log(V)).

[0043] Response extraction: a band-pass filter (center frequency 200Hz) is used to extract the component ΔLog(P) of the Log(P) signal corresponding to the perturbation frequency.

[0044] The gradient approximation value is calculated by an analog divider, with the specific formula being:

[0045]

[0046] The linear voltage / current signal is converted to a logarithmic domain signal by a logarithmic amplifier (e.g. the voltage / current of a photovoltaic string may be as low as a few millivolts / milliamperes in low light, and up to hundreds of volts / amperes in high light). The logarithmic compression characteristic allows the circuit to handle signals across multiple orders of magnitude, avoiding the precision loss caused by amplifier saturation or excessive noise ratio in traditional linear circuits at low signal amplitudes.

[0047] The power calculation is converted from "linear domain multiplication" to "logarithmic domain addition", which can be implemented using only an analog adder;

[0048] The high-precision power and gradient signals provide reliable feedback for the MPPT algorithm, allowing the system to converge more quickly and accurately to the maximum power point.

[0049] The analog calculation module also includes a multi-peak scanning unit configured to perform the following operations in response to a power drop or environmental mutation trigger signal: control the photovoltaic string operating voltage to scan within a preset range; capture the maximum power value during the scanning process through an analog peak detection circuit; when the real-time power reaches a set threshold of the maximum power value, lock the corresponding operating voltage and switch to running at that voltage point. The multi-peak scanning unit can effectively handle the multi-peak power curve situation of the photovoltaic string under partial shadow conditions, avoiding the system from falling into a local maximum.​​​

[0050] Specifically, the scanning range covers 20% to 90% of the open-circuit voltage of the string, and the scanning rate is 2V / s, which ensures the completion of full-range scanning in a short time without affecting the stability of the system. The threshold is usually set to 95%, that is, when the real-time power reaches 95% of the maximum power value detected during the scanning process, the system will lock the corresponding operating voltage and switch to running at that point.

[0051] Wherein, the trigger signal is generated by differentiating the smoothed power signal and detecting its negative duration exceeding a preset threshold, or generated according to the mutation detection signal of the light sensor and the temperature sensor. In an embodiment, the smoothed power signal is processed by a first-order low-pass filter to filter out high-frequency noise, with a cutoff frequency of 0.5Hz. The differential circuit calculates the power change rate, and when a negative change rate lasting more than 2 seconds is detected, the multi-peak scanning operation is triggered. The mutation detection of the light sensor and the temperature sensor is based on a change rate threshold, when the light intensity change rate exceeds 50W / (m²·s) or the temperature change rate exceeds 2℃ / s, the multi-peak scanning operation is also triggered. Through this double triggering mechanism, the system can respond to environmental changes and partial shadow conditions in time.

[0052] The central control unit is connected to the string optimization unit through a communication network, which is used to perform full-site monitoring and strategy scheduling. The central control unit can use an industrial-grade computing platform. The communication network adopts a hierarchical architecture, including a field bus layer and an Ethernet layer, to ensure the real-time and reliability of data transmission. The strategy scheduling function formulates the optimal power generation plan and maintenance strategy according to the power grid demand, weather forecast and historical operation data.

[0053] In a specific communication network, the string optimization unit is divided into several communication areas, and a regional agent node is set in each communication area to collect data from the sub-nodes in the area. The communication area is divided based on physical distance and network topology, and each communication area contains 5 to 10 string optimization units. The regional agent node is automatically selected by the system according to the communication quality and computing capacity, and is responsible for collecting and preprocessing the data of all sub-nodes in the area, reducing the communication burden of the central control unit. The regional agent node includes a local cache module for temporarily storing data in case of communication interruption, and automatically uploading after the communication is restored, ensuring data integrity.

[0054] The regional agent node uses a random backoff method to transmit data to the central control unit to avoid channel conflict. The random backoff method can be implemented based on the CSMA / CA protocol. When the regional agent node needs to transmit data, it first detects whether the channel is idle. If the channel is idle, it waits for a random time and then detects again. If the channel is still idle, it starts data transmission; if the channel is busy, it waits until the channel is idle and then performs the random backoff process. The random backoff time is randomly selected in the range of 5 ms to 50 ms, effectively reducing the probability of channel conflict.

