Power grid operation and maintenance intelligent scheduling system based on big data

Abstract

The invention relates to the technical field of power dispatching management, in particular to a power grid operation and maintenance intelligent dispatching system based on big data, which comprises a data acquisition module, an anomaly identification module, a state grading module, a trend prediction module and a path planning module. According to the method, the dynamic data of the power grid equipment are collected in real time, state monitoring and fault diagnosis of the power equipment are optimized, the abnormal state of the equipment is detected in real time according to the voltage data, the fault time node is recognized, the accuracy and efficiency of fault diagnosis are improved, and the fault diagnosis accuracy and efficiency are improved by combining the voltage phase deviation value and the temperature gradient rising rate of the equipment. According to the method, the equipment state is judged in time, the equipment overheating risk is monitored, the reliability and safety of a power grid system are improved, efficient execution of operation and maintenance tasks is ensured by predicting the equipment aging condition, optimizing the maintenance strategy, reducing the equipment outage risk and utilizing task cluster path planning and resource scheduling, and the power grid operation and maintenance efficiency and the reasonability of resource configuration are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power dispatching management, and particularly to an intelligent dispatching system for power grid operation and maintenance based on big data. Background Art

[0002] The technical field of power dispatching management includes real-time dispatching monitoring and operation and maintenance management of power production, transmission, and power consumption during the production and operation process of the power system. The core content of this technical field includes unified dispatching control and management of power grid operation status monitoring, power load regulation, electric energy distribution and transmission, and related equipment maintenance, etc., to achieve the safety, stability, and reliability of the power grid system operation. Specifically, it involves power grid operation and maintenance dispatching plan arrangement, real-time monitoring, fault diagnosis, processing dispatching instructions, and various technical matters, to comprehensively master the power grid operation status and coordinate the control of multiple links of electric energy production, transmission, and distribution, ensuring the safety and stability of power supply.

[0003] Among them, the intelligent dispatching system for power grid operation and maintenance based on big data refers to collecting multi-dimensional data such as the operation status of power equipment, historical fault records, and load change conditions, and performing processing such as feature extraction, clustering analysis, and distribution modeling on the data, establishing a judgment logic and dispatching rules for operation and maintenance status, supporting the dynamic allocation of dispatching tasks and the optimization of response processes. The technical matters targeted by the system cover multiple links such as data source interface access, data caching processing, task priority setting, work order generation mechanism, and dispatching path planning, and classify and clean the original data through data warehouse technology to support the condition matching of the dispatching decision logic module and the logic execution of the dispatching process control module.

[0004] The traditional intelligent dispatching management technology for power grid operation and maintenance has insufficient timeliness in real-time monitoring of power grid operation and fault response. In the face of equipment failures or emergencies, it lacks sufficient intelligent dispatching decision-making capabilities, relies on static data collection and analysis methods, and is difficult to obtain and analyze dynamic data of equipment in real-time, resulting in longer response times during power load changes and equipment failures, affecting the stability of the power grid. The fault diagnosis and processing mechanism rely on manual intervention and lack automated and intelligent judgment capabilities, resulting in low fault handling efficiency, leading to situations of dispatching lag or decision-making errors, and being unable to dynamically adjust operation and maintenance tasks according to real-time data. It often conducts dispatching based on preset templates and is unable to flexibly adjust task priorities and the configuration of work resources according to changes in equipment status, resulting in the formulation of maintenance plans often lagging behind actual needs. Summary of the Invention

[0005] The purpose of the present invention is to solve the deficiencies existing in the prior art and propose an intelligent dispatching system for power grid operation and maintenance based on big data.

[0006] To achieve the above object, the present invention adopts the following technical solutions: An intelligent dispatching system for power grid operation and maintenance based on big data includes: The data acquisition module obtains power grid operation data, extracts the current amplitude, voltage volatility, and active power change of multiple power grid devices, analyzes the volatility of each type of data in real time, calculates the variation amplitude of the operation data, adjusts the data sampling frequency, and obtains a power parameter data set; The anomaly recognition module calls the power parameter data set, extracts the voltage monitoring data sequence of multiple power grid devices, calculates the voltage change slope, and calculates the slope difference of the voltage curve in a continuous time period, detects abnormal voltage data and abnormal power equipment, identifies the abnormal time node, and obtains an abnormal data location record; The status classification module, based on the abnormal data location record and the power parameter data set, detects the voltage phase offset value and the temperature gradient rise rate of the power equipment in real time, analyzes the equipment status, and obtains the equipment status classification level; The trend prediction module, based on the equipment status classification level, extracts the leakage current and operating voltage of the equipment in real time, calculates the insulation degradation factor of the equipment, combines the temperature gradient rise rate of the equipment, evaluates the aging status of the equipment, and calculates the remaining life cycle of the equipment to obtain a maintenance trigger time window.

[0007] As a further solution of the present invention, the power parameter data set includes an equipment voltage waveform data set, leakage current detection data, and an equipment operating temperature data set. The abnormal data location record is specifically an abnormal time node, abnormal voltage data, and abnormal power equipment. The equipment status classification level is specifically an equipment health status level, voltage phase offset value, and temperature gradient rise rate. The maintenance trigger time window is specifically the remaining life cycle of the equipment, insulation degradation factor, and maintenance trigger time.

