Integrated management system for intelligent power distribution

Abstract

The invention discloses a comprehensive management system for intelligent power distribution, and the system comprises a sensing layer which is provided with a plurality of sensors, the sensors are disposed on power distribution equipment, and are used for collecting the operation data of a power distribution system in real time, and the operation data comprise a current I, a voltage U, a power P, an equipment temperature T, a power factor cos phi, and an insulation resistance R; the transmission layer is used for transmitting the operation data to a data layer; the data layer is used for preprocessing the operation data to generate standardized data, and preprocessing comprises data cleaning and normalization conversion; and the application layer comprises a fault early warning module, a predictive maintenance module, an energy consumption analysis module and a remote control module, and the application layer realizes collaborative management by calling the standardized data. Multi-dimensional data fusion, a dynamic intelligent algorithm and a module interaction mechanism are introduced, and full-process intelligence of power distribution management is achieved.

Description

Technical Field

[0001] This invention relates to the field of management systems, and more specifically to a comprehensive management system for intelligent power distribution. Background Technology

[0002] Smart power distribution is a digital and automated upgrade of traditional power distribution systems. It integrates sensors, communication equipment, and intelligent algorithms into the power distribution network to achieve real-time sensing, data interconnection, and dynamic control of the operating status of power distribution equipment. Its core features include: comprehensive acquisition of multi-dimensional parameters, visualized monitoring of the power distribution network, rapid response to abnormal conditions, and flexible allocation of power resources. It aims to improve the reliability, safety, and energy efficiency of power distribution systems, adapting to the demands of new power systems such as renewable energy grid integration and diversified load access.

[0003] Intelligent power distribution management is based on the operation and optimization process of intelligent power distribution systems, encompassing the entire chain of processes including data acquisition and processing, status assessment, and decision execution. Its core objective is to solve the problems of passive response, reliance on manual labor, and low efficiency in traditional power distribution management through intelligent means. Therefore, a comprehensive management system for intelligent power distribution is proposed. Summary of the Invention

[0004] The present invention solves the above-mentioned technical problems through the following technical solutions, the present invention comprising: The sensing layer is equipped with several sensors, which are deployed on the power distribution equipment to collect real-time operating data of the power distribution system. The operating data includes current I, voltage U, power P, equipment temperature T, power factor cosΦ, and insulation resistance R. The transport layer, which is communicatively connected to the perception layer, is used to transmit the running data to the data layer; The data layer, which is communicatively connected to the transport layer, is used to preprocess the running data to generate standardized data. The preprocessing includes data cleaning and normalization transformation. The application layer, which communicates with the data layer, includes a fault early warning module, a predictive maintenance module, an energy consumption analysis module, and a remote control module. The application layer achieves collaborative management by calling the standardized data, specifically: The fault warning module calculates the fault warning value S based on the standardized data of current I, voltage U, and equipment temperature T by integrating time-series features and equipment parameters. When S reaches the preset warning threshold, a fault warning is triggered. The predictive maintenance module calculates the remaining lifespan L of the equipment based on the standardized data of the power P, equipment temperature T, and insulation resistance R, combined with the equipment aging model, and generates a dynamic maintenance plan based on L. The energy consumption analysis module calculates the energy consumption E per unit time based on the standardized data of current I, voltage U, and power factor cosΦ through an energy efficiency evaluation model, and outputs an energy consumption optimization strategy that matches the electricity consumption scenario. Based on the early warning results of the fault early warning module and the maintenance plan of the predictive maintenance module, the remote control module generates remote operation instructions for the power distribution equipment, realizing intelligent management of the entire process from data acquisition to closed-loop control.

[0005] Furthermore, the sensors in the sensing layer include a current sensing unit, a voltage sensing unit, a temperature sensing unit, a power sensing unit, a power factor sensing unit, and an insulation resistance sensing unit. Each sensing unit synchronously collects corresponding parameters at a preset sampling frequency, and the collected data carries a timestamp and device identification information. The timestamp is used to realize the timing alignment processing of the data layer, and the device identification information is used to realize the device association analysis of the application layer. The preset sampling frequency is dynamically adjusted according to the real-time requirements of the parameters.

[0006] Furthermore, the transmission layer adopts a hybrid communication network architecture, including a wireless communication sublayer and a wired communication sublayer. The wireless communication sublayer uses low-power wide area network technology to transmit non-real-time operation data, while the wired communication sublayer uses industrial Ethernet to transmit real-time operation data. The transmission layer encrypts the transmitted data, and the encryption process combines a symmetric encryption algorithm with a device authentication mechanism. The device authentication mechanism is implemented through dual verification using a preset key and a dynamic token.

