Adaptive overcurrent protection setting method for new energy box-type substation

Abstract

The application discloses a new energy box-type substation self-adaptive over-current protection setting method, and relates to the technical field of relay protection of power systems, and the method comprises the steps of establishing a multi-channel communication link with each new energy converter; collecting the running state and real-time apparent power through each communication interface polling; summarizing the total real-time apparent power and correcting the line loss; and generating the over-current protection setting value according to the corrected total real-time apparent power and the measured apparent power on the low-voltage side. The technical scheme of the application combines multi-channel collection, power summarization and correction, and self-adaptive setting, so that the protection threshold dynamically matches the actual output, the fault protection sensitivity under the low-output working condition is significantly improved, the over-current protection response capability and action reliability are improved, and the safe and stable operation of the transformer is ensured.

Description

Technical Field

[0001] This application relates to the field of power system relay protection technology, and in particular to an adaptive overcurrent protection setting method for new energy prefabricated substations. Background Technology

[0002] New energy prefabricated substations are key hub equipment in new energy power plants such as photovoltaic and wind power plants. Overcurrent protection, as the most core main protection of prefabricated substations, directly undertakes the task of rapid detection and isolation of major electrical faults such as transformer short circuits, inter-turn faults, and lead faults. It is a key line of defense to ensure equipment safety, prevent fault expansion, and avoid burn-out accidents.

[0003] However, existing overcurrent protection systems in new energy prefabricated substations are generally set to fixed values ​​based on the transformer's rated capacity. During periods of low output from new energy sources such as photovoltaic and wind power, when early internal faults such as minor inter-turn short circuits or winding deformation occur, the change in fault current is far below the operating threshold of traditional overcurrent protection. This leads to insufficient timely response and a tendency for protection to fail to operate. Prolonged fault persistence can easily escalate into catastrophic accidents such as phase-to-phase short circuits and equipment burnout. Furthermore, traditional protection schemes do not dynamically calculate based on the real-time operating status and output data of the new energy converter. They lack multi-channel communication acquisition, total real-time apparent power aggregation, and line loss correction mechanisms. They cannot adaptively adjust protection settings according to actual operating capacity, making it difficult to match the characteristics of large output fluctuations in new energy sources. Consequently, protection sensitivity and operational safety cannot be simultaneously achieved. Summary of the Invention

[0004] The purpose of this application is to provide an adaptive overcurrent protection setting method for new energy prefabricated substations, aiming to solve the technical problem of insufficient overcurrent protection sensitivity in the prior art.

[0005] To achieve the above objectives, this application provides a method for setting adaptive overcurrent protection settings in a new energy prefabricated substation, comprising: Establish multi-channel communication links with various new energy converters; The new energy converter is polled through each communication interface of the multi-channel communication link to collect the operating status and real-time apparent power of the new energy converter, and the real-time apparent power of the new energy converter in the operating state is calculated. Based on the calculated real-time apparent power of the new energy converters in operation, the total real-time apparent power is obtained by summing them up, and the total real-time apparent power is corrected for line loss to obtain the corrected total real-time apparent power. Based on the corrected total real-time apparent power and the measured apparent power on the low-voltage side of the prefabricated substation, an overcurrent protection setting is generated.

[0006] In one embodiment, the renewable energy converter is polled through each communication interface of the multi-channel communication link to collect the operating status and real-time apparent power of the renewable energy converter, and the real-time apparent power of the renewable energy converter in operation is calculated, including: A fixed polling period is preset, and a timed acquisition trigger signal is generated based on the polling period; In response to the trigger signal, each of the new energy converters is polled in parallel through the communication interfaces of the multi-channel communication link to collect the operating status identifier and real-time apparent power data of each new energy converter; The collected operating status identifiers are judged, and the new energy converters that are in operation are selected. The real-time apparent power of the selected new energy converters is accumulated. When a single data collection fails or the data is abnormal, the most recently valid historical data will be used in subsequent calculations.

[0007] In one implementation, when a single data collection fails or the data is abnormal, the most recently valid historical data is used in subsequent calculations, including: For a new energy converter that fails to collect data in a single cycle, the valid data from the previous cycle of the new energy converter is used for calculation, and the number of consecutive failures is accumulated. At the same time, the next new energy converter is directly polled to ensure cycle stability. When the number of consecutive data acquisition failures of a single new energy converter reaches a preset threshold, it is determined that the communication of the new energy converter is interrupted, and it switches to the standby operation mode with fixed safety settings and issues a communication interruption alarm. When the new energy converter that was interrupted resumes continuous and normal communication, it will automatically switch back to the adaptive tuning mode and clear the corresponding alarms.

[0008] In one embodiment, after polling the new energy converter through each communication interface of the multi-channel communication link to collect the operating status and real-time apparent power of the new energy converter, and calculating the real-time apparent power of the new energy converter in the operating state, the method further includes: optimizing the polling frequency based on the rate of change of the real-time apparent power of the new energy converter, reducing the polling frequency for the new energy converter with stable power, and adaptively switching the polling cycle according to the change state of the real-time apparent power.

[0009] In one embodiment, based on the calculated real-time apparent power of the operating renewable energy converters, a total real-time apparent power is obtained by summing the data, and line loss correction is applied to the total real-time apparent power to obtain a corrected total real-time apparent power, including: The accumulated real-time apparent power of the aforementioned new energy converter is taken as the total real-time apparent power; When all the new energy converters in operation have complete communication, stable power, and the system is in a non-faulty state, the original loss coefficient is calculated based on the ratio of the measured power on the low-voltage side of the prefabricated substation to the total real-time apparent power. The original loss coefficients that meet the preset value range are smoothed to obtain real-time loss coefficients, which are then updated according to a preset update cycle. The total real-time apparent power is corrected using the real-time loss coefficient to obtain the corrected total real-time apparent power.

[0010] In one embodiment, the polling frequency is optimized based on the rate of change of the real-time apparent power of the renewable energy converter. The polling frequency is reduced for renewable energy converters with stable power, and the polling cycle is adaptively switched according to the change status of the real-time apparent power. This includes: Calculate the rate of change and absolute change of the real-time apparent power of the new energy converter in the current cycle and the previous cycle, and determine whether to update the real-time apparent power data according to the preset update conditions; For new energy converters that fail to meet the real-time apparent power data update conditions for multiple consecutive cycles, extend the polling cycle to reduce the communication frequency; When the real-time apparent power of the new energy converter meets the preset update conditions or reaches the preset forced calibration cycle, it returns to the polling cycle before the power was reduced.

[0011] In one embodiment, based on the calculated real-time apparent power of the operating renewable energy converters, the total real-time apparent power is obtained by summing the data, and line loss correction is applied to the total real-time apparent power. After obtaining the corrected total real-time apparent power, the method further includes: The three-phase current and three-phase voltage on the low-voltage side of the prefabricated substation are collected, and the measured apparent power is obtained based on the collection results. The measured apparent power is output according to the period synchronized with the polling period after the adaptive switching.

[0012] In one embodiment, an overcurrent protection setting is generated based on the corrected total real-time apparent power and the measured apparent power on the low-voltage side of the prefabricated substation, including: The smaller value between the corrected total real-time apparent power and the measured apparent power is selected as the real-time available capacity. The reference current is calculated based on the real-time available capacity, and an overcurrent protection setting is generated based on the reference current.

