CT (Computed Tomography) countercurrent prevention system and method
By combining multi-protocol communication and deep learning models, high-precision reverse current protection of photovoltaic inverters in complex electromagnetic environments is achieved, which solves the shortcomings of existing systems in terms of communication reliability and response efficiency, and ensures the reliability and rapid response of the system in distributed layout and strong interference environments.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-08
- Publication Date
- 2026-03-10
AI Technical Summary
Existing photovoltaic inverter anti-reverse current systems have low communication reliability and limited signal coverage in complex electromagnetic environments, which cannot meet the needs of distributed applications and are inconvenient to maintain, leading to control failures and grid faults.
The system employs a multi-mode networking module to maintain multi-protocol communication through a directional frequency wave algorithm. Combined with data cleaning from the power detection module and a deep learning model from the anti-reverse flow module, it monitors and identifies reverse flow risks in real time. The system also performs hierarchical strategy matching through an anomaly control module and synchronizes early warning information in real time through an information sharing module.
It achieves high-precision reverse current protection in complex power consumption scenarios, improves the system's adaptability and operation and maintenance transparency, and ensures reliability and response efficiency in distributed layout and strong interference environments.
Smart Images
Figure CN121643255A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of anti-backflow technology, and in particular to a CT anti-backflow system and method. Background Technology
[0002] Currently, the mainstream solution for preventing backflow of electricity into the public power grid using photovoltaic inverters is to employ a dedicated anti-backflow controller. This solution involves deploying high-precision power sensors or current transformers (CTs) at the grid connection point to monitor the direction and magnitude of the total power transmitted from the grid to the local load in real time.
[0003] When the anti-reverse current controller detects that the total power supplied by the grid to the local load is lower than a certain set threshold (usually a positive value close to zero, such as 100W), the system determines that the current photovoltaic power generation exceeds the local load's absorption capacity, indicating a tendency to send power back to the grid. At this point, the controller must intervene in advance to prevent any substantial reverse current from occurring.
[0004] A micro inverter generally refers to an inverter in a photovoltaic power generation system with a power output of 1000 watts or less; its full name is micro photovoltaic grid-connected inverter. Photovoltaic inversion converts the direct current (DC) generated by photovoltaic modules into alternating current (AC) for use by AC loads.
[0005] Existing patents disclose a distributed photovoltaic (PV) power generation grid-connected anti-reverse current control system and method. The system includes a load forecasting and optimization component, a reverse current detection and response component, and a control and optimization scheduling component. The method includes receiving inputs from multiple data sources such as grid dispatch data, user electricity consumption habits, and weather forecasts; forecasting load demand and grid capacity in real time for a future period to provide decision-making basis for the control and scheduling component; receiving the load demand and grid capacity forecasted by the load forecasting and optimization component, as well as real-time grid status data; sending an instruction to adjust PV output to the control and optimization scheduling component when an impending reverse current is detected in the grid; receiving the forecasted load demand and grid capacity, and the instruction to adjust PV output; optimizing the PV power generation scheduling strategy, dynamically adjusting the PV power generation output, and sending control signals to the PV power generation equipment. This invention can reduce grid faults and losses caused by reverse current.
[0006] The existing technical solutions mentioned above have the following drawbacks: 1. Existing simple wireless communication has low reliability under strong industrial interference and is prone to control failure due to packet loss; in addition, its signal coverage is limited and cannot meet the needs of distributed applications; while wired deployment and maintenance are inconvenient and a single communication method lacks universality in complex scenarios. Summary of the Invention
[0007] To address the shortcomings of existing technologies, this application employs the following technical solution, which enables it to be used not only for single-phase circuit reverse current prevention but also for three-phase power grid reverse current prevention: In a first aspect, this application provides a CT anti-backflow system; comprising: The multi-mode networking module has a first input terminal indirectly connected to the output terminal of the power grid side via the power detection module, a second input terminal connected to the load side, a first communication terminal bidirectionally connected to the first communication terminal of the portable terminal, and a second communication terminal bidirectionally connected to the first communication terminal of the external smart gateway. It is used to maintain several communication protocols and data transmission with different devices according to the preset directional frequency wave algorithm to obtain the original power parameters. The power detection module has its input end connected to the output end of the power grid side and its output end connected to the input end of the anti-reverse current module. It is used to collect and clean the original power parameters of the power grid side and the load side to obtain the standard parameters of the power grid and the standard parameters of the load. The anti-reverse current module has a first output terminal connected to the input terminal of the abnormal control module and a second output terminal connected to the first input terminal of the information sharing module. It is used to construct a reverse current detection model based on historical power parameters, monitor the power grid standard parameters and the load standard parameters, identify and output reverse current risk power parameters, and generate reverse current warning information. The abnormal control module has its output end connected to the second input end of the information sharing module. It is used to classify the reverse current risk energy parameters according to the preset reverse current risk level, match the corresponding anti-reverse current strategy, and generate an anti-reverse current progress notification. The information sharing module has its communication terminals connected to the second communication terminal of the external smart gateway and the second communication terminal of the portable terminal, respectively, for sharing the backflow warning information and the anti-backflow progress notification to the portable terminal and the cloud.
