Intelligent fusion terminal system

Abstract

The invention provides an intelligent fusion terminal system, which relates to the technical field of power systems and comprises a data acquisition module, an edge calculation module, a storage module, a communication module, a cooperative calculation module and a safety protection module. The edge calculation module adopts an edge calculation framework to realize data on-site analysis and decision, and the storage module is used for circularly storing event records and fixed-point data; the edge computing architecture is adopted to construct core processing capacity, most data analysis is completed on site, cloud remote processing is not needed, data transmission time delay is remarkably reduced, the fault event response speed is improved, abnormal states such as voltage out-of-limit and circuit breaker tripping in a transformer area can be sensed in real time, a control instruction can be rapidly triggered, and the fault diagnosis efficiency is improved. Equipment damage or power failure range expansion caused by response lag is effectively avoided, and a millisecond-level protection mechanism is provided for safe operation of transformer area equipment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power systems, and in particular to an intelligent fusion terminal system. BACKGROUND

[0002] The low-voltage distribution area is the area of the distribution transformer and its low-voltage power grid, and needs to be configured with a terminal to realize real-time monitoring of the operating state, accurate positioning of faults, analysis of power quality, and intelligent control, to support the digital transformation of the power grid and energy efficiency management.

[0003] The existing low-voltage distribution area terminal has multiple technical bottlenecks: first, data processing relies on cloud centralized computing, resulting in insufficient real-time performance, such as the long response time of traditional terminals, which is several seconds, far higher than the millisecond-level requirement for the safety protection of the area equipment; second, the multi-source device protocol compatibility is poor, and the communication protocols of different manufacturers of electric energy meters and switching devices are not unified, resulting in a success rate of less than 60% for device linkage control, and long time consumption for fault positioning; third, the edge computing capability is missing, and complex business needs frequent interaction with the cloud, causing a surge in network load, making it difficult to meet the needs of real-time monitoring and accurate control of the new power system. Therefore, the present application proposes an intelligent fusion terminal system to solve the problems in the prior art. SUMMARY

[0004] To solve the above problems, the present application proposes an intelligent fusion terminal system, which solves the problems of low data processing efficiency, insufficient storage capacity, weak multi-device collaboration capability, and insufficient on-site analysis capability in the prior art, and realizes intelligent acquisition and control of low-voltage distribution areas.

[0005] To achieve the purpose of the present application, the following technical solutions are adopted: an intelligent fusion terminal system, comprising a data acquisition module, an edge computing module, a storage module, a communication module, a collaborative computing module, and a security protection module, the data acquisition module is used to acquire voltage, current, and device state data, the edge computing module uses an edge computing architecture to realize data on-site analysis and decision-making, the storage module is used to cyclically store event records and fixed-point data;

[0006] The communication module uses 4G / 5G, optical fiber, and power line carrier communication to realize the interaction of the system with the uplink master station and the downlink device, the collaborative computing module is used to perform cloud training, edge reasoning, and multi-terminal data sharing on the collected data, and the security protection module uses a security chip and container technology to ensure data security.

[0007] Further improvement lies in that the data acquisition module comprises an analog quantity acquisition unit, a sensor interface unit, an electric energy meter acquisition unit, a switching quantity acquisition unit, a wireless temperature measurement unit and a mutual inductor monitoring unit, the analog quantity acquisition unit integrates voltage and current acquisition circuits, and real-time measurement of three-phase voltage and current effective values and 2-19 harmonic components is realized, and the formula is:

[0008]

[0009] wherein, U rms is the voltage effective value, I rms is the current effective value, u i , i i are instantaneous voltage and current values, and N is the number of sampling points.

[0010] Further improvement lies in that the sensor interface unit communicates with the power distribution transformer sensor to monitor oil temperature and gas protection state parameters; the electric energy meter acquisition unit acquires electric energy meter data through an RS-485 interface and supports active / passive uploading; the switching quantity acquisition unit is configured with four groups of switching quantity inputs to monitor circuit breaker positions and cabinet door opening and closing states; the wireless temperature measurement unit adopts a low-power wireless module to realize real-time monitoring of equipment temperature; and the mutual inductor monitoring unit identifies TA secondary circuit states through high-frequency signal injection technology.

[0011] Further improvement lies in that the edge computing module comprises a data preprocessing unit, a state analysis unit and a decision control unit, the data preprocessing unit is used for filtering and denoising processing of collected data, and the formula is:

[0012]

[0013] wherein, h(k) is a filter coefficient, x(n) is an input signal, y(n) is an output signal, and M is the order of the filter.