[0013] In one embodiment, a reference current is calculated based on the real-time available capacity, and an overcurrent protection setting is generated according to the reference current, including: The real-time available capacity is converted into a reference current that is adapted to the low-voltage side of the prefabricated substation; Based on the reference current, overcurrent protection settings corresponding to different thresholds and delay levels are configured respectively; The overcurrent protection settings generated in this round will be used for protection determination in the next polling cycle.

[0014] In one embodiment, after generating the overcurrent protection setting based on the corrected total real-time apparent power and the measured apparent power on the low-voltage side of the prefabricated substation, the method further includes: Fault discrimination feature quantities are constructed based on the collected three-phase current and three-phase voltage, and the fault judgment threshold is determined based on the real-time available capacity. Fault identification is completed by cross-verifying multiple fault identification feature quantities; If the fault is determined to be valid, the process of generating the overcurrent protection setting is blocked, the overcurrent protection setting before the fault occurs is used to perform the protection action, and the generation of the overcurrent protection setting is restored after the fault is cleared and the situation remains stable.

[0015] The above-mentioned technical solution of this application has at least the following beneficial technical effects: The technical solution of this application establishes a multi-channel communication link to acquire the operating status and real-time apparent power of each renewable energy converter in real time. By summarizing the real-time apparent power of the converters in operation and correcting for line losses, the calculation deviation caused by line transmission losses is eliminated, making the power benchmark more consistent with the actual operating conditions of the low-voltage side of the transformer. Then, based on the corrected total real-time apparent power and the measured apparent power on the low-voltage side, an overcurrent protection setting is adaptively generated, allowing the protection threshold to dynamically match the actual output of renewable energy. This significantly improves the protection sensitivity to early faults under low-output conditions, thereby improving the response capability and reliability of overcurrent protection and ensuring the safe and stable operation of the transformer. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating one implementation method of the adaptive overcurrent protection setting method for new energy prefabricated substations provided in this application. Figure 2 This is a flowchart illustrating a specific embodiment of the adaptive overcurrent protection setting method for new energy prefabricated substations provided in this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to specific embodiments and accompanying drawings. It should be understood that these descriptions are merely exemplary and not intended to limit the scope of this application. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of this application.

[0018] The embodiments described in this application are only some, not all, of the embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments described herein without inventive effort are within the scope of protection of this application.

[0019] New energy prefabricated substations are key hub equipment in new energy power plants such as photovoltaic and wind power plants. Overcurrent protection, as the most core main protection of prefabricated substations, directly undertakes the task of rapid detection and isolation of major electrical faults such as transformer short circuits, inter-turn faults, and lead faults. It is a key line of defense to ensure equipment safety, prevent fault expansion, and avoid burn-out accidents.

[0020] However, existing overcurrent protection systems in new energy prefabricated substations are generally set to fixed values ​​based on the transformer's rated capacity. During periods of low output from new energy sources such as photovoltaic and wind power, when early internal faults such as minor inter-turn short circuits or winding deformation occur, the change in fault current is far below the operating threshold of traditional overcurrent protection. This leads to insufficient timely response and a tendency for protection to fail to operate. Prolonged fault persistence can easily escalate into catastrophic accidents such as phase-to-phase short circuits and equipment burnout. Furthermore, traditional protection schemes do not dynamically calculate based on the real-time operating status and output data of the new energy converter. They lack multi-channel communication acquisition, total real-time apparent power aggregation, and line loss correction mechanisms. They cannot adaptively adjust protection settings according to actual operating capacity, making it difficult to match the characteristics of large output fluctuations in new energy sources. Consequently, protection sensitivity and operational safety cannot be simultaneously achieved.

[0021] To address the aforementioned technical issues, this application provides an adaptive overcurrent protection setting method for new energy prefabricated substations. The technical solution establishes a multi-channel communication link to acquire the real-time operating status and apparent power of each new energy converter. By summarizing the real-time apparent power of the operating converters and correcting for line losses, calculation deviations caused by line transmission losses are eliminated, making the power benchmark more closely match the actual operating conditions of the substation's low-voltage side. Then, based on the corrected total real-time apparent power and the measured apparent power on the low-voltage side, an overcurrent protection setting is adaptively generated, allowing the protection threshold to dynamically match the actual output of the new energy source. This significantly improves the protection sensitivity to early faults under low-output conditions, thereby enhancing the response capability and operational reliability of the overcurrent protection and ensuring the safe and stable operation of the transformer.

[0022] In one implementation, please refer to Figure 1 The adaptive overcurrent protection setting method for this new energy prefabricated substation includes the following steps: Step S1: Establish multi-channel communication links with each renewable energy converter. In this step, after the transformer substation's monitoring and protection device is powered on, it automatically identifies and configures multiple RS485 communication interfaces. A daisy-chain topology is used to establish high-speed communication links with each renewable energy converter, completing device address scanning and online status confirmation. A stable and reliable data transmission channel is achieved through Modbus RTU or GOOSE protocols. By constructing a multi-channel independent parallel communication architecture, a stable and high-speed hardware foundation is provided for the real-time acquisition of converter operating data, ensuring the real-time performance and integrity of data acquisition.

[0023] Step S2: Poll the renewable energy converters through each communication interface of the multi-channel communication link to collect their operating status and real-time apparent power, and calculate the real-time apparent power of the renewable energy converters in operation. In this step, the device polls all converters in parallel through each communication interface according to a preset cycle to obtain the operating status identifier and real-time apparent power, filters the operating status devices and calculates their power. For abnormal or failed data collection, historical valid values ​​are used as a fallback to maintain the continuity of calculation. Through standardized and orderly concurrent polling and data filtering, the collected data is ensured to be true and valid, providing an accurate and reliable data source for total power aggregation and improving the stability of basic calculations.

[0024] In one embodiment, step S2 includes the following specific steps: Step S21: Pre-set a fixed polling period and generate a timed acquisition trigger signal based on the polling period; In this step, a fixed standard polling period is set according to the single-interface communication time and multi-channel parallel processing capability. The device generates an acquisition trigger signal according to the period to synchronously start full-channel data acquisition. Through a unified and fixed timed trigger mechanism, the acquisition rhythm of the entire system is kept stable and controllable, providing a unified time base for subsequent power calculation and polling optimization.

[0025] Step S22: In response to the trigger signal, each renewable energy converter is polled in parallel through the communication interfaces of the multi-channel communication link to collect the operating status flag and real-time apparent power data of each renewable energy converter. In this step, each communication interface starts synchronously after receiving the trigger signal and independently polls the corresponding downstream converter in sequence to read the operating status flag and real-time apparent power raw data of each device. The multi-channel parallel polling method greatly shortens the overall acquisition time, improves the real-time performance of data, and meets the acquisition needs of rapid changes in renewable energy output.

[0026] Step S23: Judge the collected operating status identifiers, filter out the new energy converters that are in the operating state, and accumulate the real-time apparent power of the selected new energy converters. In this step, standby and operating equipment are distinguished based on the collected status identifiers. Only the real-time apparent power of the converters marked as operating is accumulated and calculated. By accurately filtering the effective operating equipment and completing the power accumulation, the high capacity caused by standby equipment is avoided, and the total real-time apparent power calculation result is guaranteed to be true and reliable.