[0008] As a further solution of the present invention, the data acquisition module includes: The power parameter extraction sub-module obtains power grid operation data, and by extracting the current amplitude, voltage volatility, and active power change data of multiple power grid devices, combines the equipment number and time stamp to obtain equipment operation characteristic data; The load variation analysis sub-module calls the equipment operation characteristic data and uses the formula: ; Calculates the data fluctuation determination value within each sliding window period to obtain a variation amplitude parameter; Among them, is the total number of samples in the sampling window, is the data point index, is the value of the th current amplitude data point, is the average value of the current amplitude, is the value of the th active power change data point, is the average value of the active power change, is the th voltage fluctuation rate data point value, is the average value of the voltage fluctuation rate, is the data fluctuation determination value; The sampling frequency control sub-module calls the mutation amplitude parameter, adjusts the data acquisition frequency in real time according to the mutation amplitude of the device operation data, and obtains the voltage waveforms, leakage currents, and operating temperature data of multiple operating devices to construct a power parameter data set.

[0009] As a further solution of the present invention, the anomaly recognition module includes: The voltage data extraction sub-module calls the power parameter data set, extracts the voltage values of each power device in the power grid at multiple time points, combines the device number and the timestamp to obtain a voltage monitoring data sequence; The voltage slope calculation sub-module calls the voltage monitoring data sequence, calculates the voltage change slope of the device in each time period according to the voltage difference between adjacent moments, and calculates the slope difference of consecutive time periods to obtain the voltage slope difference; The abnormal node recognition sub-module calls the voltage slope difference, detects abnormal voltage data in real time, combines the device number and the timestamp, marks abnormal power events, and obtains abnormal data location records.

[0010] As a further solution of the present invention, the state classification module includes: The phase shift detection sub-module calls the abnormal data location record and the power parameter data set, extracts the absolute value of the phase difference between each cycle voltage waveform and the standard sine wave according to the voltage waveform sequence, calculates the sliding window mean of the phase difference of consecutive cycles, calculates the sliding average of the absolute value of the cycle phase difference, combines the identified operating temperature data, and calculates the temperature gradient rise rate to obtain a set of state input parameters; The state judgment calculation sub-module calls the set of state input parameters, calculates the temperature rise increment and voltage fluctuation intensity of adjacent cycles, combines the real-time active power of the device, and uses the formula: ; Calculate the state comprehensive determination value; Among them, represents the state comprehensive determination value, represents the voltage phase shift value, represents the temperature gradient value, represents the real-time active power of the device, represents the temperature rise increment of adjacent cycles, Indicates the voltage fluctuation intensity of the current cycle; Based on the state comprehensive determination value, the risk level classification sub-module combines a preset state scoring interval, identifies the device state, and matches the device state label in real time to establish a device state classification level.

[0011] As a further solution of the present invention, the trend prediction module includes: The insulation degradation extraction sub-module calls the device state classification level, calls the leakage current and operating voltage data of the power equipment, obtains the average value of the leakage current and the average amplitude of the voltage, and obtains the insulation degradation factor value by analyzing the change rate of the leakage current and the change trend of the voltage amplitude, combined with the voltage deviation direction; Based on the insulation degradation factor value, the aging state evaluation sub-module extracts the periodic temperature gradient change, obtains the temperature growth amount and time span of multiple cycles, combines the device voltage fluctuation parameters, calculates the temperature rise speed and the voltage deviation ratio fluctuation level, and uses the formula: ; Calculate the device aging trend offset amount, and perform an average evaluation on the fluctuation amplitude of consecutive cycles to obtain the comprehensive aging state index; Among them, Indicates the comprehensive aging state index, Indicates the Leakage current of the cycle, Indicates the temperature gradient, Indicates the operating voltage deviation value, Indicates the temperature rise change value, Indicates the cycle time span, Indicates the number of monitoring cycles, Is the index number of the monitoring cycle; The life cycle calculation and estimation sub-module calculates the remaining life cycle of the device according to the comprehensive aging state index, retrieves the standard life benchmark parameters of each device, and combines the current time node to obtain the maintenance trigger time window.

[0012] As a further solution of the present invention, the system further includes: The path planning module calls the abnormal data location record and the maintenance trigger time window, constructs an operation and maintenance work order according to the type, location information, and required tool type of the abnormal device, calculates the job attribute similarity of multiple work orders, and combines the geographical coordinates of the device to plan the task cluster path and obtain the power grid operation and maintenance scheduling parameters; The power grid operation and maintenance scheduling parameters are specifically the operation and maintenance work order construction record, the task cluster path, and the job attribute similarity.

[0013] As a further solution of the present invention, the path planning module includes: The operation and maintenance work order construction sub-module locates and records according to the abnormal data and the maintenance trigger time window, constructs an operation and maintenance work order based on the type, location information, and required tool type of the abnormal device, and obtains a list of operation and maintenance work orders; The job attribute similarity calculation sub-module extracts the job attributes of each work order based on the list of operation and maintenance work orders, including tool requirements, voltage level, and job risk level, and uses the formula: ; Calculate the job attribute similarity between multiple work orders; Among them, represents the similarity value between work order and work order , represents the value of work order on the th job attribute, represents the value of work order on the th job attribute, is the average value of work order on all job attributes, is the average value of work order on all job attributes, is the total number of job attributes, is the index of the job attribute; The task cluster path planning sub-module plans the task cluster path based on the job attribute similarity, combines the geographical coordinates of the device, the priority of the task, and the resource requirements, and obtains the grid operation and maintenance scheduling path parameters.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In the present invention, by collecting the dynamic data of grid devices in real time, the state monitoring and fault diagnosis of power equipment are optimized. The abnormal state of the device is detected in real time according to the voltage data, and the fault time node is identified, improving the accuracy and efficiency of fault diagnosis. Combining the voltage phase offset value and the temperature gradient rise rate of the device, the device state is judged in time, and the overheating risk of the device is monitored, enhancing the reliability and security of the grid system. By predicting the aging condition of the device, the maintenance strategy is optimized, reducing the risk of device outage. Using task cluster path planning and resource scheduling, the efficient execution of operation and maintenance tasks is ensured, improving the operation and maintenance efficiency of the grid and the rationality of resource allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is the system flow chart of the present invention; Figure 2 is the flow chart of the data acquisition module of the present invention; Figure 3 is the flow chart of the abnormal identification module of the present invention; Figure 4 It is the flowchart of the state classification module of the present invention; Figure 5 It is the flowchart of the trend prediction module of the present invention; Figure 6 It is the flowchart of the path planning module of the present invention. Specific embodiments