[0008] By adopting the above technical solution, the multi-mode networking module dynamically maintains a multi-protocol communication network based on a directional frequency wave algorithm to acquire the original power parameters of the power grid and the load side in real time. After data cleaning and standardization by the power detection module, the anti-reverse current module uses an LSTM deep learning model trained on historical data to build a reverse current detection model, monitors and identifies reverse current risk power parameters in real time, and generates warnings. The abnormal control module uses a fuzzy control algorithm to classify reverse current risks into multiple levels and automatically matches graded strategies from power fine-tuning to emergency tripping, while generating anti-reverse current progress notifications. Finally, the information sharing module synchronizes all early warning information and handling progress to portable terminals and the cloud in real time through a two-way communication link, realizing full-domain status visualization management. Through the integrated closed-loop control of "intelligent perception - algorithm decision-making - graded control - cloud collaboration", the system significantly improves the reverse current protection accuracy and dynamic response efficiency in complex power consumption scenarios, while enhancing the system's adaptability and operation and maintenance transparency.
[0009] Furthermore, the multi-mode networking module includes: The interference detection unit is used to perform interference detection on the power grid side and the load side based on the wireless signal quality, and to obtain electromagnetic noise signals and frequency band reflection signals. The feature mining unit is used to perform feature mining on the electromagnetic noise signal and the frequency band reflection signal to obtain electromagnetic noise features and frequency band reflection features; The simulation inversion unit is used to derivatize and neutralize the electromagnetic noise characteristics and the frequency band reflection characteristics according to the time series, and generate a corresponding interference cancellation signal. The obstacle location unit is used to locate the equipment on the load side and search for and control the obstacle distribution map in the target space. The directional matching unit is used to match the corresponding communication protocol to different devices, and uses a preset directional frequency wave algorithm to spread the wireless signal in combination with the obstacle distribution map, and complete the network connection with different devices. The data interaction unit is used to eliminate interference in the data transmission process of different devices according to the interference cancellation signal, so as to obtain the original power parameters.
[0010] By adopting the above technical solution, the interference detection unit and feature mining unit collect and analyze electromagnetic noise and frequency band reflection signals in real time, and extract signal features using adaptive filtering and neural network algorithms; the simulation inversion unit dynamically neutralizes interference features and generates cancellation signals based on a time series prediction model, while the obstacle localization unit constructs an obstacle distribution map through spatial scanning; the directional matching unit combines the distribution map with a directional frequency wave algorithm to implement intelligent spread spectrum and multi-protocol adaptation; finally, the data interaction unit uses the cancellation signal to eliminate interference in the transmission link in real time; through the end-to-end intelligent processing of "interference prediction-feature neutralization-spatial modeling-directional spread spectrum", the stability of multi-device communication and data acquisition accuracy in complex electromagnetic environments are significantly improved.
[0011] Furthermore, the power detection module includes: The data acquisition unit is used to collect power parameters from the grid side and the load side to obtain the original power signal from the grid and the original operating signal from the load. The data cleaning unit is used to filter and denoise the original power signal of the power grid and the original operating signal of the load to obtain a noise-free power grid signal and a noise-free load signal. The data conversion unit is used to convert the noiseless power grid signal and the noiseless load signal to obtain the original power grid parameters and the original load parameters. The format detection unit is used to perform format correction and normalization on the original power grid parameters and the original load parameters according to the preset parameter format, so as to obtain the power grid standard parameters and the load standard parameters.