[0027] Step S24: When a single acquisition fails or data is abnormal, the most recently acquired valid historical data is used in subsequent calculations. In this step, if a single acquisition times out, frames are lost, or data is abnormal, the valid data from the previous cycle of the converter is directly used in this round of calculations. This does not interrupt the polling process and maintains cycle stability. Through the abnormal data fallback replacement mechanism, calculation interruptions caused by single communication jitter are avoided, improving the system's robustness in complex electromagnetic environments.

[0028] In one embodiment, step S24 further includes the following specific steps: For a new energy converter that fails to acquire data in a single cycle, the valid data from the previous cycle of the new energy converter is used for calculation, and the number of consecutive failures is accumulated. Simultaneously, the next new energy converter is directly polled to ensure cycle stability. When the number of consecutive acquisition failures for a single new energy converter reaches a preset threshold, it is determined that the new energy converter's communication is interrupted, and it switches to a standby operation mode using a fixed safety setting and issues a communication interruption alarm. When the communication interrupted new energy converter resumes continuous normal communication, it automatically switches back to the adaptive setting mode and clears the corresponding alarm. In this specific step, the number of consecutive failures for acquisition failure devices is counted. Short-term failures use historical data, while long-term interruptions switch to a fixed safety setting and issue an alarm. After normal operation resumes, it automatically returns to the adaptive mode. Through hierarchical communication anomaly handling and standby setting switching, a seamless transition under communication failures is achieved, ensuring uninterrupted protection functions and safer system operation.

[0029] In one embodiment, the following steps are included after step S2: Step A1: Optimize the polling frequency based on the rate of change of real-time apparent power of the renewable energy converter. Reduce the polling frequency for renewable energy converters with stable power, and adaptively switch the polling cycle according to the changes in real-time apparent power. In this step, the stability of equipment operation is dynamically judged based on the fluctuation amplitude and trend of the converter's real-time apparent power. The polling frequency is automatically reduced for stable equipment, and regular polling is quickly restored when data changes. Forced calibration is performed at fixed intervals. Through adaptive polling optimization driven by power changes, the communication load is significantly reduced while ensuring calculation accuracy, thus improving the overall system operating efficiency.

[0030] In one embodiment, step A1 further includes the following specific steps: Step A11: Calculate the rate of change and absolute change of real-time apparent power of the new energy converter between the current cycle and the previous cycle, and determine whether to update the real-time apparent power data based on preset update conditions. In this step, the rate of change and absolute change of power in adjacent cycles are calculated for each unit, and the data is judged whether it needs to be updated based on preset thresholds. If the conditions are met, the data is refreshed; otherwise, the original value is used. Through precise change judgment logic, only effective fluctuations that affect the total capacity are updated, reducing meaningless communication and calculation overhead.

[0031] Step A12: For new energy converters that have not met the real-time apparent power data update conditions for multiple consecutive cycles, extend the polling cycle to reduce the communication frequency. In this step, devices that have been stable and have not been updated for multiple consecutive cycles are determined to be in a stable operating state. Their polling cycle is automatically extended and the number of communication requests is reduced. By reducing the polling frequency of stable devices, the bus communication pressure and device processing load are effectively reduced, and the long-term operating stability of the system is improved.

[0032] Step A13: When the real-time apparent power of the new energy converter meets the preset update conditions or reaches the preset mandatory calibration cycle, the polling cycle is restored to the state before the frequency reduction. In this step, once the equipment power fluctuates significantly or the mandatory calibration time is reached, the standard high-speed polling cycle is immediately restored. Through the dual mechanisms of fluctuation wake-up and timed calibration, both communication load reduction and data accuracy are taken into account, avoiding the cumulative errors caused by long-term frequency reduction.

[0033] Step S3: Based on the calculated real-time apparent power of the operating new energy converters, the total real-time apparent power is obtained by summarizing the data and then applying line loss correction to the total real-time apparent power. In this step, the real-time apparent power of the operating equipment is first summarized into the total real-time apparent power. Then, the dynamic loss coefficient is calculated and smoothed by combining it with the measured power on the low-voltage side of the transformer substation. The total power is corrected using the real-time loss coefficient to obtain the corrected total real-time apparent power. By combining power summarization and dynamic loss correction, line transmission and measurement deviations are eliminated, making the power benchmark more consistent with the actual on-site operating conditions and improving the accuracy of the setpoint calculation.

[0034] In one embodiment, step S3 includes the following specific steps: Step S31: The accumulated real-time apparent power of the new energy converter is used as the total real-time apparent power. In this step, the power value of the operating equipment accumulated in step S23 is directly used as the current total real-time apparent power of the system, providing a unified original calculation benchmark for subsequent loss correction. Through simple and direct summarization, the source of the benchmark power is clear and the calculation process is simple and reliable.

[0035] Step S32: When all operating new energy converters have complete communication, stable power, and the system is in a fault-free state, calculate the original loss coefficient based on the ratio of the measured power on the low-voltage side of the prefabricated substation to the total real-time apparent power. In this step, effective operating conditions with complete communication, stable power, and no faults are strictly selected. The original loss coefficient that truly reflects the line loss is calculated using the ratio of the measured power on the low-voltage side to the total communication power. Through screening of effective data under multiple conditions, the accuracy and reliability of the loss coefficient calculation results are ensured, and interference from transient and fault data is avoided.

[0036] Step S33: Smooth the original loss coefficients that meet the preset value range to obtain the real-time loss coefficients, and update them according to the preset update cycle; In this step, the original loss coefficients that fall into the reasonable range are subjected to exponential weighted smoothing, and the stable real-time loss coefficients are updated according to the fixed cycle. By combining smoothing filtering with periodic updates, instantaneous fluctuations are filtered out, so that the loss coefficient changes smoothly and reliably, and the long-term correction stability is improved.

[0037] Step S34: Correct the total real-time apparent power using the real-time loss factor to obtain the corrected total real-time apparent power. In this step, the total real-time apparent power is multiplied by the real-time loss factor to obtain the corrected total real-time apparent power that is closer to the actual input power on the low-voltage side of the transformer. Through dynamic loss correction, the calculation deviation caused by transmission loss is eliminated, making the power reference highly consistent with the actual operating conditions on site.

[0038] In one embodiment, the following steps are included after step S3: Step B1: Collect the three-phase current and three-phase voltage on the low-voltage side of the prefabricated substation, and obtain the measured apparent power based on the collection results. In this step, the three-phase current and three-phase voltage signals are collected in real time by the low-voltage side CT and PT. The measured apparent power that truly reflects the load condition is obtained through internal calculation of the device. The measured power is obtained by directly collecting and calculating local electrical quantities, providing an independent measured power source that is not affected by communication, and providing a reliable comparison basis for the dual power selection strategy.

[0039] Step B2: Output the measured apparent power according to the period synchronized with the polling period after adaptive switching. In this step, the output period of the measured apparent power is kept synchronized with the communication polling period to ensure that the two types of power data participate in the calculation under the same time base. Through the period synchronization mechanism, the calculation error caused by timing misalignment is avoided, and the accuracy of dual-power fusion judgment is improved.

[0040] Step S4: Generate overcurrent protection settings based on the corrected total real-time apparent power and the measured apparent power on the low-voltage side of the prefabricated substation. In this step, the smaller value between the corrected total power and the measured power is selected as the real-time available capacity. After converting the reference current, adaptive overcurrent protection settings are generated according to the protection coordination principle and take effect in the next cycle. By taking the smaller of the two power values ​​and generating dynamic settings, the protection threshold is matched with the actual output in real time, significantly improving fault sensitivity under low output conditions.