[0055] When the central control unit detects that the regional agent node fails, the sub-nodes in the communication region switch to direct communication mode with the central control unit and use a random backoff method to communicate with the central control unit until a new regional agent node is assigned.

[0056] In an embodiment, the regional agent node failure detection is based on a heartbeat mechanism. When the central control unit does not receive the heartbeat signal of the regional agent node for 3 consecutive times, it is determined that the node fails. After failure determination, the central control unit broadcasts switching instructions to all sub-nodes in the communication region. After receiving the instructions, the sub-nodes immediately switch to direct communication mode. In direct communication mode, the sub-nodes also use a random backoff method to communicate with the central control unit, and the backoff time range is expanded to 10 ms to 100 ms to accommodate more nodes communicating at the same time. The central control unit will select a new regional agent node within 30 seconds according to the current network state and notify all sub-nodes to switch back to normal communication mode.

[0057] The data generated by the sub-nodes in the communication region includes critical data and periodic data. The critical data is transmitted directly from the sub-nodes to the central control unit without passing through the regional agent node. The critical data is encrypted using a communication key in an XOR manner before being sent to the central control unit.

[0058] The critical data includes information such as failure alarms, security events, and emergency state changes that require immediate response. Direct transmission ensures the timeliness of this information. Periodic data includes routine operating parameters, performance indicators, and environmental data. The data is aggregated by the regional agent node before being transmitted to the central control unit, reducing network load. XOR encryption uses bit operations, which has a small computational overhead and is suitable for resource-constrained edge devices. It also provides basic data protection functions.

[0059] The communication key acquisition step includes obtaining the unique hardware identifier of the group string optimization unit as input, combining the digest value of the periodic data transmitted in the last communication process, and calculating the communication key through a key derivation function.

[0060] In this embodiment, the unique hardware identifier is generated according to the physical characteristics of the device, such as the serial number, MAC address or unique ID in the dedicated security chip. The periodic data digest value is calculated using the SHA-256 algorithm, with a length of 32 bytes. The key derivation process ensures that even if an attacker intercepts the communication data, it is difficult to derive the key for the next communication, improving system security.

[0061] The key derivation function is a function that uses the HMAC algorithm to hash the combination of the last period data digest and the time factor using the hardware identifier as the key. The HMAC algorithm is implemented based on the SHA-256 hash function, providing high-strength cryptographic protection. The time factor is generated by the system clock, with a precision of seconds, ensuring that even with the same hardware identifier and data digest, the keys generated at different times are not the same. The output length of the hash operation is 32 bytes, which is used to generate a 256-bit communication key to meet the security requirements of modern cryptography.

[0062] In an embodiment, referring to Figure 4 , the key synchronization process between the central control unit and the child node includes:

[0063] S1: The central control unit and the child node exchange initial seed keys and synchronization parameters (such as time window, hash algorithm) through a secure channel when first connected;

[0064] S2: Both parties agree to use the average environmental data within a predetermined time period (configurable) to generate keys; for example:

[0065] Key K1 is generated based on environmental data at time T1, with a validity period of T1-T1+5 minutes;

[0066] Key K2 is generated based on data at time T1+5 minutes, and so on.

[0067] S3: If the clock deviation between the central control unit and the child node exceeds the threshold value (such as ±5 seconds), trigger the re-synchronization process.

[0068] The time stamp verification field is attached during data transmission to ensure that both parties use the same time window data to generate the key.

[0069] In a possible implementation, the edge perception unit uses a low-power ARM series microcontroller to handle data processing and communication with the string optimization unit.

[0070] The sensor module includes:

[0071] Hall current sensor for current acquisition;

[0072] Voltage transformer for voltage acquisition.

[0073] The environmental parameter collection includes:

[0074] The light sensor is used for measuring the light intensity; the digital temperature sensor is used for monitoring the temperature; and the temperature and humidity sensor is used for acquiring the environmental temperature and humidity data.

[0075] The physical state monitoring includes: the inclination sensor is used for monitoring the inclination angle of the photovoltaic panel; and the mechanical vibration sensor is used for detecting whether the panel body has abnormal vibration, and judging whether there is loosening or damage.