[0016] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0017] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. In addition, in the description of the present invention, "a plurality of" means two or more, unless otherwise specifically defined.

[0018] Please refer to Figure 1 , an intelligent dispatching system for power grid operation and maintenance based on big data includes: The data acquisition module obtains power grid operation data, extracts the current amplitude, voltage volatility, and active power change amount of multiple power grid devices, analyzes the volatility of each type of data in real time, calculates the variation amplitude of the operation data, adjusts the data sampling frequency, and obtains a power parameter data set; The anomaly recognition module calls the power parameter data set, extracts the voltage monitoring data sequences of multiple power grid devices, calculates the voltage change slope, and calculates the slope difference of the voltage curve in a continuous time period, detects abnormal voltage data and abnormal power equipment, identifies the abnormal time node, and obtains the anomaly data location record; The state classification module, based on the anomaly data location record and the power parameter data set, detects the voltage phase offset value and the temperature gradient rise rate of the power equipment in real time, analyzes the equipment state, and obtains the equipment state classification level; The trend prediction module, based on the equipment state classification level, extracts the leakage current and operating voltage of the equipment in real time, calculates the insulation degradation factor of the equipment, combines the temperature gradient rise rate of the equipment, evaluates the aging state of the equipment, and calculates the remaining life cycle of the equipment to obtain the maintenance trigger time window; The path planning module calls the abnormal data location record and the maintenance trigger time window, constructs an operation and maintenance work order based on the type, location information, and required tool type of the abnormal device, calculates the similarity of the operation attributes of multiple work orders, and combines the geographical coordinates of the device to plan the path of the task cluster and obtain the power grid operation and maintenance scheduling parameters.

[0019] The power parameter data set includes the device voltage waveform data set, the leakage current detection data, and the device operating temperature data set. The abnormal data location record specifically includes the abnormal time node, the abnormal voltage data, and the abnormal power equipment. The device status classification level specifically includes the device health status level, the voltage phase offset value, and the temperature gradient rise rate. The maintenance trigger time window specifically includes the remaining life cycle of the device, the insulation degradation factor, and the maintenance trigger time. The power grid operation and maintenance scheduling parameters specifically include the operation and maintenance work order construction record, the task cluster path, and the similarity of the operation attributes.

[0020] Please refer to Figure 2 , the data acquisition module includes: The power parameter extraction sub-module obtains the power grid operation data, and obtains the device operation characteristic data by extracting the current amplitude, voltage volatility, and active power change data of multiple power grid devices and combining the device number and time stamp. First, extract the current amplitude, voltage volatility, and active power change data from each power grid device, and record them in combination with the device number and time stamp. This module collects the electrical parameters of the device through periodic monitoring to form the operation characteristic data of the device. For each device, the current amplitude, voltage volatility, and active power change will be recorded as a time series to form a device characteristic data set. In practical applications, assume that the current amplitude of device A at a certain moment is 10A, the voltage volatility is 5V, and the active power change is 2kW. These data are used to describe the operation state of the device at this time point. In this way, each device can generate a power parameter data set, which contains information such as the device number, time stamp, current amplitude, voltage volatility, and active power change. Using the formula: ; Calculate the power parameter data set of the device, where is the th current amplitude data point, is the th voltage volatility data point, is the th active power change data point. Assume that the current amplitude of device A is 10A, the voltage volatility is 5V, and the active power change is 2kW. Substituting into the formula, the power parameter data set is obtained: ; The results show that the power parameter dataset of the device contains a current amplitude of 10 A, a voltage volatility of 5 V, and a change in active power of 2 kW. These data will provide the basic data for subsequent analysis.

[0021] The load variation analysis sub-module calls the device operation characteristic data and uses the formula: ; Calculate the data fluctuation determination value within each sliding window period to obtain the variation amplitude parameter; Among them, is the total number of samples in the sampling window, is the data point index, is the th value of the current amplitude data point, is the average value of the current amplitude, is the th value of the active power change data point, is the average value of the active power change, is the th value of the voltage volatility data point, is the average value of the voltage volatility, is the data fluctuation determination value; In the load variation analysis sub-module, the system uses the device operation characteristic data obtained from the data acquisition sub-module to further analyze the changes in the current amplitude, voltage volatility, and active power change. The system performs difference processing on these data according to time windows, and evaluates the volatility of the device state by calculating the data fluctuation determination value within each sliding window. For each sliding window, the system calculates the differences in the current amplitude, active power change, and voltage volatility, and combines them with the formula for calculation to obtain the fluctuation determination value. Suppose within a sliding window period, the current amplitude changes by 5 A, the active power change changes by 0.5 kW, and the voltage volatility changes by 1 V. Using the formula: ; Suppose within a certain sliding window, the current amplitude changes by 5 A, the active power changes by 0.5 kW, the voltage volatility changes by 1 V, and the average values of the current amplitude, power, and voltage volatility within the window are 10 A, 2 kW, and 5 V respectively. Substitute them into the formula for calculation: ; ; The calculation results show that the fluctuation determination value is 1.45. By comparing this value with the set threshold, the system can further evaluate the trend of the device state change, and thus provide a basis for sampling frequency adjustment and device maintenance.