[0041] In one embodiment, step S4 includes the following specific steps: Step S41: Select the smaller value between the corrected total real-time apparent power and the measured apparent power as the real-time available capacity. In this step, both the communication-corrected power and the local measured power are used, and the minimum value of the two is taken as the real-time available capacity of the system. By adopting a conservative strategy of taking the smaller value, the overload risk under the power limit of the converter is avoided, and the protection sensitivity is guaranteed to follow the actual load, thus achieving a balance between safety and sensitivity.

[0042] Step S42: Obtain the reference current based on the real-time available capacity, and generate overcurrent protection settings based on the reference current. In this step, the real-time available capacity is converted into a reference current according to the rated voltage on the low-voltage side. Based on the reference current, three-stage overcurrent protection settings with different thresholds and delays are generated. Through standardized conversion from capacity to current and hierarchical setting configuration, the protection settings are accurately adapted to the current operating conditions, improving the selectivity and reliability of protection actions.

[0043] In one embodiment, step S42 further includes the following specific steps: converting the real-time available capacity into a reference current suitable for the low-voltage side of the prefabricated substation; configuring overcurrent protection settings corresponding to different thresholds and delay levels based on the reference current; and using the overcurrent protection settings generated in this round for protection determination in the next polling cycle. This facilitates continuous adaptive updating of protection settings, ensuring optimal protection performance throughout the entire process.

[0044] In one embodiment, the following steps are included after step S4: Step C1: Construct fault discrimination feature quantities based on the collected three-phase current and three-phase voltage, and determine the fault judgment threshold based on the real-time available capacity. In this step, fault feature quantities such as negative sequence current, zero sequence voltage, and current mutation are calculated using three-phase voltage and current, and the fault discrimination threshold is dynamically set based on the real-time available capacity. Through the construction of multiple feature quantities and dynamic threshold setting, the fault judgment criteria are adapted to the current output level, thereby improving the accuracy and consistency of fault identification.

[0045] Step C2: Fault identification is completed by cross-verifying multiple fault discrimination features. In this step, features such as negative sequence, zero sequence, and current mutation are cross-fused and their reliability is verified to distinguish between real faults and power fluctuations and measurement interference. Through cross-verification of multiple criteria and reliability evaluation, the probability of misjudgment is effectively reduced and the reliability of fault identification is improved.

[0046] Step C3: If a valid fault is identified, the overcurrent protection setting generation process is blocked. The overcurrent protection setting before the fault occurs is used to perform the protection action, and the generation of the overcurrent protection setting is restored after the fault is cleared and the system remains stable. In this step, the setting update is immediately blocked when a valid fault is identified, ensuring reliable operation of the setting before the fault. The adaptive setting is automatically restored after the fault is cleared and stabilized. Through the fault-state blocking and reset mechanism, the calculation benchmark is prevented from being contaminated by fault data, ensuring reliable protection operation during the fault period and improving the overall safety protection capability of the system.

[0047] The technical solution of this application will be described in detail below with reference to specific embodiments. Please refer to... Figure 2 This embodiment provides a method for setting adaptive overcurrent protection settings in a new energy prefabricated substation, including the following specific steps: Step 1: Establish a multi-channel high-speed communication network. After the transformer substation's monitoring and protection device is powered on, it automatically identifies and establishes high-speed communication links with each new energy converter. This method adopts an active polling mode, where the monitoring and protection device acts as the master station, sending read requests to each converter at fixed intervals to obtain its operating data in real time. The specific implementation process is as follows: 1.1: Hardware interface configuration. The new energy transformer substation's monitoring and protection device is generally configured with 10 independent RS485 communication interfaces. Some of these interfaces (approximately 6-7) can be used to connect to the converters, while the remaining interfaces are reserved for communication with other intelligent devices within the transformer substation (such as environmental monitoring systems, electricity meters, etc.). 1.2: Communication method. Each RS485 interface uses a daisy-chain topology to connect multiple converters, i.e., multiple devices are connected in series on one bus, with a 120Ω terminating resistor connected at the end. The communication protocol uses Modbus RTU or GOOSE, with a baud rate set to 115200bps or higher to ensure high-speed and real-time data transmission. Each converter is assigned a different communication address, and the protection device distinguishes different devices by address. 1.3: Device Address Management. After power-on, the monitoring and protection device automatically scans and identifies the online devices under each interface, obtaining the number Ni of converters under each interface, with the total number N = ΣNi. Based on actual engineering experience, a single transformer substation typically connects to 1 to 18 converters. When using 6 to 7 RS485 interfaces, the average number of converters under each interface does not exceed 3.

[0048] Step 2: Polling Period Setting. Based on the above hardware configuration and engineering experience, this method adopts a fixed polling period, requiring no complex calculations. The specific implementation process is as follows: 2.1: Polling Period Determination. Considering that each RS485 interface can connect up to 3 converters, the communication time of a single converter can be controlled within 10ms (including sending requests, waiting for responses, and data parsing). Adding interface switching and software processing overhead, the maximum time for a single interface to complete one round of polling (i.e., communicating with its 3 connected devices sequentially) is approximately 30ms. Since this method uses a multi-channel concurrent mechanism, with 6-7 RS485 interfaces working independently in parallel, the total time for the system to complete one round of complete data acquisition from all converters is determined by the interface with the longest time, i.e., not exceeding 30ms. To ensure stability and allow for margin, this method sets a fixed polling period T_poll = 50ms. The protection device starts a round of data acquisition from all converters in parallel every 50ms. 2.2: Data Validity Judgment. For each converter, the monitoring and protection device records its most recently successfully read data. If a converter fails to read data in a single polling cycle, the previously successful data is used in the current round of calculation, and the number of consecutive failures N_fail_i is recorded.

[0049] Step 3: Multi-channel concurrent polling. The specific implementation process is as follows: 3.1: Concurrent polling mechanism. Each RS485 interface of the protection device works independently and in parallel. Each interface polls the converters connected to it in sequence. All interfaces poll simultaneously without interfering with each other. At the start of each round of polling, each interface starts synchronously and completes its own task within a 50ms cycle. 3.2: Data acquisition content. Each converter reads at least the following core information as the basis for subsequent capacity calculation and setting adjustment: (1) Real-time apparent power S_i: The core benchmark for protection setting calculation. It directly reflects the real-time output of the converter under the current operating conditions. The protection device dynamically calculates the overcurrent protection setting based on the sum of the real-time apparent power of all converters to ensure that the setting always follows the actual load change. (2) Standby / operation status flag: Indicates whether the converter is in standby or normal operation. The output of the converter in standby state is zero and should not be included in the total capacity calculation. Otherwise, it will lead to an overestimation of capacity and an overestimation of protection setting, causing safety hazards. The corresponding real-time apparent power is only included in the total capacity calculation when the flag is in the "running" state.