[0012] By adopting the above technical solution, the raw signals from the power grid and load side are acquired in real time by the data acquisition unit. Then, the data cleaning unit uses Kalman filtering and wavelet transform algorithms to denoise the signals and generate noise-free signals. The data conversion unit converts the noise-free signals into digital parameters through ADC analog-to-digital conversion and fast Fourier transform (FFT). Finally, the format detection unit corrects the format and scales the parameters based on the rule engine and normalization algorithm, and outputs standardized power grid and load parameters. Through multi-level algorithm collaborative processing, high-precision acquisition and standardized output of power parameters are achieved in a high-noise environment, which significantly improves the reliability of subsequent anti-reverse current control and the overall stability of the system.
[0013] Furthermore, the anti-backflow module includes: The dataset construction unit is used to divide historical electrical energy parameters according to a preset division ratio and reverse label to obtain training set and test set; The feature extraction unit is used to extract features from the training set based on the hidden layers of the network to obtain the power parameter matrix; The data separation unit is used to separate the power parameter matrix according to the data source to obtain the power grid parameter matrix and the load parameter matrix; The data extraction unit is used to mark the load parameter matrix according to the device type and extract the inverter parameter matrix; An iterative training unit is used to perform correlation training on the power grid parameter matrix and the inverter parameter matrix according to a preset number of iterations to obtain a comprehensive power weight matrix. The identification and prediction unit is used to predict and identify the test set based on the comprehensive power weight matrix and the cross-entropy loss function, and to calculate the accuracy and error rate. The evaluation and update unit is used to judge the accuracy and the error rate according to the preset accuracy range and error range. If both are within the corresponding range, the current comprehensive power weight matrix is determined to be compliant, and the reverse flow detection model is obtained. The segmentation and aggregation unit is used to segment the load standard parameters according to the equipment type to obtain the inverter standard parameters, and then aggregate them with the grid standard parameters to obtain the standard parameters of the electrical energy to be measured. The real-time monitoring unit is used to predict the standard parameters of the electrical energy to be tested based on the reverse current detection model and a preset prediction period, output the reverse current risk electrical energy parameters, and generate reverse current warning information.
[0014] By adopting the above technical solution, historical power parameters are divided by the dataset construction unit, the feature extraction unit extracts the feature matrix using the hidden layer of the neural network, and the power grid and inverter parameter matrices are obtained by the data separation and extraction unit. The iterative training unit performs correlation training using the Gauss-Newton optimization algorithm for a preset number of times to generate a comprehensive power weight matrix. The identification and prediction unit uses this matrix and the cross-entropy loss function to predict the test set and calculate the accuracy and error rate. The evaluation and update unit verifies the compliance of the model and outputs the final reflux detection model. In the real-time application stage, the segmentation and aggregation unit processes the real-time standard parameters, and the real-time monitoring unit uses this model to make periodic predictions, accurately outputs reflux risk parameters, and generates warnings. Through the whole-process machine learning closed loop from historical data mining to real-time monitoring, high-precision prediction and early warning of reflux risk are achieved, significantly improving the accuracy of anti-reflux control and the foresight of system protection.
[0015] Furthermore, the anomaly control module includes: The timing fitting unit is used to periodically fit the reverse flow risk power parameters to obtain the timing functions of the grid parameters and the inverter parameters, determine the inverter timestamp, and calculate the inverter waiting interval. A grading unit is used to match the inverter waiting interval according to a preset reverse flow risk level to obtain the reverse flow risk level; The equipment scanning unit is used to perform a status scan on all equipment on the grid side and the load side to obtain equipment operating status data. The risk adjustment unit is used to determine and adjust the backflow risk level based on the equipment operating status data to obtain the final risk level; The strategy generation unit is used to perform risk simulation of the safety control loop based on the equipment layout diagram and the risk level of the parameters, and obtain several anti-backflow strategies. The strategy matching unit is used to combine and match all the anti-backflow strategies according to the final risk level to obtain risk control strategy sets of different levels; The strategy execution unit is used to convert the risk control strategy set, generate risk control timing instructions, and control the corresponding safety protection switches in the safety control loop according to the time sequence, and generate anti-backflow progress notifications.
[0016] By adopting the above technical solution, the timing fitting unit uses Fourier transform to periodically decompose the reverse current parameters, establishes the timing function of the power grid and the inverter, and calculates the inverter waiting interval; the level classification unit maps the interval to the risk level based on the fuzzy control algorithm; after the equipment scanning unit collects equipment status data in real time, the risk adjustment unit dynamically corrects the risk level through the decision tree algorithm; the strategy generation unit uses graph theory algorithm to perform risk pre-simulation on the equipment layout diagram and generates multiple sets of anti-reverse current strategies; the strategy matching unit achieves accurate matching between strategies and risk levels through association rule mining; finally, the strategy execution unit converts the strategy set into timing instructions to control the protective switches in the safety loop; through the intelligent closed loop of "timing prediction - dynamic evaluation - strategy optimization - precise execution", the system achieves full-link adaptive management from risk identification to control execution, significantly improving the system's response speed and control accuracy in dealing with reverse current risks.