[0076] The fault isolation module is composed of a solid state relay (SSR) and a current detection sensor. When the current detection sensor finds abnormal current, the main control chip controls the solid state relay to cut off the photovoltaic panel circuit connection, so as to realize physical disconnection.

[0077] The scanning module includes an adjustable load module composed of a programmable resistor (such as AD5290) and a power MOSFET. The load resistor is adjusted by the main control chip to generate current and voltage curves under different working conditions.

[0078] The string optimization unit can use an FPGA chip to meet the real-time processing requirements of the simulation calculation module, and control the data interaction of each edge perception unit in the string.

[0079] The logarithmic transformation module in the simulation calculation module can use a logarithmic operational amplifier (such as AD8331) to convert the string voltage signal and the string current signal into a logarithmic voltage signal and a logarithmic current signal.

[0080] The power calculation unit uses an analog adder chip (such as AD8276) to perform summation operation on the logarithmic voltage signal and the logarithmic current signal to generate a logarithmic power signal.

[0081] The gradient generation unit generates a controllable disturbance signal by a signal generator (such as AD9833) and injects it into the logarithmic voltage signal. The power gradient signal is calculated by an analog multiplier (such as AD633) and an integrator (such as LF356).

[0082] The multi-peak scanning unit includes a voltage scanning control module (specifically composed of a DAC chip such as AD5623 and an operational amplifier) to control the scanning of the string working voltage in a preset range; and an analog peak detection circuit using a comparator (such as LM339) and a sample and hold circuit (such as LF398) to capture the maximum power value in the scanning process.

[0083] The intra-regional communication uses a CAN bus communication chip to realize fast and reliable data interaction with the edge perception units in the region.

[0084] If it is a regional proxy node, an Ethernet communication module is configured to perform high-speed data transmission with the central control unit.

[0085] The central control unit comprises:

[0086] The master server is used for running photovoltaic power station monitoring and management software, processing all station data and executing strategy scheduling.

[0087] The communication module is used for communicating with the group string optimization unit, and high-speed and long-distance communication with the group string optimization unit is realized through deployment of a fiber transceiver and an Ethernet switch, and TCP / IP protocol is supported.

[0088] The remote monitoring interface is equipped with a 4G / 5G communication module or a satellite communication module, realizes remote data transmission, and is convenient for remote monitoring and management of operation and maintenance personnel.

[0089] The display and storage module is used for connecting an industrial-grade liquid crystal display screen to display the running state of the power station in real time, and is configured with a large-capacity hard disk array and a database server to store historical running data for data analysis and fault tracing.

[0090] Embodiment 3: see Figure 3 A communication method of a photovoltaic panel control and management system, specifically comprising the following steps:

[0091] S101: The group string optimization unit is divided into multiple communication areas, and each area is assigned a regional proxy node, which is responsible for gathering data of the sub-nodes in the area;

[0092] When the system is initialized, all group string optimization units are divided into multiple communication areas (such as areas A, B and C), and each area is assigned a regional proxy node (such as the proxy node of area A is node A1);

[0093] S102: The sub-nodes periodically collect their own data and send them to the regional proxy nodes, and the proxy nodes receive and integrate the data;

[0094] The sub-nodes collect their own data (such as power generation and fault status) at a fixed period (such as every second / minute) and send them to the proxy nodes of the corresponding areas;

[0095] The proxy nodes receive all the data of the sub-nodes in the area, perform data deduplication, format unification, preliminary analysis (such as calculating the average power of the area), compress the data volume, and then upload them to the central control unit, so as to reduce the transmission volume of the original data, reduce the channel occupancy rate, and improve the transmission efficiency;

[0096] S103: The regional proxy nodes generate a backoff time by using a random backoff algorithm, detect the channel after the backoff is over, send data if the channel is idle, and backoff again if the channel is busy;

[0097] S104: After the central control unit detects a fault of the regional proxy node, it sends a direct communication mode switching instruction to the sub-nodes in the area;

[0098] S105: The fault area sub-nodes stop sending data to the original proxy node, and each uses a random backoff algorithm to attempt to communicate directly with the central control unit, and if there is a conflict, the backoff is restarted;

[0099] S106: The central control unit selects a new area proxy node from the fault area and configures it to be activated, the sub-nodes resume sending data to the new proxy node, and the system resumes the hierarchical communication mode. The central control unit selects a new proxy node from the fault area sub-nodes (the election conditions may include communication stability, remaining computing power, etc.). The central control unit issues a configuration instruction to the new proxy node to activate its proxy function. The sub-nodes stop the direct connection mode, resume sending data to the new proxy node, and the system returns to the hierarchical communication architecture.

