New energy intelligent ring main unit management system

Abstract

The invention relates to the technical field of ring main units, and discloses a new energy intelligent ring main unit management system comprising an acquisition module comprising a sensor; an energy taking module; the control module is electrically connected with the acquisition module and the energy taking module, and the control module comprises a monitoring unit, a judgment unit, an early warning unit and an adjustment unit; the monitoring unit is configured to collect sensor data; the judgment unit is configured to input the sensor data into a pre-trained state model for comparison, and determine a current running state of the ring main unit based on a comparison result of the state model; the early warning unit is configured to determine an abnormal region based on an abnormal state and send corresponding early warning information when the current operation state of the ring main unit is abnormal; and the adjusting unit is configured to control the energy taking module to switch the power supply circuit based on the early warning information. The power utilization reliability of the ring main unit is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ring main units, and more particularly, to a new energy intelligent ring main unit management system. Background Art

[0002] The ring main unit power supply system is a highly efficient and reliable power distribution method widely used in urban distribution networks. Its core feature is to achieve multi-path power supply through a ring network structure, ensuring that when a fault occurs in a certain line, power can continue to be transmitted through other paths, thereby improving the reliability and stability of power supply. Currently, there are mainly three ways for existing ring main unit power supply systems: (1) The ring main unit is equipped with a PT cabinet without a DTU, and is powered by the 220V power supply on the secondary side of the PT in the PT cabinet. The branch switch supplies power for the lighting of the ring main unit, the environmental monitoring system (including environmental temperature and humidity monitoring, smoke detection, water immersion monitoring, intelligent dehumidification control inside the cabinet, ventilation control, lighting control); (2) The ring main unit is equipped with a PT + DTU box, and is powered by the 220V power supply on the secondary side of the PT in the PT cabinet. The branch switch supplies power for the lighting of the ring main unit, the environmental control system, and the power supply of the distribution automation terminal (DTU). The DC power supply of the DTU supplies power for the switch control operation power supply; (3) The ring main unit has no PT cabinet and DTU, and there is no station power supply inside the ring main unit.

[0003] However, the above three ring main unit power supply systems all have drawbacks. When the bus of the ring main unit loses power or the PT cabinet fails, it is impossible to ensure the normal power supply for lighting, environmental monitoring, and switch control, posing a safety hazard.

[0004] Therefore, it is necessary to design a new energy intelligent ring main unit management system to solve the problems existing in the current technology. Summary of the Invention

[0005] In view of this, the present invention proposes a new energy intelligent ring main unit management system, aiming to solve the problem of poor power supply reliability of the current ring main unit.

[0006] The present invention proposes a new energy intelligent ring main unit management system, including:

[0007] An acquisition module, including sensors;

[0008] An energy extraction module for standby power generation of the ring main unit;

[0009] A control module electrically connected to the acquisition module and the energy extraction module. The control module includes a monitoring unit, a judgment unit, a warning unit, and an adjustment unit;

[0010] The monitoring unit is configured to collect sensor data and preprocess the sensor data;

[0011] The judgment unit is configured to extract the sensor data and input the sensor data into a pre-trained state model for comparison;

[0012] The judgment unit is further configured to determine the current operating state of the ring network box based on the comparison result of the state model;

[0013] The early warning unit is configured to determine an abnormal area based on the abnormal state and send corresponding early warning information based on different abnormal areas when the current operating state of the ring network box is abnormal;

[0014] The adjustment unit is configured to control the energy acquisition module to switch the power supply line based on the early warning information.

[0015] Furthermore, the sensor data is preprocessed in at least the following ways:

[0016] Data cleaning, used to remove invalid values;

[0017] Time synchronization, used to synchronize timestamps to ensure the time consistency of data;

[0018] The sampling frequency is unified to keep the data frequency consistent;

[0019] Data normalization is used to unify the dimensions.

[0020] Further, the judgment unit is configured to extract the sensor data and input the sensor data into a pre-trained state model for comparison, including:

[0021] Acquire two sets of time series of the sensor data, wherein one of the time series includes the time series waveform of the bus voltage and the time series waveform of the secondary side voltage of the PT cabinet, and the other time series includes the load current, the ambient temperature and humidity, and the circuit breaker status;

[0022] In the underlying feature learning stage, voltage time series feature learning and auxiliary data encoding are obtained, and finally the voltage feature and auxiliary feature are spliced to form enhanced time series features;

[0023] In the high-level feature learning stage, a dense network with a two-layer structure and a Dropout regularization technique are used to extract high-order nonlinear features layer by layer based on the enhanced time series features to form the state model;

[0024] The sensor data is input into the state model to perform waveform reconstruction and obtain a voltage prediction value.