[0007] Furthermore, the preprocessing process of the data layer includes: Data cleaning: outliers and missing values ​​are removed. Outliers are identified using a deviation detection algorithm based on the 3σ criterion, and missing values ​​are supplemented using a time-series interpolation algorithm based on a sliding window. Normalization transformation: Maps the cleaned data to a preset numerical range. The transformation formula is as follows: ; Where X is the original running data, Xmin is the historical minimum value of this type of data, Xmax is the historical maximum value of this type of data, and X′ is the normalized data. The historical minimum and historical maximum values ​​are dynamically adjusted through a rolling update mechanism.

[0008] Furthermore, the process by which the fault warning module calculates the fault warning value S is as follows: First, calculate the current deviation coefficient based on the current I and the rated current I0: ; Calculate the voltage deviation coefficient based on voltage U and rated voltage U0: ; Calculate the temperature influence coefficient based on the equipment temperature T, normal operating temperature T0, and maximum withstand temperature Tmax: ; Then, the fault warning value is obtained through weighted fusion: ; Wherein, α, β and γ are preset weight coefficients, which are dynamically adjusted according to the type of power distribution equipment using the analytic hierarchy process. For transformer equipment, γ>α>β is satisfied, and for switchgear equipment, α>γ>β is satisfied. The preset early warning threshold is generated by training with historical fault data. The training process integrates the temporal features before the fault and the equipment attribute parameters, and uses the gradient boosting tree algorithm to construct the threshold prediction model.

[0009] Furthermore, the process by which the predictive maintenance module calculates the remaining lifespan L of the equipment is as follows: Based on the equipment design life L0, power influence factor, temperature influence factor and insulation influence factor are introduced for attenuation correction. The correction formula is as follows: ; Wherein, P0 is the rated power, R0 is the initial insulation resistance, k1, k2 and k3 are the attenuation coefficients, and m, n and p are the nonlinear influence indices. The above parameters are obtained by training the life attenuation model corresponding to the equipment type. The life attenuation model integrates the historical operating data and aging test data of the same type of equipment, and uses the Bayesian optimization algorithm to optimize the parameters.

[0010] Furthermore, the process by which the energy consumption analysis module calculates the energy consumption E per unit time is as follows: The calculation is based on the integral operation of instantaneous power, and the formula is as follows: ; Where t1 and t2 are the endpoints of the time interval; the energy consumption optimization strategy is generated by constructing an energy consumption optimization model, the energy consumption optimization model takes the minimum energy consumption per unit output as the objective function, the constraints include the capacity limit of the power distribution system, the power consumption scenario requirements and the equipment operating characteristics, the solution process adopts an improved particle swarm optimization algorithm, the improved particle swarm optimization algorithm introduces adaptive inertia weight and local search operator.

[0011] Furthermore, the instruction generation logic of the remote control module is as follows: when the fault warning value S reaches the first preset threshold, a fault alarm instruction is generated and pushed to the operation and maintenance terminal; when S reaches the second preset threshold, an emergency operation instruction is automatically generated and executed, and a backup power switching mechanism is triggered at the same time; when the remaining lifespan L of the equipment is lower than the preset lifespan threshold, a maintenance plan instruction is generated and associated with the maintenance resource scheduling logic, wherein the maintenance resource scheduling logic dynamically generates a work assignment scheme based on the location of operation and maintenance personnel, spare parts inventory and equipment priority, and the first preset threshold is lower than the second preset threshold.

[0012] Furthermore, the data layer also includes a data storage unit and a data association unit. The data storage unit stores standardized data in chronological order, and the storage period is set according to the importance of the data. The storage period for real-time parameters is not less than a preset duration a1, and the storage period for non-real-time parameters is not less than a preset duration a2. The data association unit establishes an association mapping between operating data and equipment files, maintenance records, and power consumption plans. It constructs a multi-dimensional data association network through knowledge graph technology to provide cross-domain data support for the application layer.

[0013] Furthermore, a data interaction mechanism exists between the modules of the application layer: the early warning result of the fault early warning module serves as a life assessment correction factor for the predictive maintenance module, through a correction coefficient. The remaining lifespan calculation results are dynamically adjusted; the optimization strategy of the energy consumption analysis module serves as the basis for the command generation of the remote control module, forming a closed-loop adjustment from energy consumption optimization to control commands and then to effect feedback. The effect feedback is achieved by comparing the energy consumption E per unit time before and after optimization, thereby improving the overall decision-making accuracy of the system.