[0022] The sampling frequency control sub-module calls the mutation amplitude parameter, adjusts the data acquisition frequency in real time according to the mutation amplitude of the device operation data, and obtains the voltage waveforms, leakage currents, and operating temperature data of multiple operating devices to construct a power parameter data set; In the sampling frequency control sub-module, the system adjusts the data acquisition frequency according to the mutation amplitude parameter. The magnitude of the mutation amplitude determines the volatility of the device operation state. When the volatility is large, the system increases the sampling frequency to capture the device state changes in real time. The system compares the mutation amplitude parameter with the set threshold. If the volatility exceeds the set threshold, the sampling frequency will increase; if the volatility is less than the threshold, the current frequency will be maintained. Suppose the mutation amplitude of device A is 1.45 and the set threshold is 1.0. The system will adjust the sampling period according to this difference. Suppose the original sampling period is 10 seconds. The system adjusts the sampling period to 1.5 times the original, which becomes 7 seconds. At this time, the system updates the power parameter data set according to the sampling period. Using the formula: ; Calculate the new sampling period, where, is the new sampling period, is the original sampling period, is the mutation amplitude, is the set threshold. Suppose the original sampling period is 10 seconds, the mutation amplitude is 1.45, and the threshold is 1.0. Substitute into the formula for calculation: ; The calculation results show that the new sampling period is 24.5 seconds. According to this sampling frequency, the system adjusts and obtains the voltage waveforms, leakage currents, and operating temperature data of multiple operating devices, and finally constructs a power parameter data set.

[0023] Please refer to Figure 3 , the anomaly recognition module includes: The voltage data extraction sub-module calls the power parameter data set, extracts the voltage values of each power device in the power grid at multiple time points, combines the device number and the timestamp to obtain the voltage monitoring data sequence; The voltage data extraction sub-module starts execution by extracting the voltage values of grid devices at different time points from the power parameter dataset. These voltage values are marked and organized according to the device number and timestamp for subsequent analysis. Specifically, the voltage data extraction sub-module obtains the voltage sequence from the operation data of each grid device, and these data points can span multiple time periods and devices. For example, the voltage value of device A in a certain time period is 230V, and the voltage value of device B is 234V. These voltage values will be classified according to the device number and timestamp, and finally form a voltage monitoring data sequence. This data sequence not only contains voltage values, but also other grid operation data, such as current amplitude and power data, thus providing a complete data source for subsequent voltage change analysis. In this way, the system can provide voltage data for each device at each monitoring time point, generate a voltage monitoring data sequence for further analysis.

[0024] The voltage slope calculation sub-module calls the voltage monitoring data sequence, calculates the voltage change slope of the device in each time period according to the voltage difference between adjacent moments, and calculates the slope difference of consecutive time periods to obtain the voltage slope difference; The voltage slope calculation sub-module calculates the voltage change slope of the device in each time period according to the voltage monitoring data sequence. By calculating the ratio of the voltage difference between each pair of adjacent time points to the time interval, the speed of voltage change is obtained. For example, between the time points and , the voltage change amount is , and the time interval is . The system calculates that the voltage change slope is 4V / s. For the next time point pair, the system will continue to use the same method to calculate the slope and summarize the slope values of each time period. Using the formula: ; Calculate the voltage change slope, where is the voltage change slope, are the voltage values of two time points, are the timestamps of two time points. Assuming the voltage data is 230V, 234V, 233V, and the time interval is 1 second, substitute into the formula to calculate the slope: ; The calculation result shows that the voltage change slope is 4V / s. Continue to apply this calculation method to other data points in the voltage monitoring data sequence to generate voltage change slope data.

[0025] The abnormal node identification sub-module calls the voltage slope difference, real-time detects abnormal voltage data, combines the device number and timestamp, marks abnormal power events, and obtains abnormal data location records; The abnormal node recognition sub-module identifies abnormal event nodes during the voltage change process based on the voltage slope difference. The system determines whether there is an abnormal event according to the mutation of the voltage change slope. Specifically, when the voltage change slope mutates, for example, the voltage slope suddenly jumps from a normal 1V / s to 10V / s, the system marks this time point as an abnormal node. When identifying abnormal nodes, the system calculates the voltage slope difference for consecutive time periods and monitors whether there are abnormal fluctuations. If the slope difference is found to be greater than the preset threshold, the system marks this time period as an abnormal node and records it. Using the formula: ; Calculate the voltage slope difference, where, is the voltage slope difference, is the voltage slope of the th time period, is the voltage slope of the previous time period. Suppose in a certain time period, the voltage slope is 1V / s, and the voltage slope in the next time period is 10V / s. Substitute into the formula for calculation: ; The calculation result shows that the voltage slope difference is 9V / s, exceeding the preset threshold, and the system marks this time point as an abnormal node. Through this process, the system can identify abnormal event nodes in the power grid and generate abnormal data location records.