[0025] Furthermore, when the sensor data is input into the state model, waveform reconstruction is performed and a voltage prediction value is obtained, the method includes:

[0026] Obtain the voltage time series waveform predicted by the state model and calculate the mean square error with the voltage time series waveform in the sensor data.

[0027] Further, when obtaining the voltage time series waveform predicted by the state model and calculating the mean square error with the voltage time series waveform in the sensor data, it includes:

[0028]

[0029] where MSE is the mean square error, T is the total number of time steps of the time series, t is the index of the current time step, is the voltage value predicted by the state model at time step t, is the actually obtained voltage value at time step t.

[0030] Further, when the determination unit is further configured to determine the current operating state of the ring main unit box based on the comparison result of the state model, it includes:

[0031] When the mean square error value of the sensor data is less than or equal to three times the mean of the mean square error predicted by the state model, it is determined that the current operating state is normal;

[0032] When the mean square error value of the sensor data is greater than three times the mean of the mean square error predicted by the state model, the abnormal data is sent to the warning unit.

[0033] Further, when the warning unit is configured to determine the abnormal area based on the abnormal state and send corresponding warning information based on different abnormal areas when the current operating state of the ring main unit box is abnormal, it includes:

[0034] When the warning unit receives the abnormal data, it triggers a fault diagnosis mechanism, and the fault diagnosis mechanism is divided into bus power failure, PT cabinet disconnection, and PT cabinet resonance;

[0035] When the three-phase voltages are all close to zero and the load current is close to zero, it is determined as a bus power failure fault;

[0036] When a single-phase voltage is close to zero, the other two-phase voltages are normal, and the fuse status of the corresponding phase is in a blown state, it is determined as a PT cabinet disconnection fault;

[0037] When the high-frequency harmonic energy exceeds the limit and the zero-sequence voltage rises, it is determined as a PT cabinet resonance fault;

[0038] After determining the abnormal area, generate corresponding warning information and send the corresponding warning information to the adjustment unit.

[0039] Further, when the warning unit is configured to determine an abnormal area based on the abnormal state and send corresponding warning information based on different abnormal areas when the current operating state of the ring main unit is abnormal, it further includes:

[0040] When the warning unit determines that it is a bus power failure fault, it generates a first-level warning message;

[0041] When the warning unit determines that it is a PT cabinet resonance fault, it generates a second-level warning message;

[0042] When the warning unit determines that it is a PT cabinet disconnection fault, it generates a third-level warning message.

[0043] Further, when the adjustment unit is configured to control the energy-taking module to switch the power line based on the warning information, it includes:

[0044] When the adjustment unit receives the warning information, it controls the energy-taking module to switch the power of the energy-taking module and invert the current into 220V voltage through an inverter device.

[0045] Further, it further includes: a remote monitoring module, the remote monitoring module is electrically connected to the warning unit, and the remote monitoring module is configured to monitor the operating state of the warning unit and send the operating state monitoring information to the Internet and the cloud data platform.

[0046] Compared with the prior art, the beneficial effects of the present invention are as follows: By deploying sensors, the system can continuously collect key parameters such as voltage, current, and temperature, providing data support for the operation and management of the ring main unit. The data preprocessing function of the monitoring unit filters out environmental interference and signal noise, improving the accuracy of subsequent analysis. The judgment unit uses a pre-trained state model to compare data, and through the fault recognition ability, it can distinguish various operating abnormalities, reducing the false alarm rate. Based on the pre-trained state model, the system can automatically analyze the operation data, identify various abnormal situations, improving the timeliness and accuracy of fault discovery. When an abnormal situation is detected, the system can automatically start the standby power supply to ensure the continuous power supply of key loads, avoiding power supply interruptions caused by untimely power switching, and improving the reliability of the ring main unit operation. Through intelligent warning, the operation and maintenance personnel can obtain the equipment status information in a timely manner, carry out maintenance work targeted, improving work efficiency, and also reducing unnecessary inspection frequencies and labor costs. By real-time monitoring parameters such as temperature and load, the system can timely discover safety hazards such as overload and overheat, and take corresponding measures, preventing the occurrence of safety accidents such as equipment damage and electrical fires through an active protection mechanism. Description of the Drawings

[0047] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the following detailed description of the preferred embodiments. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. Also, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0048] Figure 1 This is a structural block diagram of the new energy intelligent ring main cabinet management system provided by an embodiment of the present invention. Specific embodiments

[0049] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be fully conveyed to those skilled in the art. It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the drawings and in combination with the embodiments.