[0014] Compared with existing technologies, this invention has the following advantages: The integrated management system for intelligent power distribution synchronously collects multi-dimensional data such as current, voltage, and equipment temperature through the sensing layer. Combined with a weighted fusion algorithm (fusing current / voltage deviation coefficients and temperature influence coefficients) and a dynamic threshold model (trained based on historical data and equipment attributes) in the fault early warning module, it can accurately identify potential anomalies in power distribution equipment. Compared to traditional threshold alarm methods, its early warning accuracy is significantly improved, and it can trigger warnings in advance, greatly shortening fault investigation time and reducing the risk of power outages. The predictive maintenance module calculates the remaining lifespan of equipment based on parameters such as power, temperature, and insulation resistance through a multi-factor lifespan decay model and dynamically generates maintenance plans. This model introduces a nonlinear influence index and a Bayesian optimization algorithm to reduce lifespan prediction errors, avoid the problems of over-maintenance or under-maintenance in traditional periodic maintenance, reduce maintenance costs, and extend the actual service life of equipment. The energy consumption analysis module accurately calculates energy consumption per unit time through instantaneous power integration and generates strategies adapted to power consumption scenarios based on an improved particle swarm optimization algorithm. By combining peak-valley electricity pricing with dynamic adjustment based on load characteristics, energy consumption per unit of product in industrial scenarios can be reduced, energy-saving effects in commercial buildings can be improved, and the efficiency of power resource utilization can be significantly enhanced. The transmission layer adopts a hybrid wireless and wired communication architecture and a dual encryption mechanism to ensure the real-time performance and integrity of data transmission and enhance anti-interference capabilities. The remote control module achieves closed-loop control based on fault early warning and maintenance plans. In emergencies, it can automatically trigger backup power switching to reduce the risk of fault propagation and ensure the continuous and stable operation of the power distribution system. The data layer provides high-quality, multi-dimensional data support for the application layer through preprocessing and knowledge graph association technology. The modules achieve collaborative decision-making through data interaction mechanisms, reduce manual intervention, improve the work efficiency of operation and maintenance personnel, and support global management across devices and scenarios. Attached Figure Description

[0015] Figure 1 This is a system block diagram of the present invention. Detailed Implementation

[0016] The embodiments of the present invention are described in detail below. These embodiments are implemented based on the technical solution of the present invention, and provide detailed implementation methods and specific operation processes. However, the scope of protection of the present invention is not limited to the following embodiments.

[0017] like Figure 1 As shown, this embodiment provides a technical solution: a comprehensive management system for intelligent power distribution, comprising: The sensing layer is equipped with several sensors, which are deployed on the power distribution equipment to collect real-time operating data of the power distribution system. The operating data includes current I, voltage U, power P, equipment temperature T, power factor cosΦ, and insulation resistance R. The transport layer, which is communicatively connected to the perception layer, is used to transmit the running data to the data layer; The data layer, which is communicatively connected to the transport layer, is used to preprocess the running data to generate standardized data. The preprocessing includes data cleaning and normalization transformation. The application layer, which communicates with the data layer, includes a fault early warning module, a predictive maintenance module, an energy consumption analysis module, and a remote control module. The application layer achieves collaborative management by calling the standardized data, specifically: The fault warning module calculates the fault warning value S based on the standardized data of current I, voltage U, and equipment temperature T by integrating time-series features and equipment parameters. When S reaches the preset warning threshold, a fault warning is triggered. The predictive maintenance module calculates the remaining lifespan L of the equipment based on the standardized data of the power P, equipment temperature T, and insulation resistance R, combined with the equipment aging model, and generates a dynamic maintenance plan based on L. The energy consumption analysis module calculates the energy consumption E per unit time based on the standardized data of current I, voltage U, and power factor cosΦ through an energy efficiency evaluation model, and outputs an energy consumption optimization strategy that matches the electricity consumption scenario. Based on the early warning results of the fault early warning module and the maintenance plan of the predictive maintenance module, the remote control module generates remote operation instructions for the power distribution equipment, realizing intelligent management of the entire process from data acquisition to closed-loop control.