[0026] Please refer to Figure 4 , the status classification module includes: The phase offset detection sub-module calls the abnormal data location record and the power parameter data set, extracts the absolute value of the phase difference between each cycle voltage waveform and the standard sine wave according to the voltage waveform sequence, calculates the sliding window mean of the phase difference for consecutive cycles, calculates the sliding average of the absolute value of the cycle phase difference, combines the identified operating temperature data, calculates the temperature gradient rise rate, and obtains the set of status input parameters; After calling the abnormal data location record and the power parameter data set, first extract the complete voltage waveform sequence of the target power device within a unit cycle, number its zero-crossing points, so as to mark the phase structure of each cycle signal, calculate the relative phase offset by comparing the zero-crossing point position within the sampling cycle with the standard sine wave, use the whole-cycle interpolation calculation method to obtain the absolute value of the phase difference for each cycle, and then form a sequence of phase differences for consecutive cycles. Subsequently, take every 5 cycles as a sliding window, and successively calculate the average value of its phase offset. Perform the same sliding processing operation on the temperature data, extract the temperature difference between adjacent cycles and calculate the average temperature rise rate, and then integrate the two to form the set of device status input parameters. To quantitatively analyze the stability of the phase offset, the following formula is set: ; Calculate the average phase offset of the sliding window, where is the average phase offset value, is the measured phase value of the cycle, is the ideal sine wave phase value, is the number of cycles of the sliding window.

[0027] Set: , , , , , , Substitute into the calculation: , The calculation result shows that the average phase offset value of the device in the current sliding cycle is 4.1 degrees, which combines with the temperature rise rate to form a set of state input parameters for subsequent state judgment; If you need to further observe the change trend of the phase offset, you can continuously roll the sampling cycle and construct a phase sliding curve graph, marking the offset rising mutation point for temperature coupling comparison.

[0028] The state judgment calculation sub-module calls the set of state input parameters, calculates the temperature rise increment and voltage fluctuation intensity of adjacent cycles, and combines with the real-time active power of the device. Using the formula: ; Calculate the state comprehensive judgment value; Among them, represents the state comprehensive judgment value, represents the voltage phase offset value, represents the temperature gradient value, represents the real-time active power of the device, represents the temperature rise increment of adjacent cycles, represents the voltage fluctuation intensity of the current cycle; Based on the obtained set of state input parameters, further calculate the comprehensive influence degree value among the voltage phase offset, voltage fluctuation intensity, active power and temperature rise index. First, extract the real-time active power values of the device within a continuous time period , temperature rise gradient , the voltage fluctuation intensity of the current cycle and the temperature rise increment of adjacent cycles , and construct the following comprehensive state scoring formula: ; Used to calculate the state comprehensive judgment value of the device in the current cycle. Among them, is the state comprehensive judgment value, is the phase offset value, is the real-time active power, is the voltage fluctuation intensity, is the temperature gradient, is the temperature rise increment; Set , , , , , substitute into the calculation: ; The calculation result is the comprehensive state determination value of 33.27, and this value is used as a reference for scoring and enters the next level classification process; After the execution of this sub-module, the digital determination of the device state is expressed, providing input conditions for the state level classification.

[0029] Based on the comprehensive state determination value, the risk level classification sub-module combines the preset state scoring intervals, identifies the device state, and matches the device state label in real time to establish the device state classification level; After calling the comprehensive state determination value obtained by the previous module, first establish a corresponding device number index table for calibrating the device to which each state value belongs. In the state level classification stage, the system needs to pre-set a set of state scoring intervals, which are divided according to factors such as device type, voltage level, and operating environment, generally divided into four levels, respectively representing four levels of device operating states from excellent to relatively high risk. The setting of the scoring intervals needs to be combined with the grid operation standards and engineering practice experience to ensure that the evaluation is representative. According to the interval boundaries set by the standard, the comprehensive state determination value is respectively compared with all level boundary values for interval matching judgment. When the state value is less than the lowest boundary value, the system automatically classifies it as level I, indicating that the device is in a stable operating state; if the value is in the second level interval, it is classified as level II, indicating that there is a slight unstable trend in the device operation; the device with the state value in the third level interval is classified as level III, and regular inspections need to be arranged to observe the operation change trend; when the state value reaches or exceeds the highest interval value, the device is classified as level IV, representing significant temperature rise offset or voltage fluctuation problems, and should be included in the key monitoring list and early intervention in maintenance. During the classification process, each state value needs to have a clear and unique level label, and be structurally bound with data such as device ID number, time stamp, and device type to form a complete data record. This record is stored in the state level database for subsequent system modules such as visual display, trend tracking, and maintenance task scheduling to call. After the classification process is completed, the system synchronously generates a device operation state classification list, which is grouped and summarized according to levels, facilitating the operation and maintenance management personnel to formulate subsequent intervention plans according to the risk priority, thus completing the unified classification of all target device states and obtaining the device state classification level.