[0050] Step 4: Communication Anomaly Handling. Given the reliability requirements of industrial-grade RS485 communication, this method employs a simple hierarchical processing mechanism. The specific implementation process is as follows: 4.1: Single Read Failure Handling. If a converter fails to read data in a single polling cycle, then: maintain the data from the last successful read of that converter in the current round of calculation; increment the number of consecutive failures N_fail_i by 1; without waiting for a retry, immediately poll the next device to ensure cycle stability. 4.2: Single Unit Interruption Handling. If a converter experiences 3 consecutive read failures (i.e., no valid data for 150ms), then the device is considered to have a communication interruption: the protection device immediately switches to standby mode, using fixed safety settings (see Step 4.4 for details). A "Converter Communication Interruption" alarm is issued; the device status is continuously monitored. 4.3: Recovery Mechanism. When the interrupted converter resumes normal communication (3 consecutive successful reads): the protection device automatically switches back to primary mode, restoring adaptive tuning; the interruption alarm for that device is cleared. 4.4: Standby Mode Settings. In standby mode, the protection device adopts fixed safety settings (which can be set according to the site conditions): instantaneous fast-break: I_set1 = 4 × I_N; short-delay overcurrent section: I_set2 = 2 × I_N (delay 0.5s); long-delay overcurrent section: I_set3 = 1.2 × I_N (delay 5s); where I_N is the rated current of the transformer.

[0051] Step 5: Polling optimization based on real-time power change rate. To further reduce communication load and improve system efficiency, this method introduces an optimization strategy based on power change rate on the basis of the fixed polling period (50ms) set in step 2, and automatically reduces the polling frequency for converters with stable power. The specific implementation process is as follows: 5.1: Calculation of real-time power change rate. For each converter, read the real-time apparent power S_new in this round and the recorded value S_old in the previous round, and process it according to the following rules: (1) If S_old = 0 (i.e., the device starts from the shutdown state); do not calculate the change rate; directly mark the device as needing to update data in this round; reset the continuous stable count of the device. (2) If S_old > 0; calculate the change rate: ΔS = |S_new - S_old| / S_old × 100%; calculate the absolute change: ΔS_abs = |S_new - S_old|. The above parameters are used for subsequent selective updates and frequency reduction judgment. 5.2: Selective update strategy. Based on the calculated rate of change ΔS and absolute change ΔS_abs, combined with the current total capacity S_total, determine whether the device data needs to be updated: Update conditions (update if any one condition is met, threshold is adjustable, see 5.4 for details): Condition A (significant impact): ΔS ≥ 5% and ΔS_abs ≥ 0.4% × S_total, indicating that the device change has a significant impact on the total capacity (≥ 0.4%), and needs to be updated; Condition B (drastic change): ΔS ≥ 10%, indicating that the device itself has drastic changes, and should be updated even if the absolute amount is small; Condition C (excessive absolute amount): ΔS_abs ≥ 0.8% × S_total, indicating that the absolute change has a significant impact on the total capacity, and should be updated regardless of the percentage. If any update condition is met, the device data is updated to S_new; otherwise, the power change is considered to have a negligible impact on the total capacity, and the device data is not updated in this round, continuing to use S_old for total capacity calculation. Step 5.3: Adaptive adjustment of polling frequency. To reduce communication load, the polling frequency is automatically reduced for converters with consistently stable power: 5.3.1 Entering Reduced Frequency State. If a converter fails to meet any of the update conditions in 5.2 for five consecutive standard polling cycles (i.e., 250ms), the device is determined to have entered a "stable operating state." The protection device will extend its polling cycle from the standard 50ms to 200ms. That is, in every subsequent four standard polling cycles, the protection device will only initiate a communication request to the device in the first cycle, and in the remaining three cycles, it will directly use the data from the most recently successfully read data for calculation, without sending any further requests. 5.3.2 Exiting Reduced Frequency State. The device will immediately resume normal 50ms polling when any of the following conditions are met: the current calculation meets any of the update conditions in 5.2 (significant power change), two consecutive read failures occur during reduced frequency polling, and normal polling is forcibly resumed every 30 seconds for data calibration to prevent data deviation caused by prolonged reduced frequency.5.3.3 Data Processing During Frequency Reduction. For converters in frequency reduction mode, during non-polling cycles, the most recently successfully read data continues to be used in the total capacity calculation. 5.4: Parameter Rationality Analysis, as shown in Table 1.

[0052] Table 1 Parameter Rationality Sub-Table Layered Progressive Adaptive Threshold Tuning Strategy: The threshold parameters in this strategy are not fixed, but rather a layered progressive adaptive tuning mechanism is introduced to dynamically optimize based on the number of devices in the actual engineering scenario, achieving the optimal balance between accuracy and efficiency. 5.4.1: Tuning Basis. Device Quantity Dimension: The more devices there are, the greater the risk of small changes from multiple devices accumulating into a large error. Let ε be the upper limit of the unupdated change of a single device, then the maximum possible cumulative error E_max = N × ε. To ensure that E_max ≤ 5%, the threshold ε needs to be tightened accordingly as N increases. Capacity Weight Dimension: The larger the capacity proportion of a single device, the more significant the direct impact of its change on the total capacity. For large-capacity devices, the threshold can be appropriately relaxed to reduce unnecessary communication; for small-capacity devices, the threshold needs to be tightened to prevent multiple devices from accumulating. Tuning Objective: Through layered progressive tuning, the maximum possible cumulative error of the system is controlled within the acceptable engineering range of 3%~5%, while maximizing communication efficiency. 5.4.2: Threshold Optimization Reference Table, as shown in Table 2.

[0053] Table 2 Threshold Optimization Reference Table 5.4.3: Engineering Implementation Recommendations. Before on-site commissioning, select the corresponding threshold level based on the actual number of devices. After stable operation, further fine-tuning and optimization can be performed based on historical data statistical analysis. Key principle: While ensuring accuracy, relax the threshold as much as possible to reduce communication load. Step 5.5: Typical Scenario Example Verification. To verify the effectiveness and parameter adjustability of this strategy, four typical scenarios are selected for testing: 5.5.1: Scenario Definition, as shown in Table 3.

[0054] Table 3 Scene Definition Table 5.5.2: Threshold settings for each scenario are shown in Table 4.

[0055] Table 4 Threshold Settings for Various Scenarios 5.5.3: Cumulative Error Control Effect Analysis. Scenario A (40 units, threshold 0.25%): Maximum unupdated change per unit: 0.24% (just below the threshold); Cumulative error for 40 units: 40 × 0.24% = 9.6%. In this case, the cumulative error has seriously exceeded the limit, but this strategy controls it through the following mechanisms: Condition B (ΔS≥10%) will capture devices with drastic changes; Condition C (0.5%) will capture devices with absolute values ​​exceeding the limit; in actual operation, the probability of all devices simultaneously changing continuously at a rate just below the threshold is extremely low; if there is a systematic and continuous change, the 30-second forced calibration cycle will correct it in time. Scenario B (10 units, threshold 0.4%): Maximum unupdated change per unit: 0.39%; Cumulative error for 10 units: 3.9%, close to but not exceeding 4%, within an acceptable range. Combined with the 30-second forced calibration, long-term operational accuracy can be ensured. Scenario C (3 units, threshold 0.6%): Maximum unupdated change per unit: 0.59%, cumulative error for 3 units: 1.77%, far below the control target of 3%, with sufficient safety margin. Scenario D (1 unit, threshold 0.8%): No cumulative risk, only time accumulation; frequency reduction period within 200ms; calculated at a change rate of 10% / second, maximum error 2%; threshold 0.8% ensures most changes are captured in time. 5.5.4: Communication load optimization effect. Taking Scenario B (10 units) as an example, if 8 units enter a stable state, as shown in Table 5.