[0014] Further improvement lies in that the state analysis unit realizes real-time calculation of voltage deviation and unbalance rate parameters based on an edge computing architecture, and the formula is:

[0015]

[0016]

[0017] wherein, U 实际 is a measured voltage, U 额定 is a rated voltage, U max , U min and U 平均 are maximum, minimum and average values of three-phase voltages respectively.

[0018] Further improvement lies in that the decision control unit generates control instructions according to analysis results of the data preprocessing unit and the state analysis unit, including adjusting a tap of a load-voltage-regulating transformer and switching a reactive power compensation device.

[0019] Further improvement lies in that the storage module comprises a local storage unit and a data cache unit, the local storage unit adopts an industrial-grade flash memory, cyclically stores at least 1024 event records, 31-day fixed-point data and extreme value data, and the data cache unit is used for temporarily storing unsent data and has a storage function after power failure.

[0020] Further improvement lies in that the communication module comprises a remote communication unit and a local communication unit, the remote communication unit adopts 4G / 5G and optical fiber communication, interacts with a master station in uplink and communicates with smart meters and switch devices in downlink, and the local communication unit performs networking of substation equipment through power line carrier and micro-power wireless.

[0021] Further improvement lies in that the collaborative computing module comprises a cloud-edge collaborative unit and a multi-terminal collaborative unit, the cloud-edge collaborative unit performs model training through a cloud end, executes inference through an edge end, adopts a model segmentation algorithm, and unloads a calculation-intensive layer to an edge server, and a formula is as follows:

[0022]

[0023] Wherein, T local is a local calculation time, T remote is remote transmission and calculation time, and T total is minimized by optimizing a segmentation point.

[0024] The multi-terminal collaborative unit shares data of adjacent terminals through a distributed coordination algorithm, and improves fault research and judgment accuracy.

[0025] Further improvement lies in that the security protection module comprises a hardware security unit and a software security unit, the hardware security unit integrates a security chip, performs data encryption and identity authentication, and the software security unit adopts a container technology to isolate applications and block malicious attacks.

[0026] The present application has the following advantages:

[0027] 1. The present application adopts an edge computing architecture to build core processing capability, completes most data analysis on site, does not need to rely on remote cloud processing, significantly reduces data transmission delay, improves fault event response speed, can realize real-time sensing and rapid triggering of control instructions on abnormal states such as voltage overrun and circuit breaker tripping in a substation, effectively avoids damage to equipment or expansion of power outage range caused by response lag, and provides a millisecond-level protection mechanism for safe operation of substation equipment.

[0028] 2、The application adopts a cloud edge collaborative mechanism, trains a model on the cloud, and executes inference on the edge, realizes safe and efficient analysis of data without leaving the area, and through a distributed coordination algorithm, adjacent terminals share data and improve fault analysis accuracy.

[0029] 3、The application takes into account the access requirements of multiple source devices, integrates multiple communication protocols and local communication technologies, realizes compatibility of different manufacturers and different types of devices, solves the data island problem caused by non-uniform protocols of traditional transformer area devices, and realizes seamless collaboration of functions such as electric energy meter state monitoring, switch signal acquisition, and transformer working condition sensing.

[0030] 4、The application is equipped with a data acquisition module and an intelligent algorithm, a fine monitoring system is constructed for key devices such as distribution transformers and transformers, accurate identification of transformer secondary circuit state is realized through high-frequency signal injection technology, and measurement errors caused by poor circuit contact or shunt are eliminated; low-power wireless temperature measurement technology is used to track device contact temperature in real time and early warning of excessive contact resistance and other hidden faults; the analog quantity acquisition circuit combines with the digital filtering algorithm to effectively filter out noise such as power grid harmonics and pulse interference, ensuring high-precision measurement of basic data such as voltage and current, and providing a reliable data source for electric energy metering and power analysis. BRIEF DESCRIPTION OF DRAWINGS

[0031] Figure 1 The present application is composed of the following drawings. DETAILED DESCRIPTION

[0032] In order to deepen the understanding of the present application, the present application will be further described in combination with the embodiments, and the present embodiment is only used to explain the present application and does not constitute a limitation on the protection scope of the present application.

[0033] Embodiment one

[0034] According to Figure 1 The present embodiment proposes an intelligent fusion terminal system, which comprises a data acquisition module, an edge computing module, a storage module, a communication module, a collaborative computing module and a security protection module, the data acquisition module is used for realizing the acquisition of voltage, current and device state data, the edge computing module realizes data analysis and decision-making on site by using edge computing architecture, and the storage module is used for cyclic storage of event records and fixed-point data.