[0050] In some embodiments of the present application, referring to Figure 1 as shown, a new energy intelligent ring main cabinet management system includes:

[0051] An acquisition module, including sensors,

[0052] An energy acquisition module for standby power generation of the ring main cabinet,

[0053] A control module, electrically connected to the acquisition module and the energy acquisition module. The control module includes a monitoring unit, a judgment unit, a warning unit, and an adjustment unit.

[0054] The monitoring unit is configured to collect sensor data and preprocess the sensor data.

[0055] The judgment unit is configured to extract the sensor data and input the sensor data into a pre-trained state model for comparison.

[0056] The judgment unit is further configured to determine the current operating state of the ring main cabinet based on the comparison result of the state model.

[0057] The warning unit is configured to, when the current operating state of the ring main cabinet is abnormal, determine the abnormal area based on the abnormal state and send corresponding warning information based on different abnormal areas.

[0058] The adjustment unit is configured to control the energy acquisition module to switch the power line based on the warning information.

[0059] Specifically, through sensors, the operating voltage, current, circuit breaker status and other information of the ring network box are monitored and obtained in real time to provide an operating data basis for the control module. The energy acquisition module includes a solar power generation unit, an energy storage unit, an inverter and a dual power automatic switching device. The solar power generation unit generates electricity through solar energy and stores electric energy in the energy storage unit. When the control module detects that the ring network box fails and cannot be powered by the PT side, it can switch to the energy storage unit for power supply through the dual power automatic switching device, and the inverter can invert the energy storage unit into a 220V voltage to drive the normal operation of the ring network box. The sensors include but are not limited to monitoring environmental temperature and humidity, water immersion, fire, partial discharge in the cabinet, temperature outside the cabinet, voltage, current and other parameters. The real-time monitored sensor data is input into the pre-trained state model through the judgment unit. By comparing the data in the state model with the real-time monitored data, the current operating status of the ring network box can be discovered in time. When the operating status is abnormal, the specific abnormal area and abnormal components are discovered in time, and early warning information is generated based on the severity. When the adjustment unit receives the early warning information, it switches the power line to ensure that the ring network box can operate normally.

[0060] In some embodiments of the present application, the sensor data is preprocessed in at least the following ways:

[0061] Data cleaning, used to remove invalid values,

[0062] Time synchronization is used to synchronize timestamps to ensure the time consistency of data.

[0063] The sampling frequency is unified to keep the data frequency consistent.

[0064] Data normalization is used to unify the dimensions.

[0065] Specifically, sensor data preprocessing is a crucial step in data analysis and model application. Its purpose is to improve the quality, consistency, and usability of data, thereby providing a reliable basis for subsequent analysis, judgment, and decision-making. Sensors may be affected by factors such as environmental interference, equipment failures, or communication delays during actual operation, resulting in problems such as noisy, missing, time-inconsistent, or non-uniform dimensions in the data. Data cleaning is used to eliminate invalid values, including outliers, missing values, or error values. These invalid values are caused by reasons such as sensor failures, communication interruptions, or environmental interference. After eliminating the invalid values, the data is more accurate and reliable, avoiding the interference of noise on the analysis results. The cleaned data can better reflect the actual situation, improving the training effect and prediction accuracy of the model. Time synchronization is used to unify the timestamps of sensor data, ensuring consistency in the time dimension. Different sensors may have clock deviations or sampling time asynchronization, resulting in data that cannot be aligned in time. Sampling frequency unification is used to adjust the data of different sensors to the same sampling frequency, ensuring consistency in the time series. Different sensors may collect data at different frequencies, resulting in inconsistent numbers of data points. Data normalization is used to unify the dimensions of data, adjusting the value ranges of data from different sensors to the same scale. Different sensors may measure different physical quantities. After normalization, data of different physical quantities can be compared and analyzed on the same scale, avoiding biases caused by dimensional differences.