[0056] Table 5 shows the operational status of scenario B (10 units) as an example. Optimization Results: Original request count per cycle: 10 times; Optimized request count per cycle: 4 times, communication load reduced by 60%. Scenario A (40 servers): If 30 servers reach a stable state: Original: 40 times / cycle; Optimized: 10 times (normal) + 30 / 4 ≈ 7.5 times ≈ 17.5 times / cycle, communication load reduced by 56%. Scenario C (3 servers): If 2 servers reach a stable state: Original: 3 times / cycle; Optimized: 1 time + 2 / 4 = 1.5 times / cycle, communication load reduced by 50%. 5.5.5: Summary of Strategy Effectiveness. This optimization strategy achieves the following goals through adjustable parameter design: Controllable accuracy: By setting a threshold based on the number of devices, the maximum possible cumulative error is controlled within the range of 3% to 6%; Optimal efficiency: Smooth device frequency reduction and polling reduces communication load by 50% to 60% in typical scenarios; Flexible adaptation: The same strategy can be adapted to various scenarios from 1 to 40 devices through parameter adjustment; Safety redundancy: 30-second forced calibration and cumulative error monitoring (step 5.4) provide dual protection.

[0057] Step 6: Real-time Capacity Calculation. The specific implementation process is as follows: 6.1: Total Real-time Apparent Power Calculation. For the valid data of this round (including the old data that remains unchanged), calculate the total real-time apparent power of all converters in operation: S_total(t) = ΣS_i(t), (i is all converters in the "operating" state), where: S_i(t) is the real-time apparent power of the i-th converter in this round (it may be a newly read value or the value maintained from the previous time). Only when the converter's status flag is "operating" is its power value included in the total capacity; the power value of converters in the "standby" or "stop" state is considered to be 0. 6.2: Line Loss Correction. 6.2.1: The correction formula is as follows: The sum of real-time apparent power S_total(t) reported by the converters is the theoretical output value. There are line losses when the actual power is transmitted to the low-voltage side of the transformer. To ensure that S_total(t) and the measured power S_meas(t) are compared on the same benchmark, a loss correction is required: S_comm_corrected(t) = S_total(t) × η(t), where η(t) is the real-time loss coefficient. 6.2.2: Principle of Dynamic Measurement Method. This method uses dynamic measurement to determine η(t), that is, it uses two types of data simultaneously collected by the protection device for real-time calculation: the sum of real-time apparent power reported by the converter S_total(t), and the real-time apparent power S_meas(t) measured on the low-voltage side of the transformer substation. The ratio of the two is the real-time comprehensive loss coefficient: η_raw(t) = S_meas(t) / S_total(t). This coefficient comprehensively reflects factors such as line loss, converter efficiency, and measurement error. 6.2.3: Data Validity Conditions. η_raw(t) is adopted to update η(t) only if all of the following conditions are met: Complete communication: In this round of calculation, the power values ​​of all converters in the "running" state participating in the S_total(t) calculation are all newly read and successfully obtained data. This means that no device is in a communication failure state (using old data), and no device is not polled due to frequency reduction optimization (for frequency reduction devices, this condition is only met in the specific period during which they are polled). Stable power: The total power change rate is <5% within 10 consecutive minutes (to avoid transient process interference); Non-faulty state: The system is not in a faulty state (determined in step 9); Reasonable data range: 0.95≤η_raw(t)≤1.0; Data range explanation: The upper limit of 1.0 is based on energy conservation (the measured power cannot be greater than the theoretical output), and the lower limit of 0.95 is based on experience in new energy engineering projects (normal line loss does not exceed 5%, and values ​​below this indicate data abnormality or equipment failure). 6.2.4: Data smoothing processing. To filter out instantaneous fluctuations, an exponentially weighted moving average is applied to η_raw(t) that meets the conditions: η(t) = β × η_raw_avg + ( )×η(t-1), where: η_raw_avg is the average of all valid η_raw in the past 60 minutes, β is the smoothing coefficient, taken as β = 0.1, η(t-1) is the loss coefficient of the previous period, and the initial value of η(0) is 0.96. Parameter description: β=0.1 can effectively smooth fluctuations while maintaining the ability to track trends. The initial value of 0.96 is the median value of conventional engineering scenarios, covering most cases. 6.2.5: Update cycle. The updated value of η(t) is calculated every 60 minutes. Each time it is updated, the average of all η_raw that meet the conditions in the past 60 minutes is taken as input. The updated η(t) is used for loss correction in the following 60 minutes. Cycle description: The 60-minute update is based on the physical characteristics of the loss coefficient - the line loss is mainly determined by the cable impedance, which changes slowly (hourly level), and there is enough time to collect sufficient samples to avoid frequent calculations. 6.2.6: Anomaly handling. If no data meeting the conditions is available for 24 consecutive hours, the last valid η value remains unchanged, and a "loss coefficient has not been updated for a long time" prompt is issued; if η_raw(t) continuously exceeds the range of 0.95~1.0, a "data abnormality" alarm is issued; during communication interruption, dynamic updates are paused, and the last valid η value is used. 6.2.7: Corrected output: S_comm_corrected(t) = S_total(t) × η(t) is used in step 8.1 to compare with S_meas(t) and take the smaller value. Step 7: Electrical quantity sampling: 7.1: Sampling data acquisition: The protection device collects three-phase currents IA, IB, IC and three-phase voltages UA, UB, UC in real time through current transformers and voltage transformers (or direct voltage sampling) connected to the low-voltage side of the transformer. The built-in measurement function module of the device can directly output the apparent power S_meas. 7.2: Measured power output: Every 50ms (synchronized with the polling cycle), the protection device reads the real-time apparent power S_meas(t) calculated by the measurement module, which is used for comparison calculation in subsequent steps.

[0058] Step 8: Protection setting calculation and upper limit constraint. The protection setting takes the smaller value of the converter side capacity and the measured capacity on the transformer substation side as the basis to ensure that the transformer is not overloaded under any circumstances. The specific implementation process is as follows: 8.1: Determination of the real-time available capacity on the source side: The protection device obtains two types of data simultaneously: (1) Converter side capacity: The sum of the smoothed apparent powers read from each converter, and after the loss correction in Step 6.2, is S_comm_corrected(t); (2) Measured capacity on the transformer substation side: The real-time apparent power S_meas(t) obtained from Step 7.2; Take the smaller value of the two as the real-time available capacity on the source side: S_source(t) = min(S_comm_corrected(t), S_meas(t)). The basis for this minimum value strategy: When the converter limits power but the actual output exceeds the standard (S_meas > S_comm), limit according to the converter instruction to prevent the transformer from being overloaded; When the actual output is lower than the converter capacity (S_meas < S_comm), track according to the actual load to maintain the protection sensitivity; In any case, ensure that the transformer does not operate beyond the limit. 8.2: Calculation of the reference current: Convert the real-time available capacity on the source side to the reference current under the rated voltage of the low-voltage side of the transformer substation: I_base(t) = S_source(t) / (√3 × U_N), where U_N is the rated voltage of the low-voltage side of the transformer substation. 8.3: Adaptive setting based on communication data: According to the principle of relay protection coordination, set the three-stage overcurrent protection (which can be set according to the on-site situation): Instantaneous cut-off section: I_set1(t) = 4 × I_base(t), Short-time delay overcurrent section: I_set2(t) = 2 × I_base(t) (delay 0.5 s), Long-time delay overcurrent section: I_set3(t) = 1 × I_base(t) (delay 5 s). 8.4: The setting takes effect immediately: The protection setting calculated in this round takes effect immediately at the start of the next polling cycle.