[0035] The communication module realizes the interaction of the system with the uplink master station and the downlink device by 4G / 5G, optical fiber, and power line carrier communication. The collaborative computing module is used for cloud training, edge inference, and multi-terminal data sharing of the collected data. The security protection module uses a security chip and container technology to ensure data security. Through edge computing, real-time decision-making is enabled, large-capacity storage supports deep analysis, multiple protocols are integrated to realize device collaboration, high-precision monitoring ensures data reliability, and modular design adapts to business expansion. Multiple technologies comprehensively break through the functional bottlenecks of traditional terminals and build a five-in-one intelligent system of collection, calculation, storage, control, and expansion, providing core technical support for the digitalization and intelligent transformation of low-voltage distribution areas in new power system construction.

[0036] The data collection module includes an analog quantity collection unit, a sensor interface unit, an electric energy meter collection unit, a switching quantity collection unit, a wireless temperature measurement unit, and a mutual inductor monitoring unit. The analog quantity collection unit integrates voltage and current collection circuits to measure real-time three-phase voltage, current effective value, and 2-19 harmonic components, which are expressed by the formula:

[0037]

[0038] where U rms is the voltage effective value, I rms is the current effective value, u i and i i are instantaneous voltage and current values, and N is the number of sampling points. The sensor interface unit communicates with the distribution transformer sensor to monitor oil temperature and gas protection state parameters. The electric energy meter collection unit collects electric energy meter data through an RS-485 interface and supports active / passive uploading. The switching quantity collection unit is configured with four groups of switching quantity inputs to monitor circuit breaker position and cabinet door opening / closing status. The wireless temperature measurement unit uses a low-power wireless module to monitor device temperature in real time. The mutual inductor monitoring unit identifies TA secondary circuit status through high-frequency signal injection technology. Multi-dimensional data fusion is used to integrate distribution, marketing, and environmental multi-source data collection to solve the problem of single function of traditional terminals. High-precision monitoring technology is used to improve the reliability of distribution area equipment through TA secondary circuit status identification and wireless temperature measurement functions.

[0039] The edge computing module includes a data preprocessing unit, a state analysis unit, and a decision control unit. The data preprocessing unit is used for filtering and denoising processing of collected data, which is expressed by the formula:

[0040]

[0041] Wherein, h(k) is the filter coefficient, x(n) is the input signal, y(n) is the output signal, and M is the order of the filter. The formula describes the operation process of a finite impulse response (FIR) filter: by multiplying the input signal x(n) at the current time n with the corresponding filter coefficient h(k) and the previous The purpose is to filter out high-frequency noise, power frequency interference (such as 50Hz / 60Hz harmonic) or abnormal pulse in the original signal, improve the accuracy and stability of the data, and provide high-quality input data for subsequent edge computing, state monitoring, etc. For example, when the intelligent fusion terminal collects voltage and current data of distribution transformers, the filter can effectively suppress harmonic interference (such as 3rd, 5th harmonic) in the power grid, ensuring the accuracy of power calculation and energy metering.

[0042] The state analysis unit calculates the voltage deviation and unbalance rate parameters in real time based on the edge computing architecture, and the formula is:

[0043]

[0044]

[0045] Wherein, U 实际 is the measured voltage, U 额定 is the rated voltage, U max , U min , U 平均 are the maximum, minimum and average values of three-phase voltage respectively. The decision control unit generates control instructions according to the analysis results of the data preprocessing unit and the state analysis unit, including adjusting the load of the voltage regulating transformer and switching the reactive power compensation device. With the edge computing architecture, more than 90% of data processing is completed on site, reducing the burden of the main station and improving the response speed.

[0046] ​​The storage module includes a local storage unit and a data cache unit. The local storage unit uses industrial-grade flash memory and cyclically stores at least 1024 event records, 31 days of fixed-point data, and extreme value data. The data cache unit is used for temporary storage of unuploaded data and has the function of storing data after power failure. Adopting an industrial-grade high-capacity storage solution, it can achieve long-term, high-density recording of transformer area operation data, supporting at least 31 days of fixed-point data (such as voltage, current, power, and other parameters collected every 15 minutes) and cyclical storage of historical events (such as equipment failures and parameter changes). This meets the power sector's long-term data backtracking needs for transformer area line loss analysis, load characteristic assessment, and equipment life assessment, providing a solid data foundation for distribution network planning and operation optimization. Simultaneously, the local storage supports data preservation after power failure, and combined with remote access functionality, it ensures data integrity and traceability, avoiding blind spots in business analysis caused by data interruption.