[0017] Secondly, this application also provides a CT anti-backflow method, which adopts the following technical solution: A CT backflow prevention method, applied to the aforementioned CT backflow prevention system, comprising: Based on a preset directional frequency wave algorithm, several communication protocols are maintained to connect to the distribution network and transmit data with different devices to obtain raw power parameters; The raw electrical energy parameters from the grid side and the load side are collected and cleaned to obtain the grid standard parameters and the load standard parameters; A reverse current detection model is constructed based on historical power parameters, and the standard parameters of the power grid and the standard parameters of the load are monitored to identify and output reverse current risk power parameters and generate reverse current warning information. The reverse current risk parameters are classified according to the preset reverse current risk level, and the corresponding anti-reverse current strategy is matched to generate an anti-reverse current progress notification.
[0018] By adopting the above technical solution, a multi-protocol communication link is established through a directional frequency wave algorithm to obtain raw power parameters. Kalman filtering is used for data cleaning to obtain standard parameters. A reflux detection model based on LSTM monitors and identifies reflux risks in real time and generates warnings. Finally, a fuzzy control algorithm is used to classify risks and match corresponding protection strategies to generate progress notifications. By constructing a complete anti-reflux closed loop from intelligent communication and precise monitoring to adaptive regulation, the system's response speed and protection reliability under complex operating conditions are significantly improved.
[0019] Thirdly, this application also provides an electronic device, comprising: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method described in the above scheme.
[0020] Fourthly, this application also provides a storage medium storing at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the method as described above.
[0021] In summary, the beneficial technical effects of this application are as follows: By integrating multi-protocol communication and intelligent data cleaning, high-precision detection and rapid response in complex electromagnetic environments are achieved, effectively solving the problems of insufficient accuracy and response delay in traditional CT transformers. By using an adaptive communication network and a hierarchical control strategy, the limitations of a single communication method are overcome, ensuring reliable multi-device collaborative control even in distributed layouts and environments with strong interference. Attached Figure Description
[0022] Figure 1 This is a schematic diagram of the CT anti-backflow system of one embodiment of this application; Figure 2 This is an electrical connection diagram of the CT anti-backflow system of one embodiment of this application; Figure 3 This is a flowchart illustrating the collaborative operation of multiple modules in Embodiment 2 of this application; Figure 4 This is a flowchart illustrating the CT anti-backflow method of Embodiment 2 of this application. Detailed Implementation
[0023] The present application will be further described in detail below with reference to the accompanying drawings.
[0024] Reference Figure 1 The present application discloses a CT anti-backflow system, comprising: The multi-mode networking module has a first input terminal indirectly connected to the output terminal of the power grid side via the power detection module, a second input terminal connected to the load side, a first communication terminal bidirectionally connected to the first communication terminal of the portable terminal, and a second communication terminal bidirectionally connected to the first communication terminal of the external smart gateway. It is used to maintain several communication protocols and data transmission with different devices according to the preset directional frequency wave algorithm to obtain the original power parameters. The power detection module has its input end connected to the output end of the power grid side and its output end connected to the input end of the anti-reverse current module. It is used to collect and clean the original power parameters of the power grid side and the load side to obtain the standard parameters of the power grid and the standard parameters of the load. The anti-reverse current module has a first output terminal connected to the input terminal of the abnormal control module and a second output terminal connected to the first input terminal of the information sharing module. It is used to construct a reverse current detection model based on historical power parameters, monitor the power grid standard parameters and the load standard parameters, identify and output reverse current risk power parameters, and generate reverse current warning information. The abnormal control module has its output end connected to the second input end of the information sharing module. It is used to classify the reverse current risk energy parameters according to the preset reverse current risk level, match the corresponding anti-reverse current strategy, and generate an anti-reverse current progress notification. The information sharing module has its communication terminals connected to the second communication terminal of the external smart gateway and the second communication terminal of the portable terminal, respectively, for sharing the backflow warning information and the anti-backflow progress notification to the portable terminal and the cloud.