[0030] Please refer to Figure 5 , the trend prediction module includes: The insulation degradation extraction sub-module calls the device status classification level, calls the leakage current and operating voltage data of the power equipment, obtains the average value of the leakage current and the average amplitude of the voltage, and obtains the insulation degradation factor value by analyzing the change rate of the leakage current and the change trend of the voltage amplitude, combined with the voltage deviation direction; In the insulation degradation extraction sub-module, the system first retrieves the status classification level of the power equipment and obtains the data of the leakage current and the operating voltage. By analyzing these data, the system can calculate the average value of the leakage current and the average value of the voltage amplitude. Next, by analyzing the change rate of the leakage current and the change trend of the voltage amplitude, the system can obtain a preliminary signal about the degradation of the equipment's insulation material. Specifically, when the change in the leakage current is large, it means that the insulation performance of the equipment is gradually degrading; and the trend of voltage fluctuation can help confirm the directionality of this degradation, such as the degree and persistence of the voltage deviation from the normal value. Combining these two data, the insulation degradation factor value is finally calculated, which reflects the health level of the equipment's insulation state and can be used for decision support in equipment maintenance. For example, in a monitoring period, assuming that the change rate of the leakage current is 5% and the voltage fluctuation amplitude is 2%, the system will combine these two parameters to obtain a certain degradation factor. Using the formula: ; Calculate the insulation degradation factor, where is the insulation degradation factor, is the average value of the leakage current, is the average value of the voltage amplitude. Assuming , , substitute the set values for calculation: ; The calculation results show that the insulation degradation factor of the equipment is 0.91%, and this low value indicates that the insulation performance of the equipment is still in a healthy state, indicating that no serious degradation problems have occurred during the operation of the equipment. Next, the system will conduct equipment health assessment based on this value and predict subsequent maintenance requirements.

[0031] The aging state assessment sub-module, based on the insulation degradation factor value, extracts the periodic temperature gradient change, obtains the temperature growth amount and time span of multiple periods, combines the equipment voltage fluctuation parameters, calculates the temperature rise speed and the voltage deviation ratio fluctuation level, using the formula: ; Calculate the equipment aging trend offset, and conduct an average assessment of the fluctuation amplitude of consecutive periods to obtain the comprehensive aging state index; where Represents the comprehensive index of the aging state, Represents the leakage current of the cycle, Represents the temperature gradient, Represents the deviation value of the operating voltage, Represents the change value of the temperature rise, Represents the cycle time span, Represents the number of monitoring cycles, Is the index number of the monitoring cycle; In the aging state evaluation sub-module, based on the extracted insulation degradation factor values, the system further calculates the temperature rise increment and time span in multiple cycles. By extracting the temperature change data within the cycle, the system can determine whether there is a risk of excessive temperature rise in the equipment. Combining the voltage fluctuation parameters, the relationship between the temperature rise rate and voltage fluctuation is calculated to understand the impact of the equipment's heat reception on aging. For example, when the voltage fluctuation of the equipment intensifies, the rate of temperature rise will also accelerate, ultimately leading to an acceleration of equipment aging. In this way, the system can obtain the comprehensive index of the equipment's aging state, which is used to evaluate the health status of the equipment in different cycles. Assuming that the temperature rise change amount of the equipment in the first cycle is 3°C and the voltage fluctuation is 5%, the temperature rise rate can be calculated and the aging process of the equipment can be inferred. Using the formula: ; Set to 5, indicating that the data of 5 cycles are used for calculation, IL is , in units of A, TG (temperature gradient) is , in units of °C, UV (voltage deviation value) is , in units of V, (temperature rise change amount) is , in units of °C, (time span) is , in units of hours. Calculation process: For cycle 1( ): ; ; ; For cycle 2( ): ; ; ; For cycle 3( ): ; ; ; ; ; ; For cycle 5 ( ): ; ; ; ; ; The comprehensive index DA of the resulting aging state is 1.810191016. This result indicates that within the past 5 cycles, the comprehensive aging value of the equipment is 1.81, indicating that the health state of the equipment is good or further monitoring is required.

[0032] The life cycle calculation sub-module retrieves the standard life benchmark parameters of each type of equipment according to the comprehensive aging state index, calculates the remaining life cycle of the equipment in combination with the current time node, and obtains the maintenance trigger time window; In the life cycle calculation sub-module, the system calculates the current used life quantity of the equipment according to the comprehensive aging state index in combination with the standard life parameters of the equipment type, and then combines the current cumulative operation time to calculate the remaining operable cycles of the equipment and output the maintenance trigger time window. First, the system calls the standard life parameters of each piece of equipment. For example, the life standard of a large transformer is set to 36000 hours, and obtains the actual operation duration of the equipment. Through the comprehensive aging state index maps the life consumption multiple per unit time currently, forming a basis for acceleration or deceleration compared with the normal aging rate. Specifically, the system multiplies the already-operated duration by the life acceleration multiple represented by the aging factor at the current moment to calculate the current equivalent consumed life, and then subtracts the equivalent consumed life from the total life to obtain the remaining life cycle. Using the formula: ; Calculate the remaining life cycle, where, is the remaining life cycle of the equipment, is the standard life cycle of the equipment, is the current cumulative operation time of the equipment, is the comprehensive aging state index, is the equipment aging gain coefficient. Set , , , , substitute the set value for calculation: ; The calculation results show that the remaining life cycle of the equipment is 19,200 hours. Based on this value and the operation plan, the system sets a reasonable maintenance trigger time window.