[0018] The sensors in the sensing layer include a current sensing unit, a voltage sensing unit, a temperature sensing unit, a power sensing unit, a power factor sensing unit, and an insulation resistance sensing unit. Each sensing unit synchronously collects corresponding parameters at a preset sampling frequency, and the collected data carries a timestamp and device identification information. The timestamp is used to realize the time sequence alignment processing of the data layer, and the device identification information is used to realize the device association analysis of the application layer. The preset sampling frequency is dynamically adjusted according to the real-time requirements of the parameters. By setting up dedicated sensing units for current, voltage, and temperature, targeted collection of multi-dimensional operating parameters of the power distribution system is achieved, ensuring no critical data is missed. Each unit collects data synchronously at a preset frequency, avoiding data analysis biases caused by asynchronous parameter collection, and providing consistent foundational data for subsequent fault warnings and energy consumption calculations. The collected data carries timestamps, allowing the data layer to eliminate time differences in collection from different devices and parameters through time-series alignment processing, ensuring the accuracy of time-series analysis. Device identification information binds data to specific power distribution equipment, avoiding data confusion between multiple devices and providing accurate basis for device correlation analysis at the application layer. The preset sampling frequency is dynamically adjusted according to the real-time requirements of the parameters, ensuring the timeliness of core data while reducing invalid data transmission and storage, lowering bandwidth pressure on the transmission layer and processing load on the data layer, and improving the overall operating efficiency of the system.

[0019] The transmission layer adopts a hybrid communication network architecture, including a wireless communication sublayer and a wired communication sublayer. The wireless communication sublayer uses low-power wide area network technology to transmit non-real-time operation data, while the wired communication sublayer uses industrial Ethernet to transmit real-time operation data. The transmission layer encrypts the transmitted data, and the encryption process uses a combination of symmetric encryption algorithm and device authentication mechanism. The device authentication mechanism is implemented through dual verification using a preset key and a dynamic token. A hybrid wireless and wired communication architecture is adopted. Low-power wide-area network (LPWAN) transmission is used for non-real-time data, balancing cost and coverage; industrial Ethernet transmission is used for real-time data, ensuring high bandwidth and low latency. This enables differentiated transmission of different data types, avoiding the limitations of a single communication method in complex scenarios. Encryption processing combining symmetric encryption algorithms and device authentication mechanisms not only prevents data theft or tampering during transmission but also prevents unauthorized terminal access to the system through device identity verification. This blocks malicious attack paths at the transmission stage, addressing the security vulnerabilities of weak encryption and authentication in traditional power distribution system data transmission.

[0020] The communication sublayer is allocated according to the real-time data requirements to avoid non-real-time data occupying high-priority communication resources and reduce bandwidth waste; the application of low-power wide area networks reduces the communication power consumption of sensors in remote areas and extends the equipment's battery life. At the same time, the directional transmission characteristics of industrial Ethernet reduce the probability of data collisions and retransmissions, improving overall communication efficiency.

[0021] The preprocessing process of the data layer includes: Data cleaning: outliers and missing values ​​are removed. Outliers are identified using a deviation detection algorithm based on the 3σ criterion, and missing values ​​are supplemented using a time-series interpolation algorithm based on a sliding window. Normalization transformation: Maps the cleaned data to a preset numerical range. The transformation formula is as follows: ; Where X is the original running data, Xmin is the historical minimum value of this type of data, Xmax is the historical maximum value of this type of data, and X′ is the normalized data. The historical minimum and historical maximum values ​​are dynamically adjusted through a rolling update mechanism. By using preset data cleaning rules, invalid data generated by interference during the collection process can be filtered out, avoiding the bias caused by dirty data in subsequent analysis; dynamic normalization processing eliminates the dimensional differences between different devices and parameters, making multi-source data comparable and providing an accurate and consistent analytical basis for upper-level modules such as fault early warning and life prediction.

[0022] By using knowledge graph technology to map data such as equipment parameters, historical faults, and maintenance records, a network of connections between equipment, status, and events can be built. This allows scattered data to form meaningful logical chains, helping the system to uncover potential patterns from isolated data and improve the depth and accuracy of decision-making.

[0023] By adopting a rolling updated historical threshold library and a dynamic weight adjustment mechanism, the data layer can automatically iterate over time as the system runs, avoiding outdated analysis problems caused by fixed thresholds, ensuring that data standards always adapt to the current system status, and improving the adaptability to equipment aging and scene changes.

[0024] The hierarchical storage design based on data type enables fast data retrieval and access, reducing data acquisition latency in upper-level modules. At the same time, the pre-processed high-quality data reduces the computational load on upper-level modules and improves the overall system response speed.