[0025] Reference Figure 2 Example 1: After the anti-reverse current system is activated, the multi-state networking module maintains Wi-Fi / BLU multi-protocol connections with the micro-inverter, charging pile, and smart home appliances simultaneously through a directional frequency wave algorithm, acquiring the raw power parameters of each device in real time. The power detection module collects data from the grid side and the load side through a CT transformer, and outputs standard parameters to the anti-reverse current module after Kalman filtering and cleaning. This module continuously monitors the power flow of the grid based on an LSTM-based reverse current detection model. When it detects that the charging pile and high-power home appliances are working simultaneously, causing a sudden drop in load, it immediately identifies the reverse current risk power parameters and generates a warning. The anomaly control module adjusts the parameters according to preset risk levels. The system classifies the current state as medium risk and automatically matches a combined strategy of "reducing the output power of the micro-inverter + delaying the charging of the charging pile." It also controls the K1 and K2 relays in the circuit through the strategy execution unit. At the same time, the information sharing module pushes the reverse current warning and control progress to the user's mobile app and cloud platform in real time, allowing maintenance personnel to remotely view the anti-reverse current execution status of the entire community through the router. This forms an intelligent protection system that is collaborative between the "end-edge-cloud" and ultimately completes the entire process from risk detection to strategy execution within 200 milliseconds, effectively preventing power reverse current while ensuring the stable operation of the user's electrical equipment.
[0026] The implementation principle of this embodiment is as follows: real-time data collection of power grid and load is carried out through a multi-protocol network. After filtering, cleaning and feature extraction, dynamic monitoring is performed using a pre-trained LSTM reflux detection model. When a power reflux risk is identified, the risk is immediately classified and a control strategy is automatically matched through a fuzzy control algorithm. Finally, the power adjustment command is executed through a relay group and the entire process status is pushed to the user terminal and cloud platform in real time, thereby completing the closed-loop anti-reflux control from detection, decision-making to execution within milliseconds.
[0027] In this embodiment, the multi-mode networking module includes: The interference detection unit is used to perform interference detection on the power grid side and the load side based on the wireless signal quality, and to obtain electromagnetic noise signals and frequency band reflection signals. The feature mining unit is used to perform feature mining on the electromagnetic noise signal and the frequency band reflection signal to obtain electromagnetic noise features and frequency band reflection features; The simulation inversion unit is used to derivatize and neutralize the electromagnetic noise characteristics and the frequency band reflection characteristics according to the time series, and generate a corresponding interference cancellation signal. The obstacle location unit is used to locate the equipment on the load side and search for and control the obstacle distribution map in the target space. The directional matching unit is used to match the corresponding communication protocol to different devices, and uses a preset directional frequency wave algorithm to spread the wireless signal in combination with the obstacle distribution map, and complete the network connection with different devices. The data interaction unit is used to eliminate interference in the data transmission process of different devices according to the interference cancellation signal, so as to obtain the original power parameters.
[0028] In this embodiment, the communication protocols include RS485, WIFI, sub-1g, LoRa, and LAN; In this embodiment, the power detection module includes: The data acquisition unit is used to collect power parameters from the grid side and the load side to obtain the original power signal from the grid and the original operating signal from the load. The data cleaning unit is used to filter and denoise the original power signal of the power grid and the original operating signal of the load to obtain a noise-free power grid signal and a noise-free load signal. The data conversion unit is used to convert the noiseless power grid signal and the noiseless load signal to obtain the original power grid parameters and the original load parameters. The format detection unit is used to perform format correction and normalization on the original power grid parameters and the original load parameters according to the preset parameter format, so as to obtain the power grid standard parameters and the load standard parameters.
[0029] The anti-backflow module includes: The dataset construction unit is used to divide historical electrical energy parameters according to a preset division ratio and reverse label to obtain training set and test set; The feature extraction unit is used to extract features from the training set based on the hidden layers of the network to obtain the power parameter matrix; The data separation unit is used to separate the power parameter matrix according to the data source to obtain the power grid parameter matrix and the load parameter matrix; The data extraction unit is used to mark the load parameter matrix according to the device type and extract the inverter parameter matrix; An iterative training unit is used to perform correlation training on the power grid parameter matrix and the inverter parameter matrix according to a preset number of iterations to obtain a comprehensive power weight matrix. The identification and prediction unit is used to predict and identify the test set based on the comprehensive power weight matrix and the cross-entropy loss function, and to calculate the accuracy and error rate. The evaluation and update unit is used to judge the accuracy and the error rate according to the preset accuracy range and error range. If both are within the corresponding range, the current comprehensive power weight matrix is determined to be compliant, and the reverse flow detection model is obtained. The segmentation and aggregation unit is used to segment the load standard parameters according to the equipment type to obtain the inverter standard parameters, and then aggregate them with the grid standard parameters to obtain the standard parameters of the electrical energy to be measured. The real-time monitoring unit is used to predict the standard parameters of the electrical energy to be tested based on the reverse current detection model and a preset prediction period, output the reverse current risk electrical energy parameters, and generate reverse current warning information.