[0059] Step 9: Fault discrimination and fault state handling. The specific implementation process is as follows: 9.1: Obtaining fault characteristic quantities. Some of the fault characteristic quantities required by this method can be directly provided by the protection and measurement control device of the transformer substation, and some need to be provided after the device adds calculation functions. Specifically as follows: 9.1.1: Negative sequence current : Modern protection and measurement control devices of transformer substations are all built-in with symmetrical component calculation functions and can output the negative sequence current value in real time. This method directly calls this calculation result and records it as . 9.1.2: Zero-sequence voltage : The device calculates the zero-sequence component through the three-phase voltages in real time and can be directly called, and is recorded as 9.1.3: Current mutation ΔI (this function needs to be added to the device). Current mutation reflects the instantaneous change of current and is an important characteristic quantity for detecting the moment of fault occurrence. Most measurement and control devices do not directly provide this parameter. This method requires the following calculation function to be added to the device: (1) Calculation formula: The current mutation is calculated by the cycle differential method, that is, the absolute value of the difference between the current value at the current sampling point and the current value one cycle ago (20ms): ΔI_A(t)=|i_A(t)-i_A(t - T)|, ΔI_B(t)=|i_B(t)-i_B(t - T)|, ΔI_C(t)=|i_C(t)-i_C(t - T)|, where: i_A(t), i_B(t), i_C(t) are the instantaneous values ​​of the three-phase current at the current sampling point, T is one power frequency cycle, taken as 20ms. The sampling frequency of the device should not be less than 48 points per cycle to ensure the calculation accuracy. (2) Composite output: The maximum value of the three-phase current abrupt change is taken as the final output: This value is calculated once per sampling point and output in real time for fault identification. (3) Physical meaning: During steady-state operation: the current waveforms of adjacent cycles are basically consistent, and ΔI(t) is close to 0; During power regulation: when the converter adjusts the output according to the dispatching instructions, the current amplitude will change stepwise, and ΔI(t) will increase instantaneously, but the waveform remains sinusoidal and the three phases are symmetrical; When a fault occurs: the current waveform is distorted (such as short circuit, grounding, etc.), ΔI(t) increases instantaneously, and is often accompanied by waveform distortion, negative sequence component or zero sequence component. 9.1.4: Other available characteristic quantities: According to the device capacity, the following can also be selected: effective value of phase current, positive sequence current, zero sequence current (if configured), harmonic content, etc., to assist in fault identification. 9.2: Fault criteria and adaptive setting: 9.2.1 Dynamic adjustment of negative sequence current setting: The negative sequence current setting is dynamically adjusted according to the current actual capacity to maintain the consistency of sensitivity under different output levels: Where: I_base(t) is the base current calculated in step 8.2 (based on real-time capacity). The negative order reliability coefficient ranges from 0.1 to 0.2, and is preferred in this embodiment. Physical meaning: When the negative sequence current reaches 15% of the current reference current, it is considered to have significant asymmetrical fault characteristics. 9.2.2: Zero-sequence voltage setting. The zero-sequence voltage setting uses a fixed value, reflecting the system grounding method: , where U_N is the rated phase voltage. 9.2.3: Current mutation rate setting: The current mutation rate setting is also dynamically adjusted based on the actual capacity: ΔI_set(t)=K_Δ×I_base(t), where K_Δ takes 2~3, and in this embodiment, K_Δ is preferably 2.5. Physical meaning: When the current mutation rate reaches 2.5 times the current reference current, it is considered that a significant current mutation has occurred, which may be a fault. 9.3: Fault identification and reliability verification: In order to distinguish between real faults and disturbances, measurement anomalies, etc., this method introduces a multi-criteria fusion and reliability verification mechanism. 9.3.1: Preliminary fault identification: If any of the following conditions are met, the fault reliability verification process is initiated: Condition A (asymmetric fault): And lasting for more than 5ms; Condition B (ground fault): And lasts for more than 5ms; Condition C (symmetrical fault): ΔI(t)>ΔI_set(t) and the three-phase current changes synchronously. 9.3.2: Fault credibility verification: When the initial judgment is triggered, the following verification logic is started to calculate the fault credibility coefficient C_fault: (1) Continuous consistency verification: The characteristic quantity lasts for more than the threshold for a period of time t_duration; t_duration≥ 10ms (half cycle), credibility +0.3 t_duration ≥ 20ms (integers), confidence +0.5; (2) Multi-feature cross-validation: when multiple criteria are met simultaneously; when 2 criteria are met, confidence +0.3; when 3 criteria are met, confidence +0.5. (3) Consistency verification of change trend: the change trend of the feature quantity is consistent with the expected fault characteristics. The negative sequence current and the current mutation rise rapidly at the same time, and the confidence level is +0.2; the zero sequence voltage and the negative sequence current appear at the same time, and the confidence level is +0.2 (grounding fault characteristics). (4) Verification by contradiction: If the following situations exist, the confidence level decreases, the characteristic quantity recovers after rapid fluctuation, which may be interference, and the confidence level is -0.3; only a single sampling point exceeds the standard, and the confidence level is -0.2. 9.3.3: Fault judgment: Comprehensive confidence level coefficient C_fault (value range 0~1): If C_fault ≥ 0.7, it is judged as a high confidence level fault and immediately enters the fault state processing flow; if 0.3 ≤ C_fault<0.7, it is judged as a suspicious event and the observation time is extended (for another 20ms). If it is still maintained, it is upgraded to a fault; if C_fault<0.3, it is judged as a disturbance or interference and does not enter the fault state. 9.4: Fault Handling: When a high-confidence fault is identified, perform the following operations: 9.4.1 Blocking Capacity Update: Immediately stop the capacity calculation and setting update in steps 6-8, maintaining the latest setting value before the fault occurred. This prevents fault data from contaminating the capacity reference. 9.4.2: Protection Logic Retention and Action: Continue using the blocking setting value before the fault occurred for real-time protection judgment. The protection device monitors the three-phase current in real time. If the fault current (such as phase current, negative sequence current, etc.) exceeds the corresponding blocking value (such as I_set1, ... (etc.) And if the duration reaches the set delay, the protection device immediately issues a trip command to clear the fault and ensure system safety. 9.4.3: Recovery after fault clearance: When the following conditions are met simultaneously, the fault state is automatically exited: all fault characteristic quantities are restored to below the threshold value, the above state lasts for more than 1 second, and after recovery, the normal capacity calculation and setting update in step 6 are restarted. 9.4.4: Anti-maloperation mechanism: If the fault state is re-entered within 100ms after exiting, then: the duration of the fault state is extended, a "repeated fault" alarm is issued, and the event is recorded for operation and maintenance analysis.