[0100] When the power station contains thousands of sub-nodes, the communication pressure on the central control unit is reduced through the area proxy node. In a scenario where wireless signals are easily disturbed, the random backoff algorithm can effectively improve the communication success rate.

[0101] The above embodiments are only 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 foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A photovoltaic panel control and management system, characterized in that, include: Edge sensing units are deployed on each photovoltaic panel to collect electrical parameters, environmental parameters, and physical status in real time; The string optimization unit connects to the edge sensing unit within the same string and is configured to perform dynamic operating point matching and fault isolation. The central control unit is connected to the string optimization unit via a communication network and is used to perform station-wide monitoring and policy scheduling. The string optimization unit is divided into several communication areas based on physical distance and network topology. The data generated by the child nodes in the communication area includes key data and periodic data. Each communication area contains 5 to 10 string optimization units, and a regional proxy node is set in each communication area to aggregate and preprocess the periodic data of the child nodes in the area. The regional proxy node is automatically selected by the system according to communication quality and computing power, and the regional proxy node includes a local cache module for temporarily storing data in the event of communication interruption and automatically uploading it after communication is restored to ensure data integrity. Key data generated by child nodes in the communication area bypasses the regional agent node, is encrypted using the communication key in an XOR manner, and is directly transmitted to the central control unit. Periodic data is aggregated through the regional agent node before being transmitted. The regional agent node uses a random backoff method when transmitting data to the central control unit to avoid channel conflicts. When the central control unit detects a regional agent node failure, the child nodes in the communication area switch to direct communication mode with the central control unit and communicate with the central control unit using a random backoff method until a new regional agent node is assigned.

2. The system according to claim 1, characterized in that, The edge sensing unit includes: a fault isolation module for triggering the physical disconnection of the photovoltaic panel based on current anomaly detection; and a scanning module for generating current and voltage curves through an adjustable load module.

3. The system according to claim 1, characterized in that, The string optimization unit includes a simulation calculation module, which is used to perform maximum power point tracking calculations.

4. The system according to claim 3, characterized in that, The simulation calculation module includes: The logarithmic transformation module is configured to convert the string voltage signal and the string current signal into logarithmic voltage signal and logarithmic current signal, respectively. The power calculation unit receives the output of the logarithmic transformation module and performs a summation operation on the logarithmic voltage signal and the logarithmic current signal through an analog adder to generate a logarithmic power signal. The gradient generation unit is configured to inject a controllable perturbation signal into the logarithmic voltage signal and output a power gradient signal based on the response of the logarithmic power signal to the perturbation.

5. The system according to claim 3, characterized in that, The simulation calculation module also includes: The multi-peak scanning unit is configured to perform the following operations in response to a power drop or environmental abrupt change trigger signal: The control string operating voltage is scanned within a preset range; The maximum power value during the scanning process is captured by an analog peak detection circuit; When the real-time power reaches the set threshold of the maximum power value, the corresponding operating voltage is locked and the system switches to that voltage point for operation.

6. The system according to claim 5, characterized in that, The trigger signal is generated by differentiating the smoothed power signal and detecting when its negative duration exceeds a preset threshold, or by generating a sudden change detection signal from the light sensor and temperature sensor.

7. The system according to claim 1, characterized in that, The data generated by the child nodes in the communication area includes key data and periodic data. The key data is transmitted directly from the child nodes to the central control unit, bypassing the regional agent nodes. The key data is then encrypted using a communication key via XOR and sent to the central control unit.

8. The system according to claim 7, characterized in that, The steps for obtaining the communication key include: The unique hardware identifier of the string optimization unit is obtained and used as input. Combined with the digest value of the cycle data transmitted in the previous communication process, the communication key is generated by calculating through the key derivation function.

9. The system according to claim 8, characterized in that, The key derivation function is a function that uses the HMAC algorithm, with the hardware identifier as the key, and performs a hash operation on the combination of the previous period's data digest and the time factor.

Images