[0030] The abnormal control module includes: The timing fitting unit is used to periodically fit the reverse flow risk power parameters to obtain the timing functions of the grid parameters and the inverter parameters, determine the inverter timestamp, and calculate the inverter waiting interval. A grading unit is used to match the inverter waiting interval according to a preset reverse flow risk level to obtain the reverse flow risk level; The equipment scanning unit is used to perform a status scan on all equipment on the grid side and the load side to obtain equipment operating status data. The risk adjustment unit is used to determine and adjust the backflow risk level based on the equipment operating status data to obtain the final risk level; The strategy generation unit is used to perform risk simulation of the safety control loop based on the equipment layout diagram and the risk level of the parameters, and obtain several anti-backflow strategies. The strategy matching unit is used to combine and match all the anti-backflow strategies according to the final risk level to obtain risk control strategy sets of different levels; The strategy execution unit is used to convert the risk control strategy set, generate risk control timing instructions, and control the corresponding safety protection switches in the safety control loop according to the time sequence, and generate anti-backflow progress notifications.
[0031] In this embodiment, one or more high-precision current transformers (CTs) are installed at the common connection point (PCC) between the power grid and the user load, and combined with a voltage measuring device, to monitor the instantaneous power flowing to the power grid.
[0032] The defined power value is positive when electrical energy flows from the grid to the load. This means that the power consumed by the local load is greater than the power generated by the inverter, and the shortfall is made up by the grid.
[0033] When the power output of the inverter is exactly equal to the power consumed by the local load, the power measured on the grid side is close to zero. The system sets a small positive number close to zero as a threshold (dead zone) to prevent frequent operation.
[0034] When the power output of the inverter exceeds the immediate consumption of the local load, the excess electrical energy flows back to the grid. At this time, the current direction detected by the current transformer (CT) is opposite to that when there is no reverse current, and the power value calculated by the system is negative. This "negative power" is the decisive signal for determining the occurrence of reverse current.
[0035] Reference Figure 3 Example 2: After the multi-mode networking module is activated, its interference detection unit scans the electromagnetic noise of the WiFi / Bluetooth bands in real time, the feature mining unit extracts the noise spectrum features through wavelet transform, the simulation inversion unit generates interference cancellation signals based on the LSTM prediction model, and the obstacle localization unit constructs an indoor obstacle distribution map using UWB positioning technology. The directional matching unit dynamically adjusts the signal beam using a directional frequency wave algorithm based on this distribution map, successfully establishing multi-protocol connections with charging piles (WiFi), micro-inverters (BLU), and smart home appliances. The data interaction unit uses the cancellation signal to eliminate transmission interference and stably acquires the original power parameters. In the power detection module, the data acquisition unit collects signals from the grid side and the load side through a CT transformer, the data cleaning unit uses Kalman filtering to remove noise, the data conversion unit converts the signals into digital parameters through FFT analysis, and the format detection unit finally outputs standardized parameters. The anti-reverse current module monitors the power flow in real time using a pre-trained LSTM reverse current detection model (trained with 100,000 sets of historical data, achieving an accuracy of 98.2%). When it detects that fast charging from the charging pile and photovoltaic power generation are superimposed, resulting in power surplus, it immediately generates a reverse current warning message. The timing fitting unit of the abnormal control module predicts the power fluctuation cycle through Fourier analysis, the level classification unit sets the current risk as Level-2, the equipment scanning unit detects that the air conditioner is about to start, the risk adjustment unit raises the risk to Level-3, the strategy generation unit generates a combined strategy of "reducing inverter output + delaying charging power + adjusting air conditioner start-up timing" based on the equipment layout diagram, the strategy execution unit completes the power adjustment within 150ms by controlling K1 and K2 relays, and pushes the execution progress to the information sharing module in real time. Finally, the entire process data and protection effect are displayed to users through the cloud platform and mobile APP, realizing intelligent anti-backflow closed-loop management from perception, decision-making to execution.