[0060] This application aims to protect an adaptive overcurrent protection setting method for a new energy prefabricated substation. The technical solution establishes a multi-channel communication link to acquire the real-time operating status and apparent power of each new energy converter. By summarizing the real-time apparent power of the operating converters and correcting for line losses, the calculation deviation caused by line transmission losses is eliminated, making the power benchmark more closely match the actual operating conditions of the low-voltage side of the prefabricated substation. Then, based on the corrected total real-time apparent power and the measured apparent power on the low-voltage side, an overcurrent protection setting is adaptively generated, allowing the protection threshold to dynamically match the actual output of the new energy source. This significantly improves the protection sensitivity to early faults under low-output conditions, thereby enhancing the response capability and operational reliability of the overcurrent protection and ensuring the safe and stable operation of the transformer.

[0061] It should be understood that the specific embodiments described above are merely illustrative or explanatory of the principles of this application and do not constitute a limitation thereof. Therefore, any modifications, equivalent substitutions, improvements, etc., made without departing from the spirit and scope of this application should be included within the protection scope of this application. Furthermore, the appended claims are intended to cover all variations and modifications falling within the scope and boundaries of the appended claims, or equivalent forms of such scope and boundaries.

Claims

1. A method for setting adaptive overcurrent protection settings in a new energy prefabricated substation, characterized in that, include: Establish multi-channel communication links with various new energy converters; The new energy converter is polled through each communication interface of the multi-channel communication link to collect the operating status and real-time apparent power of the new energy converter, and the real-time apparent power of the new energy converter in the operating state is calculated. Based on the calculated real-time apparent power of the new energy converters in operation, the total real-time apparent power is obtained by summing them up, and the total real-time apparent power is corrected for line loss to obtain the corrected total real-time apparent power. Based on the corrected total real-time apparent power and the measured apparent power on the low-voltage side of the prefabricated substation, an overcurrent protection setting is generated.

2. The adaptive overcurrent protection setting method for new energy prefabricated substations according to claim 1, characterized in that, The renewable energy converter is polled through each communication interface of the multi-channel communication link to collect its operating status and real-time apparent power. The real-time apparent power of the renewable energy converter in operation is then calculated, including: A fixed polling period is preset, and a timed acquisition trigger signal is generated based on the polling period; In response to the trigger signal, each of the new energy converters is polled in parallel through the communication interfaces of the multi-channel communication link to collect the operating status identifier and real-time apparent power data of each new energy converter; The collected operating status identifiers are judged, and the new energy converters that are in operation are selected. The real-time apparent power of the selected new energy converters is accumulated. When a single data collection fails or the data is abnormal, the most recently valid historical data will be used in subsequent calculations.

3. The adaptive overcurrent protection setting method for new energy prefabricated substations according to claim 2, characterized in that, When a single data collection fails or the data is abnormal, the most recently valid historical data will be used in subsequent calculations, including: For a new energy converter that fails to collect data in a single cycle, the valid data from the previous cycle of the new energy converter is used for calculation, and the number of consecutive failures is accumulated. At the same time, the next new energy converter is directly polled to ensure cycle stability. When the number of consecutive data acquisition failures of a single new energy converter reaches a preset threshold, it is determined that the communication of the new energy converter is interrupted, and it switches to the standby operation mode with fixed safety settings and issues a communication interruption alarm. When the new energy converter that was interrupted resumes continuous and normal communication, it will automatically switch back to the adaptive tuning mode and clear the corresponding alarms.

4. The adaptive overcurrent protection setting method for new energy prefabricated substations according to claim 3, characterized in that, The new energy converter is polled through each communication interface of the multi-channel communication link to collect its operating status and real-time apparent power. After calculating the real-time apparent power of the new energy converter in operation, the method further includes: optimizing the polling frequency based on the rate of change of the real-time apparent power of the new energy converter; reducing the polling frequency for new energy converters with stable power; and adaptively switching the polling cycle according to the change status of the real-time apparent power.

5. The adaptive overcurrent protection setting method for new energy prefabricated substations according to claim 2, characterized in that, Based on the calculated real-time apparent power of the operating renewable energy converters, the total real-time apparent power is obtained by summing the data. Line loss correction is then applied to the total real-time apparent power to obtain the corrected total real-time apparent power, which includes: The accumulated real-time apparent power of the aforementioned new energy converter is taken as the total real-time apparent power; When all the new energy converters in operation have complete communication, stable power, and the system is in a non-faulty state, the original loss coefficient is calculated based on the ratio of the measured power on the low-voltage side of the prefabricated substation to the total real-time apparent power. The original loss coefficients that meet the preset value range are smoothed to obtain real-time loss coefficients, which are then updated according to a preset update cycle. The total real-time apparent power is corrected using the real-time loss coefficient to obtain the corrected total real-time apparent power.

6. The adaptive overcurrent protection setting method for new energy prefabricated substations according to claim 5, characterized in that, The polling frequency is optimized based on the rate of change of the real-time apparent power of the renewable energy converter. For renewable energy converters with stable power, the polling frequency is reduced, and the polling cycle is adaptively switched according to the change status of the real-time apparent power, including: Calculate the rate of change and absolute change of the real-time apparent power of the new energy converter in the current cycle and the previous cycle, and determine whether to update the real-time apparent power data according to the preset update conditions; For new energy converters that fail to meet the real-time apparent power data update conditions for multiple consecutive cycles, extend the polling cycle to reduce the communication frequency; When the real-time apparent power of the new energy converter meets the preset update conditions or reaches the preset forced calibration cycle, it returns to the polling cycle before the power was reduced.

7. The adaptive overcurrent protection setting method for new energy prefabricated substations according to claim 6, characterized in that, Based on the calculated real-time apparent power of the operating renewable energy converters, the total real-time apparent power is obtained by summing the data. Line loss corrections are then applied to the total real-time apparent power to obtain the corrected total real-time apparent power. The process then includes: The three-phase current and three-phase voltage on the low-voltage side of the prefabricated substation are collected, and the measured apparent power is obtained based on the collection results. The measured apparent power is output according to the period synchronized with the polling period after the adaptive switching.

8. The adaptive overcurrent protection setting method for new energy prefabricated substations according to claim 7, characterized in that, Based on the corrected total real-time apparent power and the measured apparent power on the low-voltage side of the prefabricated substation, overcurrent protection settings are generated, including: The smaller value between the corrected total real-time apparent power and the measured apparent power is selected as the real-time available capacity. The reference current is calculated based on the real-time available capacity, and an overcurrent protection setting is generated based on the reference current.

9. The adaptive overcurrent protection setting method for new energy prefabricated substations according to claim 8, characterized in that, The reference current is calculated based on the real-time available capacity, and an overcurrent protection setting is generated based on the reference current, including: The real-time available capacity is converted into a reference current that is adapted to the low-voltage side of the prefabricated substation; Based on the reference current, overcurrent protection settings corresponding to different thresholds and delay levels are configured respectively; The overcurrent protection settings generated in this round will be used for protection determination in the next polling cycle.

10. The adaptive overcurrent protection setting method for new energy prefabricated substations according to claim 9, characterized in that, Based on the corrected total real-time apparent power and the measured apparent power on the low-voltage side of the prefabricated substation, after generating the overcurrent protection setting, the following steps are also included: Fault discrimination feature quantities are constructed based on the collected three-phase current and three-phase voltage, and the fault judgment threshold is determined based on the real-time available capacity. Fault identification is completed by cross-verifying multiple fault identification feature quantities; If the fault is determined to be valid, the process of generating the overcurrent protection setting is blocked, the overcurrent protection setting before the fault occurs is used to perform the protection action, and the generation of the overcurrent protection setting is restored after the fault is cleared and the situation remains stable.

Images