[0066] It is understandable that sensor data preprocessing plays a crucial role in data analysis and model applications. Its core objective is to improve the quality, consistency, and usability of data through various means, thereby providing a reliable basis for subsequent analysis, judgment, and decision-making. Sensors may be affected by factors such as environmental interference, equipment failures, or communication delays during actual operation, resulting in problems such as noisy, missing, time-inconsistent, or dimensionally inconsistent data. Data cleaning improves the accuracy and reliability of data by removing invalid values (such as outliers, missing values, or error values), avoiding interference from noise on analysis results, enabling the cleaned data to better reflect the actual situation, and thereby enhancing the training effect and prediction accuracy of the model. Time synchronization ensures the consistency of data in the time dimension by unifying the timestamps of sensor data, solving the problem of data misalignment caused by clock deviations or sampling time asynchronization among different sensors, thus guaranteeing the comparability and analysis accuracy of multi-source data at the same time point and providing more reliable support for real-time monitoring and decision-making. Sampling frequency unification solves the problem of differences in the number of data points caused by inconsistent sensor sampling frequencies by adjusting the data of different sensors to the same sampling frequency, simplifies the data analysis process, improves the comparability of data, and supports the efficient operation of real-time processing systems. Data normalization eliminates the bias caused by dimensional differences by adjusting the value ranges of data from different sensors to the same scale, enabling data of different physical quantities to be compared and analyzed on the same scale, while also improving the convergence speed and stability of the model, enhancing the robustness and prediction accuracy of the model. The comprehensive application of these preprocessing methods not only improves the quality and consistency of data but also provides a data foundation with high quality, consistency, and strong usability for subsequent data analysis, model training, and decision-making, thereby improving the overall performance and reliability of the system.

[0067] In some embodiments of the present application, when the judgment unit is configured to extract sensor data and input the sensor data into a pre-trained state model for comparison, it includes:

[0068] Obtaining two time series of sensor data, one time series including the time series waveform of the bus voltage and the time series waveform of the secondary side voltage of the PT cabinet, and the other time series including load current, ambient temperature and humidity, and the breaker status

[0069] In the underlying feature learning stage, obtaining voltage time series feature learning and auxiliary data encoding, and finally splicing the voltage feature and the auxiliary feature to form an enhanced time series feature.

[0070] In the high-level feature learning stage, using a dense network with a two-layer structure and Dropout regularization technology based on the enhanced time series feature to extract high-order non-linear features layer by layer to form a state model.

[0071] Input the sensor data into the state model for waveform reconstruction and obtain the voltage prediction value.

[0072] Specifically, by obtaining the bus voltage (from the primary side of the PT in the ring main unit, three-phase voltages Ua, Ub, Uc), as well as the secondary side voltage of the PT (from the measurement circuit of the PT cabinet, single-phase / three-phase voltage) and auxiliary timing data. The auxiliary timing data includes but is not limited to load current (Ia, Ib, Ic): reflecting the load condition of the ring main unit, used to analyze the operating state and load changes of the equipment; ambient temperature: monitoring the ambient temperature around the ring main unit, used to evaluate the impact of temperature on equipment operation; ambient humidity: monitoring the ambient humidity around the ring main unit, used to evaluate the impact of humidity on the insulation performance of the equipment; breaker status (open / closed): recording the switch status of the breaker, used to judge the operating state and fault conditions of the equipment; power factor: reflecting the power utilization efficiency of the equipment, used to analyze the operating efficiency and energy consumption of the equipment; active power and reactive power: used to evaluate the actual power output and energy consumption of the equipment; frequency: monitoring the frequency of the power grid, used to evaluate the stability of the power grid; equipment temperature: monitoring the temperature of key components inside the ring main unit, used to evaluate the operating state and overheating risk of the equipment; vibration data: monitoring the vibration condition of the equipment, used to judge the mechanical state and potential faults of the equipment, etc. The state model is an LSTM-Autoencoder model. In the underlying feature learning stage, by inputting the original voltage waveform, a low-dimensional timing feature vector is obtained. And for the auxiliary data encoding, by inputting data such as load current, temperature, and humidity, it is processed through a fully connected encoder. Finally, the voltage feature and the auxiliary feature are concatenated to obtain an enhanced timing feature. Furthermore, in the high-level feature learning, it is trained through a two-layer Dense network + Dropout regularization technique. Its objective function is mean square error (MSE) + waveform similarity constraint (DTW loss):

[0073]

[0074] where N is the number of samples, that is, the total number of sensor data, is the predicted value of the i-th sample, is the true value of the i-th sample, is the square of the calculated Euclidean distance, and DTW(Vpred, Vtrue) is the dynamic time warping distance, used to measure the similarity between the predicted sequence V pred and the true sequence V true . DTW can handle the non-linear alignment problem of time series on the time axis. is the overall loss function, which is used to measure the comprehensive difference between the predicted voltage waveform and the true voltage waveform. λ is the weight coefficient, which is used to control the contribution degree of the DTW loss value. It is trained and optimized by the AdamW optimizer. When the validation loss does not decrease for 5 consecutive rounds, the training is terminated to obtain the trained state model. Finally, the voltage waveform + auxiliary data that are obtained in real time and not involved in the training can be input into the model for voltage prediction. By plotting the predicted waveform and the true waveform, the abnormal interval can be detected, and then the abnormal range can be determined through the abnormal interval.