[0036] Reference Figure 4 A CT anti-backflow method, applied to the aforementioned CT anti-backflow system, comprising: S1: Based on a preset directional frequency wave algorithm, maintain several communication protocols to connect with different devices in the power distribution network and transmit data to obtain raw power parameters; S2: Collect and clean the raw electrical energy parameters on the grid side and the load side to obtain the grid standard parameters and the load standard parameters; S3: Construct a reverse current detection model based on historical power parameters, monitor the standard parameters of the power grid and the standard parameters of the load, identify and output reverse current risk power parameters, and generate reverse current warning information; S4: Divide the reverse current risk power parameters according to the preset reverse current risk level, match the corresponding anti-reverse current strategy, and generate an anti-reverse current progress notification.
[0037] The implementation principle of this embodiment is as follows: By dynamically maintaining cross-protocol communication connections with various devices through a directional frequency wave algorithm, raw power parameters are acquired in real time. Then, data from the grid side and the load side are collected and cleaned to generate standardized parameters. Subsequently, a reverse current detection model trained based on historical data continuously monitors these parameters, accurately identifies reverse current risks and outputs warning information. Finally, the corresponding anti-reverse current strategy is automatically matched according to the preset risk level, and real-time progress notifications are generated, thereby forming a closed-loop protection system from intelligent sensing, dynamic analysis to adaptive control, ensuring that the system can efficiently and reliably prevent power reverse current.
[0038] An electronic device, comprising: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method described in the above scheme.
[0039] A storage medium storing at least one instruction, at least one program, a code set, or an instruction set, wherein the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the method as described above.
[0040] The embodiments described in this specific implementation are preferred embodiments of this application and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.
Claims
1. A CT anti-flow system, comprising: a multi-state networking module, a first input end is indirectly connected with an output end of a power grid side through a power detection module, a second input end is connected with a load side, a first communication end is bidirectionally connected with a first communication end of a portable terminal, a second communication end is bidirectionally connected with a first communication end of an external intelligent gateway, used for maintaining several communication protocols and connecting with different devices for network configuration and data transmission according to a preset directional frequency wave algorithm, and obtaining original power parameters; the power detection module, an input end is connected with an output end of the power grid side, and an output end is connected with an input end of an anti-flow module, used for collecting and cleaning the original power parameters of the power grid side and the load side, to obtain power grid standard parameters and load standard parameters; the anti-flow module, a first output end is connected with an input end of an abnormality regulation module, and a second output end is connected with a first input end of an information sharing module, used for constructing an anti-flow detection model according to historical power parameters, monitoring the power grid standard parameters and the load standard parameters, identifying and outputting anti-flow risk power parameters, and generating anti-flow warning information; the abnormality regulation module, an output end is connected with a second input end of the information sharing module, used for dividing the anti-flow risk power parameters according to a preset anti-flow risk level, matching corresponding anti-flow strategies, and generating anti-flow progress notifications; the information sharing module, a communication end is connected with a second communication end of the external intelligent gateway and a second communication end of the portable terminal, used for sharing the anti-flow warning information and the anti-flow progress notifications to the portable terminal and a cloud.
2. The CT anti-reflux system of claim 1, wherein, the multi-state networking module comprises: an interference detection unit, used for performing interference detection on the power grid side and the load side according to wireless signal quality, to obtain electromagnetic noise signals and frequency band reflection signals; a feature mining unit, used for performing feature mining on the electromagnetic noise signals and the frequency band reflection signals, to obtain electromagnetic noise features and frequency band reflection features; a simulation inversion unit, used for performing derivation and neutralization on the electromagnetic noise features and the frequency band reflection features according to time series, to generate corresponding interference cancellation signals; an obstacle positioning unit, used for positioning devices of the load side, and searching and controlling an obstacle distribution map in a target space; a directional matching unit, used for matching corresponding communication protocols for different devices, and performing spread spectrum on wireless signals by using a preset directional frequency wave algorithm in combination with the obstacle distribution map, to complete network configuration with different devices; a data interaction unit, used for performing interference elimination on a data transmission process of different devices according to the interference cancellation signals, to obtain original power parameters.