[0033] Please refer to Figure 6 , the path planning module includes: The operation and maintenance work order construction sub-module locates records according to the abnormal data and the maintenance trigger time window, and constructs operation and maintenance work orders based on the type, location information, and required tool type of the abnormal equipment to obtain a list of operation and maintenance work orders; In the operation and maintenance work order construction sub-module, first, relevant data needs to be extracted according to the aforementioned abnormal data location records and the maintenance trigger time window to identify the time node when the equipment fails, and to locate the fault type and location of the equipment. Through these two pieces of information, the system can accurately identify the specific equipment that needs to be repaired in the power grid and the specific fault content of each piece of equipment. Subsequently, the system automatically generates operation and maintenance work orders according to the type, location of the faulty equipment, and the nature of the maintenance tasks. These work orders include equipment location information, equipment fault types, required repair tools, operation time required, and priorities, etc. Maintenance personnel can quickly understand the detailed information of the maintenance tasks through the work orders, thereby improving work efficiency. Suppose a transformer in the power grid fails, the system will automatically generate a work order containing fault description, repair requirements, required tools, repair time, etc. according to its location and fault type. If the equipment has abnormal voltage fluctuations and the insulation system needs to be inspected, the system will specify the required tools as "insulation detector" and "voltage tester", and arrange relevant staff to go to the designated location to inspect and repair the equipment according to the work order requirements. During this process, the generation of the maintenance work order ensures the timely and accurate execution of the maintenance work, not only improving the maintenance efficiency but also reducing the probability of misoperation. Finally, all generated work orders will be assigned to the maintenance team according to the priority to ensure the stable operation of the system and prepare for subsequent work. Obtain the list of operation and maintenance work orders as the final result.

[0034] The operation attribute similarity calculation sub-module extracts the operation attributes of each work order based on the list of operation and maintenance work orders, including tool requirements, voltage level, and operation risk level, and uses the formula: ; Calculate the operation attribute similarity between multiple work orders; Among them, represents the similarity value between work order and work order , represents the value of work order on the th operation attribute, Representative work order The value on the th job attribute is the mean value of the work order on all job attributes, is the mean value of the work order on all job attributes, is the total number of job attributes, is the index of the job attribute; During the execution of this sub-module, it is first necessary to extract the job attributes of each work order from the aforementioned operation and maintenance work order list. The main job attributes include tool requirements, voltage level, and job risk level. For each pair of work orders, the system calculates the differences in each job attribute between them, and then obtains the similarity value between the two work orders. The core of this process is to transform the job attribute differences of each pair of work orders into a comparable similarity score through standardization calculations, reflecting the similarity of tasks. Using the similarity value, the system can reasonably allocate resources, optimize the task execution order, and improve efficiency by minimizing resource waste. Using the formula: ; Calculate the job similarity. Assume the job attributes of work order A and work order B are as follows: for work order A, the tool requirement level is 2, the voltage level is 10 kV, and the risk level is high; for work order B, the tool requirement level is 3, the voltage level is 10 kV, and the risk level is medium. The calculation process is as follows: , , , Obtain the job attribute similarities among multiple work orders.

[0035] Based on the job attribute similarities and combined with the geographical coordinates of the equipment, the priorities of the tasks, and the resource requirements, the task cluster path planning sub-module plans the task cluster path and obtains the power grid operation and maintenance scheduling path parameters; During the execution of this sub-module, first based on the job attribute similarities calculated above and combined with information such as the geographical coordinates, task priorities, and required resources of each task, the system will classify the tasks into different clusters through clustering analysis. The tasks within each cluster usually have similar job requirements and geographical locations, so they can be completed together to improve resource utilization and save time. During the clustering analysis process, the system considers multiple factors, including the urgency of the tasks, the tools required for maintenance, the execution time of the tasks, etc. These factors work together to ensure that the task clusters can be optimized and scheduled according to the actual situation. Using the clustering analysis formula: ; Among them, is the distance metric between task X and task Y, and is the value of task X and Y on the th job attribute, is the total number of job attributes. Through the clustering analysis algorithm, the system clusters tasks to ensure that tasks with high priority and high similarity are physically close, so that multiple tasks can be completed at one time in the same area, reducing transportation time and frequent switching of operations. Assume the attributes of task A and task B in the task cluster are as follows: for task A, the location is (3, 4) and the priority is high; for task B, the location is (5, 6) and the priority is medium. The calculation process is as follows: ; According to this distance metric, the system will further incorporate task A and task B into different task clusters (if their job attributes and priorities match). After calculating the task distance metric, the system will continue with path planning, sorting and scheduling tasks based on the distance metric between tasks and job priorities to ensure that tasks can be executed on the most efficient path, optimizing resource allocation and reducing unnecessary time waste. In this way, the system can ensure that jobs are executed in priority order while minimizing resource waste and execution time. Obtain the path parameters of the task cluster.

[0036] The above is only a preferred embodiment of the present invention, and it is not intended to limit the present invention in other forms. Any person skilled in the art may use the disclosed technical content to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution content of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. An intelligent dispatching system for power grid operation and maintenance based on big data, characterized in that, The system includes: The data acquisition module obtains power grid operation data, extracts the current amplitude, voltage volatility, and active power change amount of multiple power grid devices, analyzes the volatility of each type of data in real time, calculates the variation amplitude of the operation data, adjusts the data sampling frequency, and obtains a power parameter data set. The anomaly identification module calls the power parameter data set, extracts the voltage monitoring data sequences of multiple power grid devices, calculates the voltage change slope, and calculates the slope difference of the voltage curve in a continuous time period, detects abnormal voltage data and abnormal power equipment, identifies abnormal time nodes, and obtains abnormal data location records. The status classification module, based on the abnormal data location records and the power parameter data set, detects the voltage phase offset value and the temperature gradient rise rate of the power equipment in real time, analyzes the equipment status, and obtains the equipment status classification level. The trend prediction module, based on the equipment status classification level, extracts the leakage current and operating voltage of the equipment in real time, calculates the insulation degradation factor of the equipment, combines the temperature gradient rise rate of the equipment, evaluates the aging status of the equipment, and calculates the remaining life cycle of the equipment, and obtains the maintenance trigger time window.