[0075] In some embodiments of the present application, when inputting the sensor data into the state model for waveform reconstruction and obtaining the voltage prediction value, it includes:

[0076] Obtain the mean square error by calculating the voltage time series waveform predicted by the state model and the voltage time series waveform in the sensor data.

[0077] In some embodiments of the present application, when obtaining the mean square error by calculating the voltage time series waveform predicted by the state model and the voltage time series waveform in the sensor data, it includes:

[0078]

[0079] Among them, MSE is the mean square error, T is the total number of time steps of the time series, t is the index of the current time step, is the voltage value predicted by the state model at time step t, is the actually obtained voltage value at time step t.

[0080] Specifically, by calculating the mean square error, the difference between the true voltage time series waveform and the voltage time series waveform in the model can be determined, and then whether the true voltage time series waveform is abnormal can be determined.

[0081] It is understandable that by performing waveform reconstruction and error analysis on real-time sensor data and state model prediction data, important technical support is provided for the ring main unit cabinet. The mean square error (MSE) is used as the core evaluation index. By quantifying the difference between the predicted waveform and the actual waveform, the judgment of the operating state of the ring main unit cabinet is realized. By establishing a point-by-point comparison mechanism for time series, the system can capture subtle abnormal changes in the voltage waveform, including transient disturbances and progressive deteriorations that are difficult to detect by traditional monitoring means. The MSE calculation method comprehensively considers the overall deviation degree of the waveform at each time point, avoiding the limitations of single-threshold judgment, making the abnormal detection more comprehensive and reliable. It realizes the leap from simple alarm to intelligent diagnosis. It can not only judge whether an abnormality exists, but also identify the type of abnormality through the error distribution characteristics, providing an important basis for subsequent fault location and handling. In practical applications, the waveform-level comparison and analysis improve the recognition accuracy of complex faults such as PT disconnection and resonance, and the false alarm rate is reduced compared with traditional methods. The time series processing ability of the system enables it to adapt to the monitoring requirements of different sampling frequencies, and can effectively capture both high-frequency transient processes and long-term trend changes. Through continuous accumulation of error data analysis, the system can automatically optimize the parameter settings of the state model and realize self-improvement of the diagnostic ability.

[0082] In some embodiments of the present application, when the judgment unit is further configured to determine the current operating state of the ring main unit cabinet based on the comparison result of the state model, it includes:

[0083] When the mean square error value of the sensor data is less than or equal to three times the mean of the mean square error predicted by the state model, it is determined that the current operating state is normal.

[0084] When the mean square error value of the sensor data is greater than three times the mean of the mean square error predicted by the state model, the abnormal data is sent to the warning unit.

[0085] It is understandable that the closer the predicted waveform is to the real waveform, the more normal the system state. If a fault is triggered when the mean square error value is only greater than the mean, it will result in a high false alarm rate, because there is about a 50% probability that normal data exceeds the mean (if the MSE follows a symmetric distribution). That is, when the mean square error (MSE) of the real-time data exceeds the mean of the MSE of the normal samples in the training set plus 3 times the standard deviation, it is determined as abnormal, which can reduce the misjudgment rate.

[0086] In some embodiments of the present application, when the warning unit is configured to determine the abnormal area based on the abnormal state and send corresponding warning information based on different abnormal areas when the current operating state of the ring main unit cabinet is abnormal, it includes:

[0087] When the warning unit receives abnormal data, it triggers a fault diagnosis mechanism. The fault diagnosis mechanism is divided into bus power failure, PT cabinet disconnection, and PT cabinet resonance.

[0088] When the three-phase voltages are all close to zero and the load current is close to zero, it is determined as a bus power loss fault.

[0089] When a single-phase voltage is close to zero, the other two-phase voltages are normal, and the fuse status of the corresponding phase is blown, it is determined as a PT cabinet disconnection fault.