3. The CT anti-reflux system of claim 1, wherein, the power detection module comprises: a data collection unit, used for collecting power parameters of the power grid side and the load side, to obtain power grid original power signals and load original operation signals; a data cleaning unit, used for performing filtering and denoising on the power grid original power signals and the load original operation signals, to obtain noise-free power grid signals and noise-free load signals; a data conversion unit, used for performing data conversion on the noise-free power grid signals and the noise-free load signals, to obtain original power grid parameters and original load parameters; A format detection unit is configured to perform format correction and normalization on the original grid parameters and the original load parameters according to a preset parameter format, to obtain grid standard parameters and load standard parameters.
4. The CT anti-reflux system of claim 1, wherein: The anti-backflow module comprises: A data set construction unit is configured to divide historical electric energy parameters according to a preset division ratio and a backflow label, to obtain a training set and a test set; A feature extraction unit is configured to perform feature extraction on the training set according to a network hidden layer, to obtain an electric energy parameter matrix; A data separation unit is configured to separate the electric energy parameter matrix according to a data source, to obtain a grid parameter matrix and a load parameter matrix; A data extraction unit is configured to mark the load parameter matrix according to a device type, and extract an inverter parameter matrix; An iterative training unit is configured to perform associated training on the grid parameter matrix and the inverter parameter matrix according to a preset number of iterations, to obtain a comprehensive electric energy weight matrix.
5. The CT anti-reflux system of claim 4, wherein: The anti-backflow module further comprises: An identification and prediction unit is configured to perform prediction and identification on the test set according to the comprehensive electric energy weight matrix and a cross-entropy loss function, to calculate an accuracy rate and an error rate; An evaluation and update unit is configured to judge the accuracy rate and the error rate according to preset accuracy and error intervals, and if both are located in the corresponding intervals, it is determined that the current comprehensive electric energy weight matrix is compliant, to obtain a backflow detection model; A segmentation and aggregation unit is configured to segment the load standard parameters according to a device type, to obtain inverter standard parameters, and aggregate the inverter standard parameters with the grid standard parameters, to obtain to-be-measured electric energy standard parameters; A real-time monitoring unit is configured to predict the to-be-measured electric energy standard parameters according to the backflow detection model and a preset prediction period, to output backflow risk electric energy parameters, and generate backflow warning information.
6. The CT anti-reflux system of claim 1, wherein: The abnormal regulation module comprises: A time sequence fitting unit is configured to perform period fitting on the backflow risk electric energy parameters, to obtain a grid parameter time sequence function and an inverter parameter time sequence function, and determine an inverter time stamp, to calculate an inverter waiting interval; A grade division unit is configured to match the inverter waiting interval according to a preset backflow risk level, to obtain a backflow risk grade; A device scanning unit is configured to perform state scanning on all devices on the grid side and the load side, to obtain device operation state data; A risk adjustment unit is configured to determine and adjust the backflow risk grade according to the device operation state data, to obtain a final risk grade; A strategy generation unit is configured to perform risk pre-play on a safety control loop according to a device layout diagram and the parameter risk grade, to obtain a plurality of anti-backflow strategies; A strategy matching unit is configured to combine and match all the anti-backflow strategies according to the final risk grade, to obtain a risk regulation strategy set of different grades; A strategy execution unit is configured to convert the risk regulation strategy set, to generate a risk regulation time sequence instruction, and regulate corresponding safety protection switches in the safety control loop according to a time sequence, to generate an anti-backflow progress notification.
7. A CT anti-reflux method applied to the CT anti-reflux system according to any one of claims 1-6, characterized in that, It comprises: A plurality of communication protocols are maintained according to a preset directional frequency wave algorithm to connect and transmit data with different devices in a power grid, to obtain original electric energy parameters; Collect and clean the original electric energy parameters of the power grid side and the load side to obtain power grid standard parameters and load standard parameters; According to the historical electric energy parameters, an inverse flow detection model is constructed, the power grid standard parameters and the load standard parameters are monitored, inverse flow risk electric energy parameters are identified and output, and inverse flow warning information is generated; According to a preset inverse flow risk level, the inverse flow risk electric energy parameters are divided, corresponding anti-inverse flow strategies are matched, and anti-inverse flow progress notifications are generated.
8. An electronic device, comprising: Comprise: One or more processors; Memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method of claim 7.
9. A storage medium having at least one instruction, at least one program, a code set or an instruction set stored therein, the at least one instruction, the at least one program, the code set or the instruction set being loaded and executed by a processor to implement the CT anti-inverse flow method of claim 7.