[0025] The process by which the fault early warning module calculates the fault early warning value S is as follows: First, calculate the current deviation coefficient based on the current I and the rated current I0: ; Calculate the voltage deviation coefficient based on voltage U and rated voltage U0: ; Calculate the temperature influence coefficient based on the equipment temperature T, normal operating temperature T0, and maximum withstand temperature Tmax: ; Then, the fault warning value is obtained through weighted fusion: ; Wherein, α, β and γ are preset weight coefficients, which are dynamically adjusted according to the type of power distribution equipment using the analytic hierarchy process. For transformer equipment, γ>α>β is satisfied, and for switchgear equipment, α>γ>β is satisfied. The preset early warning threshold is generated by training with historical fault data. The training process integrates the time sequence features before the fault and the equipment attribute parameters, and uses the gradient boosting tree algorithm to construct the threshold prediction model. Through a closed-loop mechanism of real-time monitoring, intelligent diagnosis, and automatic work order dispatch, the system can quickly locate the fault point and cause in the early stages of a fault, and automatically generate a maintenance work order containing the fault type, scope of impact, and processing priority, which is then simultaneously pushed to the maintenance personnel's terminals. Compared with the traditional manual inspection, hierarchical reporting, and passive handling mode, this significantly shortens the fault response time, reduces power outage losses caused by fault escalation, and reduces the blindness of manual troubleshooting.

[0026] Based on energy consumption data preprocessed at the data layer and a dynamic weighting model, the energy consumption analysis module can generate personalized energy-saving solutions for different regions and time periods, and execute them automatically through the remote control module. Simultaneously, it provides real-time feedback on energy-saving effects, forming an optimization closed loop and avoiding localized overload or resource waste caused by traditional one-size-fits-all energy-saving models.

[0027] The remote control module employs a dual verification and hierarchical permission mechanism. Operation commands must be verified by the data layer and the administrator's permissions, and all operations are fully logged. This avoids system risks caused by accidental or unauthorized operations, ensures the traceability of remote control, and solves the security risks of ambiguous permissions and lack of operation records in traditional remote control.

[0028] The predictive maintenance module calculates the remaining lifespan L of the equipment as follows: Based on the equipment design life L0, power influence factor, temperature influence factor and insulation influence factor are introduced for attenuation correction. The correction formula is as follows: ; Where P0 is the rated power, R0 is the initial insulation resistance, k1, k2 and k3 are the attenuation coefficients, and m, n and p are the nonlinear influence indexes. The above parameters are obtained by training the life attenuation model corresponding to the equipment type. The life attenuation model integrates the historical operating data and aging test data of the same type of equipment and uses the Bayesian optimization algorithm to optimize the parameters. Employing a proactive monitoring and coordinated defense mechanism, the system forms a comprehensive security defense covering the network, devices, and personnel through multi-dimensional monitoring including network-layer intrusion detection, device-layer status verification, and operational-layer compliance auditing. When a warning is triggered in any dimension, the system can automatically coordinate with the transmission layer to cut off suspicious communications, the data layer to isolate abnormal data, and the application layer to freeze relevant control commands. This prevents the risk spread caused by the failure of a single protection layer and addresses the vulnerability issues of traditional security modules operating independently. Based on a dynamically updated emergency plan library, the system can automatically match the optimal handling plan in case of emergencies. For example, by combining real-time load distribution data from the perception layer, it can quickly calculate the priority recovery area, coordinate with the remote control module at the application layer to adjust the power supply path, and simultaneously push step-by-step operation guidelines to the operations and maintenance team. Compared to the lag in traditional manual plan formulation, this design shortens the emergency response time and reduces the impact of sudden failures.

[0029] Employing a modular interface design and adaptive protocol conversion mechanism, it is compatible with power distribution equipment from different manufacturers, resolving the data silo problem caused by differences in equipment models. Simultaneously, it reserves interfaces for AI algorithm upgrades and hardware expansion slots, enabling the system to support new feature iterations without a complete system overhaul, reducing later upgrade costs and extending the system's lifespan.

[0030] The energy consumption analysis module calculates the energy consumption E per unit time as follows: The calculation is based on the integral operation of instantaneous power, and the formula is as follows: ; Where t1 and t2 are the endpoints of the time interval; the energy consumption optimization strategy is generated by constructing an energy consumption optimization model, the energy consumption optimization model takes the minimum energy consumption per unit output as the objective function, the constraints include the capacity limit of the power distribution system, the power consumption scenario requirements and the equipment operating characteristics, the solution process adopts an improved particle swarm optimization algorithm, the improved particle swarm optimization algorithm introduces adaptive inertial weights and local search operators; As the system's global hub, this module integrates energy consumption data from the data layer, fault warning results from the application layer, and equipment status information from the security layer. It dynamically adjusts power distribution strategies through multi-objective optimization algorithms: for example, automatically balancing transformer loads in different areas during peak electricity consumption periods; and prioritizing the scheduling of photovoltaic / wind power resources when new energy sources are connected to the grid. Compared to traditional independent regional scheduling, global scheduling reduces overall system energy consumption and minimizes equipment losses caused by load imbalances.