[0090] When the high-frequency harmonic energy exceeds the limit and the zero-sequence voltage increases, it is determined as a PT cabinet resonance fault.

[0091] After determining the abnormal area, generate the corresponding warning information and send the corresponding warning information to the adjustment unit.

[0092] It can be understood that by adopting a fault diagnosis logic that integrates multi-dimensional criteria and setting differentiated determination conditions for different fault types, for example, for bus power loss, it is necessary to simultaneously meet the double verification of three-phase voltages being close to zero and the load current returning to zero, avoiding the possibility of misjudgment by a single parameter and improving the accuracy of fault identification. For PT cabinet disconnection faults, the voltage abnormality and fuse status signal are correlated and analyzed. Through the combined determination of the disappearance of a single-phase voltage and the blowing of the fuse of the corresponding phase, it is possible to distinguish between PT secondary circuit disconnection and primary-side equipment faults, providing a clear direction for on-site maintenance. When dealing with complex faults such as PT resonance, through a composite criterion that combines high-frequency harmonic energy monitoring and zero-sequence voltage change, the problem that traditional methods are not sensitive to resonance characteristics is solved. In particular, it has excellent detection ability for overvoltage phenomena caused by ferromagnetic resonance of electromagnetic voltage transformers in ungrounded neutral systems, and can give early warnings in the initial stage of resonance to avoid serious consequences such as equipment insulation breakdown.

[0093] In some embodiments of the present application, when the warning unit is configured to determine the abnormal area based on the abnormal state and send the corresponding warning information based on different abnormal areas when the current operating state of the ring network cabinet is abnormal, it further includes:

[0094] When the warning unit determines it as a bus power loss fault, generate a first-level warning information.

[0095] When the warning unit determines it as a PT cabinet resonance fault, generate a second-level warning information.

[0096] When the warning unit determines it as a PT cabinet disconnection fault, generate a third-level warning information.

[0097] Specifically, by establishing a hierarchical early warning mechanism, a fault handling system is provided for the management of ring network cabinets. The early warning information is divided into three levels according to the fault type and severity, achieving full coverage monitoring from emergency accidents to general defects, and improving the emergency response ability and operation reliability of the power system. The first-level early warning targets major faults such as bus power loss that directly affect the continuity of power supply. It adopts the highest priority alarm strategy and ensures the immediate transmission of information through multiple channels, prompting the operation and maintenance team to handle it first and minimizing the fault time. The second-level early warning is specifically for potential hazards such as PT cabinet resonance that may cause equipment damage but do not affect power supply temporarily. When the system issues an alarm, it can start the harmonic elimination device and provide detailed parameters such as resonance frequency and amplitude for the operation and maintenance personnel to support precise defect elimination. The third-level early warning deals with local defects such as PT cabinet disconnection. It marks the fault phase and fuse position through a visual interface to guide rapid replacement on-site and avoid the development of more serious faults. The differentiated early warning strategy optimizes the allocation of operation and maintenance resources, enabling the limited maintenance force to handle the most urgent faults first and improving the emergency repair efficiency. Secondly, the precise matching of the early warning level and the fault impact avoids overreaction. For example, a lower-level early warning is used for PT disconnection, which not only ensures the timely handling of problems but also prevents unnecessary emergency responses. The hierarchical early warning mechanism avoids the frequent switching of standby power supplies caused by false alarms by accurately distinguishing the nature of faults, thus extending the service life of equipment. The standardized classification of early warning information also lays a foundation for fault statistical analysis. By mining historical data, weak links in specific lines or equipment can be identified to guide targeted technical transformation.

[0098] In some embodiments of the present application, when the adjustment unit is configured to control the switching unit to switch the power line based on the early warning information, it includes:

[0099] When the adjustment unit receives the early warning information, it controls the switching unit to switch the power of the energy-taking module and invert the current into 220V voltage through an inverter device.

[0100] It is understandable that this technical solution provides guarantee for the continuous and reliable operation of the ring main unit through an intelligent power switching and inverter control mechanism. The early warning information and power switching linkage system established by the system realizes the full-automatic processing from fault detection to emergency power supply. When the adjustment unit receives the early warning information, it can immediately control the switching unit to start the standby power supply and convert the electric energy into a standard 220V voltage output through the inverter device to ensure that the power supply of critical loads is not interrupted. Through the closed-loop management of real-time monitoring and automatic switching, the system has changed the disadvantages of slow response and low efficiency of traditional manual operations. This flexible power configuration scheme enables the system to adapt to the application requirements of different scenarios. Whether it is a high-reliability required site in the core area of the city or an unattended ring main unit in remote areas, reliable power guarantee can be obtained. The system downtime caused by power interruption is minimized, ensuring the continuity and integrity of monitoring data and providing complete data support for fault analysis. The automatic switching mechanism eliminates the risk of human operation errors, and the built-in multiple electrical protection functions can prevent abnormal conditions such as short circuits and overloads to ensure the safety and reliability of the switching process.