[0047] The communication module includes a remote communication unit and a local communication unit. The remote communication unit uses 4G / 5G and fiber optic communication, interacting with the master station uplink and communicating with smart meters and switching equipment downlink. The local communication unit uses power line carrier and low-power wireless to network the equipment in the distribution area.

[0048] The collaborative computing module includes a cloud-edge collaborative unit and a multi-terminal collaborative unit. The cloud-edge collaborative unit trains the model in the cloud and performs inference at the edge. It uses a model segmentation algorithm to offload the computationally intensive layer to the edge server, as shown in the formula:

[0049]

[0050] Among them, T local For local time calculation, T remote To optimize the time for remote transmission and computation, the split point is optimized to make T... total Minimum accuracy; the multi-terminal collaborative unit enables adjacent terminals to share data through a distributed coordination algorithm, improving the accuracy of fault diagnosis. A cloud-edge collaborative mechanism is adopted, with the model trained in the cloud and inference executed at the edge, achieving secure and efficient data analysis without leaving the area.

[0051] The security protection module includes a hardware security unit and a software security unit. The hardware security unit integrates a security chip for data encryption and identity authentication, while the software security unit uses container technology to isolate applications and block malicious attacks. It adopts a modular design, with hardware platform-based and software app-based, supporting flexible functional expansion to adapt to different scenario needs.

[0052] Example 2

[0053] according to Figure 1 As shown in the figure, this embodiment proposes an intelligent fusion terminal system with the following configuration:

[0054] Hardware: A40i quad-core processor (1.2 GHz), 2 GB of memory, 8 GB of storage, supports 4G / fiber dual communication link.

[0055] Software: Based on lightweight Linux system, deploy power distribution and marketing APP, realize transformer area topology identification, line loss analysis and other functions.

[0056] Verification data

[0057] Data acquisition efficiency: support 1000+ devices concurrent access, acquisition success rate reaches 99.9%, 3 times higher than traditional system.

[0058] Storage capacity: cyclic storage of 31 days of data, storage space occupancy rate is less than 80%, meet the standard of State Grid.

[0059] Edge computing performance: real-time analysis of voltage qualification rate, load rate and other parameters, response time less than 100ms, 90% shorter than cloud processing.

[0060] Collaborative computing effect: through cloud-edge collaboration, distributed photovoltaic access transformer area voltage out-of-limit times reduced by 80%, reverse overload occurrence rate reduced by 70%.

[0061] In a certain transformer area test, the first 5 layers of CNN model are unloaded to the edge server, the inference time is reduced from 1200ms to 350ms, and the accuracy rate remains 98%. The TA secondary circuit state identification accuracy is 100%, effectively avoiding measurement error.

[0062] The intelligent fusion terminal system adopts an edge computing architecture to build core processing capability, completes most data analysis on site, does not need to rely on remote cloud processing, significantly reduces data transmission delay, improves fault event response speed, can realize real-time sensing on abnormal states such as voltage overrun and circuit breaker tripping in a transformer area and quickly trigger control instructions, effectively avoids equipment damage or power outage range expansion caused by response lag, and provides a millisecond-level protection mechanism for safe operation of transformer area equipment. Meanwhile, the application adopts a cloud-edge collaborative mechanism, trains a model on the cloud, executes inference on the edge, realizes safe and efficient analysis of data without leaving the transformer area, and makes adjacent terminals share data through a distributed coordination algorithm to improve fault judgment accuracy. In addition, the application takes into account the access requirements of multiple source devices, integrates multiple communication protocols and local communication technologies, realizes compatibility of different manufacturers and different types of equipment, solves the data island problem caused by non-uniform protocols of traditional transformer area equipment, and realizes seamless collaboration of functions such as electric energy meter state monitoring, switch quantity signal acquisition and transformer working condition sensing. Finally, the application is equipped with a data acquisition module and intelligent algorithms, builds a fine monitoring system for key devices such as distribution transformers and transformers, realizes accurate identification of the state of the transformer secondary circuit through high-frequency signal injection technology, eliminates measurement errors caused by poor circuit contact or shunt; adopts low-power wireless temperature measurement technology to track the temperature of the device contact in real time, and early warning of hidden faults such as excessive contact resistance; the analog quantity acquisition circuit combined with the digital filter algorithm effectively filters out noise such as power grid harmonics and pulse interference, ensures high-precision measurement of basic data such as voltage and current, and provides a reliable data source for electric energy metering, power analysis and other businesses.