[0031] The module integrates full lifecycle data, including equipment purchase records, installation time, maintenance history, and failure frequency. Using AI algorithms, it predicts the remaining lifespan of the equipment and automatically generates preventative maintenance plans. Simultaneously, it evaluates the cost-effectiveness of the equipment based on long-term operational data, providing a basis for future procurement decisions. This design avoids the reactive approach of post-failure repair, reducing equipment maintenance costs and extending the overall lifespan of the system.

[0032] A layered interactive interface is designed for different users: maintenance personnel can view real-time fault tickets and operation guides; managers can obtain energy consumption analysis reports and optimization suggestions; and residents can query their own household electricity data and energy-saving tips. The interface supports voice control, visual charts (such as load trend curves), and has intelligent question-and-answer functionality. Compared to traditional professional code-based operations, the simplified interactive design shortens the learning time for non-professionals, improving the system's accessibility and ease of use.

[0033] The module interfaces with the smart city platform through standardized interfaces, enabling cross-domain data sharing: for example, synchronizing load forecast data from the power distribution system to the transportation department; receiving electricity demand from the building automation system and adjusting power distribution plans in advance. This cross-system collaboration upgrades power distribution management from an isolated system to a core node in the smart city's energy network, improving the overall energy efficiency of the city.

[0034] When new devices are added to the system or regional functions are adjusted, the load forecasting model and scheduling strategy are automatically updated. For example, after the renovation of a commercial area, the system automatically adjusts peak electricity consumption periods from nighttime to daytime and adapts to higher air conditioning load demands. This solves the cumbersome problem of manual reprogramming required after changes in traditional system scenarios, ensuring that the system always maintains optimal operating status during urban planning adjustments.

[0035] The instruction generation logic of the remote control module is as follows: when the fault warning value S reaches the first preset threshold, a fault alarm instruction is generated and pushed to the operation and maintenance terminal; when S reaches the second preset threshold, an emergency operation instruction is automatically generated and executed, and the backup power switching mechanism is triggered at the same time; when the remaining lifespan L of the equipment is lower than the preset lifespan threshold, a maintenance plan instruction is generated and associated with the maintenance resource scheduling logic. The maintenance resource scheduling logic dynamically generates a work assignment plan based on the location of operation and maintenance personnel, spare parts inventory and equipment priority. The first preset threshold is lower than the second preset threshold. By setting first and second preset thresholds, fault warning values ​​are categorized, and differentiated instructions are generated for different levels of faults: low-level warnings only trigger alarm notifications to avoid excessive intervention; high-level warnings automatically execute emergency operations and activate backup power to quickly contain the spread of the fault. This tiered response mechanism ensures timely handling of emergencies while reducing unnecessary system actions, balancing response speed and stability.

[0036] When a maintenance plan is triggered based on the remaining lifespan of the equipment, the system dynamically generates a work assignment plan by associating maintenance resource scheduling logic with the location of maintenance personnel, spare parts inventory, and equipment priority, thus avoiding blind resource allocation. This transforms maintenance work from passive waiting to proactive planning, reducing maintenance delays and ensuring that critical equipment receives priority support.

[0037] The command generation and execution process forms a complete closed loop: alarm command push ensures relevant personnel are informed in a timely manner; emergency operations are linked to the backup power supply mechanism to reduce the impact of failures; maintenance plans are linked to resource scheduling to ensure maintenance measures are implemented. The fully traceable command flow avoids the loophole of "no tracking after command is issued" in traditional control, improving the reliability and manageability of remote control.

[0038] The data layer also includes a data storage unit and a data association unit. The data storage unit stores standardized data in chronological order, and the storage period is set according to the importance of the data. The storage period for real-time parameters is not less than a preset duration a1, and the storage period for non-real-time parameters is not less than a preset duration a2. The data association unit establishes an association mapping between operating data and equipment files, maintenance records, and power consumption plans. It constructs a multi-dimensional data association network through knowledge graph technology to provide cross-domain data support for the application layer, where a1 < a2. The data storage unit sets storage periods according to data importance, using longer storage periods for real-time parameters to ensure historical data analysis needs, and using appropriate storage periods for non-real-time parameters to avoid redundant storage. This hierarchical storage mechanism ensures the long-term traceability of critical data while reducing unnecessary storage resource consumption, balancing data retention and system load. The data association unit constructs a multi-dimensional network of connections between operational data, equipment files, maintenance records, and power consumption plans using knowledge graph technology, breaking the limitations of information silos in traditional data storage. This enables scattered parameter data to form logically linked chains, providing the application layer with more comprehensive analytical basis and improving the depth and accuracy of decision-making. The multi-dimensional data association network provides cross-domain data support for various modules in the application layer: the fault early warning module can call related historical maintenance records to assist in judging the cause of faults; the energy consumption analysis module can combine power consumption plan data to optimize strategies. This data interoperability enables modules to shift from single data dependence to multi-source data collaboration, improving the overall intelligence level and decision reliability of the system.