[0101] In some embodiments of the present application, it further includes: a remote monitoring module, which is electrically connected to the early warning unit and is configured to monitor the operating state of the early warning unit and send the operating state monitoring information to the Internet and the cloud data platform.

[0102] It is understandable that by introducing the remote monitoring module and deeply integrating it with the cloud data platform, a complete ecological system for the intelligent management of the ring main unit is constructed, realizing the all-weather and all-round monitoring and management of the operating state of the equipment. The direct connection between the remote monitoring module and the early warning unit ensures the real-time transmission of abnormal information, enabling managers to master the operating conditions of the ring main unit at any time and anywhere. The system uploads the operating state monitoring information of the early warning unit to the Internet and the cloud data platform in real time. Based on the centralized monitoring function of the cloud platform, the unified management of multiple ring main units deployed dispersedly is realized. The operation and maintenance personnel can directly understand the real-time status of each site through the visual interface, improving the management efficiency. Secondly, the historical data stored in the cloud provides a solid foundation for in-depth analysis. By mining the data such as early warning information and switching records accumulated over a long time, the operation rules of the equipment can be identified, the life cycle can be predicted, and the formulation of preventive maintenance plans can be guided. In terms of emergency response, the remote monitoring module realizes the instant push of fault information and supports multi-channel alarms such as text messages, emails, and mobile applications to ensure that relevant personnel are informed of abnormal situations in the first time and strive for precious time for rapid disposal.

[0103] In the above embodiments, by deploying sensors, the system can continuously collect key parameters such as voltage, current, and temperature, providing data support for the operation and management of the ring main unit. The data preprocessing function of the monitoring unit filters out environmental interference and signal noise, improving the accuracy of subsequent analysis. The judgment unit uses a pre-trained state model to compare data and, through its fault recognition ability, can distinguish various abnormal operations, reducing the false alarm rate. Based on the pre-trained state model, the system can automatically analyze operation data, identify various abnormal situations, improving the timeliness and accuracy of fault discovery. When an abnormal situation is detected, the system can automatically start the standby power supply to ensure continuous power supply to critical loads, avoiding power outages caused by untimely power switching and improving the reliability of the ring main unit operation. Through intelligent early warning, operation and maintenance personnel can obtain device status information in a timely manner, carry out maintenance work in a targeted manner, improving work efficiency and reducing unnecessary inspection frequencies and labor costs. By continuously monitoring parameters such as temperature and load, the system can promptly detect safety hazards such as overload and overheating and take corresponding measures, preventing the occurrence of safety accidents such as equipment damage and electrical fires through an active protection mechanism.

[0104] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, system, or computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0105] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of flows and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the specified functions in Figure 1 one or more of the flows Figure 1 or multiple flows and / or blocks

[0106] These computer program instructions can also be stored in a computer-readable memory capable of guiding a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device that implements the specified functions in Figure 1 one or more of the flowsFigure 1 The functions specified in one or more boxes.

[0107] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide for implementing the steps of the functions specified in one or more processes and / or boxes Figure 1 One or more processes and / or boxes Figure 1 The steps of the functions specified in one or more boxes.

[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: it is still possible to modify the specific implementation manners of the present invention or make equivalent replacements. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention shall be covered by the protection scope of the claims of the present invention.

Claims

1. A new energy intelligent ring main cabinet management system, characterized in that, Including: A collection module, including sensors; An energy harvesting module for standby power generation of the ring main unit cabinet; A control module electrically connected to the collection module and the energy harvesting module, the control module including a monitoring unit, a judgment unit, a warning unit, and an adjustment unit; The monitoring unit is configured to collect sensor data and preprocess the sensor data; The judgment unit is configured to extract the sensor data and input the sensor data into a pre-trained state model for comparison; The judgment unit is further configured to determine the current operating state of the ring main unit cabinet based on the comparison result of the state model; The warning unit is configured to, when the current operating state of the ring main unit cabinet is abnormal, determine the abnormal area based on the abnormal state and send corresponding warning information based on each abnormal area; The adjustment unit is configured to control the energy harvesting module to switch the power line based on the warning information.