[0063] The above shows and describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above embodiments, and the above embodiments and descriptions in the specification are only to illustrate the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. An intelligent converged terminal system comprising a data acquisition module, an edge computing module, a storage module, a communication module, a collaborative computing module and a security protection module, characterized in that: The data acquisition module is used to realize the acquisition of voltage, current, and device state data, the edge computing module realizes data analysis and decision-making on site based on an edge computing architecture, and the storage module is used to cyclically store event records and fixed-point data. The communication module realizes the interaction of the system with the upper master station and the lower device through 4G / 5G, optical fiber, and power line carrier communication, the collaborative computing module is used to perform cloud training, edge reasoning, and multi-terminal data sharing on the collected data, and the security protection module uses a security chip and container technology to ensure data security.

2. The intelligent converged terminal system of claim 1, wherein: The data acquisition module includes an analog quantity acquisition unit, a sensor interface unit, an electric energy meter acquisition unit, a switching quantity acquisition unit, a wireless temperature measurement unit, and a mutual inductor monitoring unit. where U rms is the voltage effective value, I rms is the current effective value, u i , i i are the instantaneous voltage, current values, and N is the number of sampling points.

3. The intelligent converged terminal system of claim 2, wherein: The analog quantity acquisition unit integrates voltage and current acquisition circuits, and is used to measure the effective values of three-phase voltage and current and 2-19 harmonic components in real time, and the formula is:

4. The intelligent converged terminal system of claim 1, wherein: The sensor interface unit communicates with the power distribution transformer sensor to monitor oil temperature and gas protection state parameters; the electric energy meter acquisition unit acquires electric energy meter data through an RS-485 interface and supports active / passive uploading; the switching quantity acquisition unit is configured with four groups of switching quantity inputs to monitor the positions of circuit breakers and the opening and closing states of cabinet doors; the wireless temperature measurement unit uses a low-power wireless module to monitor the temperature of the device in real time; and the mutual inductor monitoring unit identifies the TA secondary circuit state through high-frequency signal injection technology. The edge computing module includes a data preprocessing unit, a state analysis unit, and a decision control unit.

5. The intelligent converged terminal system of claim 4, wherein: The data preprocessing unit is used to filter and denoise the collected data, and the formula is: where U 实际 is the measured voltage, U 额定 is the rated voltage, U max , U min , U 平均 are the maximum, minimum and average values of the three-phase voltage, respectively.

6. The intelligent converged terminal system of claim 5, wherein: wherein h(k) is a filter coefficient, x(n) is an input signal, y(n) is an output signal, and M is the order of the filter.

7. The intelligent converged terminal system of claim 1, wherein: The state analysis unit calculates voltage deviation and unbalance rate parameters in real time based on the edge computing architecture, and the formula is:

8. The intelligent converged terminal system of claim 1, wherein: The decision control unit generates control instructions according to the analysis results of the data preprocessing unit and the state analysis unit, including adjusting the tap position of the on-load voltage regulating transformer and switching the reactive power compensation device.

9. The intelligent converged terminal system of claim 1, wherein: The storage module includes a local storage unit and a data cache unit. Where T local is the local computation time, T remote is the remote transmission and computation time, and T total is minimized by optimizing the split point. The local storage unit uses an industrial-grade flash memory to cyclically store at least 1024 event records, 31 days of fixed-point data, and extreme value data, and the data cache unit is used to temporarily store data that has not been uploaded and has a power failure storage function. The communication module includes a remote communication unit and a local communication unit. The remote communication unit uses 4G / 5G and optical fiber communication to interact with the master station, and the local communication unit performs group networking of the substation equipment through power line carrier and micro-power wireless communication. The collaborative computing module includes a cloud-edge collaborative unit and a multi-terminal collaborative unit. The cloud-edge collaborative unit trains a model through the cloud and executes reasoning through the edge, and uses a model segmentation algorithm to offload the calculation-intensive layer to the edge server, and the formula is: The multi-terminal collaborative unit enables adjacent terminals to share data through a distributed coordination algorithm to improve the accuracy of fault research and judgment.

10. The intelligent converged terminal system of claim 1, wherein: The security protection module comprises a hardware security unit and a software security unit, the hardware security unit integrates a security chip, performs data encryption and identity authentication, and the software security unit adopts a container technology to isolate applications and block malicious attacks.

Images