[0039] There is a data interaction mechanism between the modules of the application layer: the early warning result of the fault early warning module serves as a life assessment correction factor for the predictive maintenance module, and is adjusted by a correction coefficient. The remaining lifespan calculation results are dynamically adjusted; the optimization strategy of the energy consumption analysis module serves as the basis for the command generation of the remote control module, forming a closed-loop adjustment from energy consumption optimization to control commands and then to effect feedback. The effect feedback is achieved by comparing the energy consumption E per unit time before and after optimization, thereby improving the overall decision-making accuracy of the system. Fault warning results serve as correction factors for predictive maintenance life assessment, while energy consumption optimization strategies provide the basis for remote control commands. This prevents modules from operating independently, creating a collaborative network where data is shared and results are used interchangeably. This avoids the problems of fault warning and maintenance planning being disconnected from energy consumption optimization and actual control in traditional systems, ensuring consistency in the goals of each module and improving the overall consistency of system decision-making. By adjusting the remaining life calculation results in real time through correction coefficients, and combining this with an iterative feedback strategy based on the effects before and after energy consumption optimization, the analysis model can be dynamically calibrated according to the actual operating conditions. Compared to static evaluation using a fixed model, this dynamic correction mechanism makes fault prediction, life assessment, and energy consumption optimization more closely aligned with the actual equipment condition, reducing the deviation between theoretical calculations and actual conditions.

[0040] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0041] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0042] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A comprehensive management system for intelligent power distribution, characterized in that, include: The sensing layer is equipped with several sensors, which are deployed on the power distribution equipment to collect real-time operating data of the power distribution system. The operating data includes current I, voltage U, power P, equipment temperature T, power factor cosΦ, and insulation resistance R. The transport layer is used to transmit the running data to the data layer; The data layer is used to preprocess the running data to generate standardized data. The preprocessing includes data cleaning and normalization transformation. The application layer includes a fault early warning module, a predictive maintenance module, an energy consumption analysis module, and a remote control module. The application layer achieves collaborative management by calling the standardized data, specifically: The fault warning module calculates the fault warning value S based on the standardized data of current I, voltage U, and equipment temperature T by integrating time-series features and equipment parameters. When S reaches the preset warning threshold, a fault warning is triggered. The predictive maintenance module calculates the remaining lifespan L of the equipment based on standardized data of power P, equipment temperature T, and insulation resistance R, combined with the equipment aging model, and generates a dynamic maintenance plan based on L. The energy consumption analysis module calculates the energy consumption E per unit time based on standardized data of current I, voltage U, and power factor cosΦ through an energy efficiency assessment model, and outputs energy consumption optimization strategies that match the electricity consumption scenario. The remote control module generates remote operation commands for the power distribution equipment based on the early warning results of the fault early warning module and the maintenance plan of the predictive maintenance module, thereby realizing intelligent management of the entire process from data acquisition to closed-loop control.

2. The integrated management system for intelligent power distribution according to claim 1, characterized in that: The sensors in the sensing layer include a current sensing unit, a voltage sensing unit, a temperature sensing unit, a power sensing unit, a power factor sensing unit, and an insulation resistance sensing unit. Each sensing unit synchronously collects corresponding parameters at a preset sampling frequency, and the collected data carries a timestamp and device identification information. The timestamp is used to realize the timing alignment processing of the data layer, and the device identification information is used to realize the device association analysis of the application layer. The preset sampling frequency is dynamically adjusted according to the real-time requirements of the parameters.

3. The integrated management system for intelligent power distribution according to claim 1, characterized in that: The transmission layer adopts a hybrid communication network architecture, including a wireless communication sublayer and a wired communication sublayer. The wireless communication sublayer uses low-power wide area network technology to transmit non-real-time operation data, while the wired communication sublayer uses industrial Ethernet to transmit real-time operation data. The transmission layer encrypts the transmitted data, and the encryption process combines a symmetric encryption algorithm with a device authentication mechanism. The device authentication mechanism is implemented through dual verification using a preset key and a dynamic token.