2. The new energy intelligent ring main cabinet management system according to claim 1, characterized in that When preprocessing the sensor data, at least the following methods are included: Data cleaning for removing invalid values; Time synchronization for synchronizing timestamps to ensure the time consistency of data; Unifying the sampling frequency for keeping the data frequency consistent; Data normalization for unifying the dimension.

3. The new energy intelligent ring main cabinet management system according to claim 2, characterized in that When the judgment unit is configured to extract the sensor data and input the sensor data into a pre-trained state model for comparison, it includes: Obtaining two time series of the sensor data, one of the time series including the time series waveform of the bus voltage and the time series waveform of the secondary side voltage of the PT cabinet, and the other time series including the load current, the ambient temperature and humidity, and the breaker state; In the underlying feature learning stage, obtaining voltage time series feature learning and auxiliary data encoding, and finally splicing the voltage feature and the auxiliary feature to form an enhanced time series feature; In the high-level feature learning stage, using a dense network with a two-layer structure and Dropout regularization technology, based on the enhanced time series feature, extracting high-order non-linear features layer by layer to form the state model; Inputting the sensor data into the state model for waveform reconstruction and obtaining the voltage prediction value.

4. The new energy intelligent ring main cabinet management system according to claim 3, characterized in that When inputting the sensor data into the state model for waveform reconstruction and obtaining the voltage prediction value, it includes: Obtaining the mean square error by calculating the voltage time series waveform predicted by the state model and the voltage time series waveform in the sensor data.

5. The new energy intelligent ring main cabinet management system according to claim 4, wherein, When obtaining the mean square error by calculating the voltage time series waveform predicted by the state model and the voltage time series waveform in the sensor data, it includes: Among them, MSE is the mean square error, T is the total number of time steps of the time series, t is the index of the current time step, is the voltage value predicted by the state model at time step t, is the actually obtained voltage value at time step t.

6. The new energy intelligent ring main cabinet management system according to claim 5, characterized in that When the judgment unit is further configured to determine the current operating state of the ring main unit cabinet based on the comparison result of the state model, it includes: When the mean square error value of the sensor data is less than or equal to three times the mean of the mean square error predicted by the state model, it is determined that the current operating state is normal; When the mean square error value of the sensor data is greater than three times the mean of the mean square error predicted by the state model, the abnormal data is sent to the warning unit.

7. The new energy intelligent ring main cabinet management system according to claim 6, characterized in that When the warning unit is configured to, when the current operating state of the ring main unit cabinet is abnormal, determine the abnormal area based on the abnormal state and send corresponding warning information based on each abnormal area, it includes: When the warning unit receives the abnormal data, it triggers a fault diagnosis mechanism, which is divided into bus power loss, PT cabinet line break, and PT cabinet resonance; When the three-phase voltages are all close to zero and the load current is close to zero, it is determined as a bus power loss fault; When a single-phase voltage is close to zero, the other two-phase voltages are normal, and the fuse status of the corresponding phase is blown, it is determined as a PT cabinet line break fault; When the high-frequency harmonic energy exceeds the limit and the zero-sequence voltage increases, it is determined as a PT cabinet resonance fault; After determining the abnormal area, corresponding warning information is generated and sent to the adjustment unit.

8. The new energy intelligent ring main cabinet management system according to claim 7, wherein, The warning unit is configured to, when the current operating state of the ring network cabinet is abnormal, determine the abnormal area based on the abnormal state and send corresponding warning information based on different abnormal areas, and further includes: When the warning unit determines it as a bus power loss fault, it generates a first-level warning information; When the warning unit determines it as a PT cabinet resonance fault, it generates a second-level warning information; When the warning unit determines it as a PT cabinet line break fault, it generates a third-level warning information.

9. The new energy intelligent ring main cabinet management system according to claim 8, wherein The adjustment unit is configured to control the energy-taking module to switch the power line based on the warning information, and includes: When the adjustment unit receives the warning information, it controls the energy-taking module to switch the power supply of the energy-taking module and invert the current into 220V voltage through an inverter device.

10. The new energy intelligent ring main cabinet management system according to claim 9, characterized in that It further includes: A remote monitoring module, which is electrically connected to the warning unit. The remote monitoring module is configured to monitor the operating state of the warning unit and send the operating state monitoring information to the Internet and the cloud data platform.

Images