4. The integrated management system for intelligent power distribution according to claim 1, characterized in that: The preprocessing process of the data layer includes: Data cleaning: outliers and missing values ​​are removed. Outliers are identified using a deviation detection algorithm based on the 3σ criterion, and missing values ​​are supplemented using a time-series interpolation algorithm based on a sliding window. Normalization transformation: Maps the cleaned data to a preset numerical range. The transformation formula is as follows: ； Where X is the original running data, Xmin is the historical minimum value of this type of data, Xmax is the historical maximum value of this type of data, and X′ is the normalized data. The historical minimum and historical maximum values ​​are dynamically adjusted through a rolling update mechanism.

5. A comprehensive management system for intelligent power distribution according to claim 1, characterized in that: The process by which the fault early warning module calculates the fault early warning value S is as follows: First, calculate the current deviation coefficient based on the current I and the rated current I0: ； Calculate the voltage deviation coefficient based on voltage U and rated voltage U0: ； Calculate the temperature influence coefficient based on the equipment temperature T, normal operating temperature T0, and maximum withstand temperature Tmax: ； Then, the fault warning value is obtained through weighted fusion: ； Wherein, α, β and γ are preset weight coefficients, which are dynamically adjusted according to the type of power distribution equipment using the analytic hierarchy process. For transformer equipment, γ>α>β is satisfied, and for switchgear equipment, α>γ>β is satisfied. The preset early warning threshold is generated by training with historical fault data. The training process integrates the temporal features before the fault and the equipment attribute parameters, and uses the gradient boosting tree algorithm to construct the threshold prediction model.

6. A comprehensive management system for intelligent power distribution according to claim 1, characterized in that: The predictive maintenance module calculates the remaining lifespan L of the equipment as follows: Based on the equipment design life L0, power influence factor, temperature influence factor and insulation influence factor are introduced for attenuation correction. The correction formula is as follows: ; Wherein, P0 is the rated power, R0 is the initial insulation resistance, k1, k2 and k3 are the attenuation coefficients, and m, n and p are the nonlinear influence indices. The above parameters are obtained by training the life attenuation model corresponding to the equipment type. The life attenuation model integrates the historical operating data and aging test data of the same type of equipment, and uses the Bayesian optimization algorithm to optimize the parameters.

7. A comprehensive management system for intelligent power distribution according to claim 1, characterized in that: The energy consumption analysis module calculates the energy consumption E per unit time as follows: The calculation is based on the integral operation of instantaneous power, and the formula is as follows: ； Where t1 and t2 are the endpoints of the time interval; the energy consumption optimization strategy is generated by constructing an energy consumption optimization model, the energy consumption optimization model takes the minimum energy consumption per unit output as the objective function, the constraints include the capacity limit of the power distribution system, the power consumption scenario requirements and the equipment operating characteristics, the solution process adopts an improved particle swarm optimization algorithm, the improved particle swarm optimization algorithm introduces adaptive inertia weight and local search operator.

8. A comprehensive management system for intelligent power distribution according to claim 1, characterized in that: The instruction generation logic of the remote control module is as follows: when the fault warning value S reaches the first preset threshold, a fault alarm instruction is generated and pushed to the operation and maintenance terminal; when S reaches the second preset threshold, an emergency operation instruction is automatically generated and executed, and the backup power switching mechanism is triggered at the same time; when the remaining lifespan L of the equipment is lower than the preset lifespan threshold, a maintenance plan instruction is generated and associated with the maintenance resource scheduling logic, wherein the first preset threshold is lower than the second preset threshold.

9. A comprehensive management system for intelligent power distribution according to claim 1, characterized in that: The data layer also includes a data storage unit and a data association unit. The data storage unit stores standardized data in chronological order, and the storage period is set according to the importance of the data. The storage period for real-time parameters is not less than a preset duration a1, and the storage period for non-real-time parameters is not less than a preset duration a2. The data association unit establishes an association mapping between operating data and equipment files, maintenance records, and power consumption plans. It constructs a multi-dimensional data association network through knowledge graph technology to provide cross-domain data support for the application layer.

10. A comprehensive management system for intelligent power distribution according to claim 1, characterized in that: There is a data interaction mechanism between the modules of the application layer: the early warning result of the fault early warning module serves as a life assessment correction factor for the predictive maintenance module, and is adjusted by a correction coefficient. The remaining lifespan calculation results are dynamically adjusted; the optimization strategy of the energy consumption analysis module serves as the basis for the command generation of the remote control module, forming a closed-loop regulation from energy consumption optimization to control commands and then to effect feedback.

Images