2. The intelligent dispatching system for power grid operation and maintenance based on big data according to claim 1, wherein The power parameter data set includes an equipment voltage waveform data set, leakage current detection data, and an equipment operating temperature data set. The abnormal data location record is specifically an abnormal time node, abnormal voltage data, and abnormal power equipment. The equipment status classification level is specifically an equipment health status level, voltage phase offset value, and temperature gradient rise rate. The maintenance trigger time window is specifically the remaining life cycle of the equipment, insulation degradation factor, and maintenance trigger time.

3. The intelligent dispatching system for power grid operation and maintenance based on big data according to claim 1, wherein, The data acquisition module includes: The power parameter extraction sub-module obtains power grid operation data, and by extracting the current amplitude, voltage volatility, and active power change amount data of multiple power grid devices, combines the equipment number and time stamp to obtain equipment operation characteristic data. The load variation analysis sub-module calls the equipment operation characteristic data and uses the formula: ； Calculates the data fluctuation determination value within each sliding window period to obtain the variation amplitude parameter. in, is the total number of samples in the sampling window, is the data point index, For the The value of the current amplitude data point, is the average value of the current amplitude, For the The value of the active power change data point, is the average value of active power change, For the The value of the voltage fluctuation rate data point, is the average value of voltage fluctuation rate, is the data fluctuation determination value; The sampling frequency control sub-module calls the variation amplitude parameter, adjusts the data acquisition frequency in real time according to the variation amplitude of the equipment operation data, and obtains the voltage waveforms, leakage currents, and operating temperature data of multiple operating devices, and constructs a power parameter data set.

4. The intelligent dispatching system for power grid operation and maintenance based on big data according to claim 3, wherein, The anomaly identification module includes: The voltage data extraction sub-module calls the power parameter data set, extracts the voltage values of each power equipment in the power grid at multiple time points, and combines the equipment number and time stamp to obtain the voltage monitoring data sequence. The voltage slope calculation sub-module calls the voltage monitoring data sequence, calculates the voltage change slope of the equipment in each time period according to the voltage difference between adjacent moments, and calculates the slope difference of the continuous time period to obtain the voltage slope difference. The abnormal node identification sub-module calls the voltage slope difference, detects abnormal voltage data in real time, combines the equipment number and time stamp, marks abnormal power events, and obtains abnormal data location records.

5. The intelligent dispatching system for power grid operation and maintenance based on big data according to claim 4, wherein The status classification module includes: The phase shift detection sub-module calls the abnormal data location record and the power parameter data set, extracts the absolute value of the phase difference between the voltage waveform of each cycle and the standard sine wave according to the voltage waveform sequence, calculates the moving window mean of the phase differences of consecutive cycles, calculates the moving average of the absolute values of the cycle phase differences, combines the recognized operating temperature data, calculates the temperature gradient rise rate, and obtains the set of state input parameters; The state judgment calculation sub-module calls the set of state input parameters, calculates the temperature rise increment and voltage fluctuation intensity of adjacent cycles, and combines the real-time active power of the device, using the formula: ； Calculate the comprehensive state determination value; Among them, represents the state comprehensive judgment value, represents the voltage phase offset value, represents the temperature gradient value, represents the real-time active power of the device, represents the temperature rise increment in adjacent cycles, represents the voltage fluctuation intensity in the current cycle; The risk level classification sub-module, based on the comprehensive state determination value, combines the preset state scoring interval, identifies the device state and matches the device state label in real time, and establishes the device state classification level.

6. The intelligent dispatching system for power grid operation and maintenance based on big data according to claim 5, characterized in that The trend prediction module includes: The insulation degradation extraction sub-module calls the device state classification level, calls the leakage current and operating voltage data of the power device, obtains the average leakage current and the average voltage amplitude, and obtains the insulation degradation factor value by analyzing the change rate of the leakage current and the change trend of the voltage amplitude, in combination with the voltage deviation direction; The aging state evaluation sub-module, based on the insulation degradation factor value, extracts the periodic temperature gradient change, obtains the temperature increase amount and time span of multiple cycles, combines the device voltage fluctuation parameters, calculates the temperature rise speed and the voltage deviation ratio fluctuation level, using the formula: ； Calculate the device aging trend offset amount, and average the fluctuation amplitude of consecutive cycles to obtain the comprehensive aging state index; Among them, represents the comprehensive index of the aging state, represents the leakage current of the th cycle, represents the temperature gradient, represents the deviation value of the operating voltage, represents the change value of the temperature rise, represents the time span of the cycle, represents the number of monitoring cycles; The life cycle calculation sub-module, according to the comprehensive aging state index, retrieves the standard life benchmark parameters of each device, combines the current time node to calculate the remaining life cycle of the device, and obtains the maintenance trigger time window.

7. The intelligent dispatching system for power grid operation and maintenance based on big data according to claim 1, characterized in that The system further includes: The path planning module calls the abnormal data location record and the maintenance trigger time window, constructs an operation and maintenance work order according to the type, location information, and required tool type of the abnormal device, calculates the similarity of the operation attributes of multiple work orders, combines the geographical coordinates of the device, and plans the task cluster path to obtain the grid operation and maintenance scheduling parameters; The grid operation and maintenance scheduling parameters are specifically the operation and maintenance work order construction record, the task cluster path, and the operation attribute similarity.

8. The intelligent dispatching system for power grid operation and maintenance based on big data according to claim 7, characterized in that The path planning module includes: The operation and maintenance work order construction sub-module constructs an operation and maintenance work order according to the abnormal data location record and the maintenance trigger time window, according to the type, location information, and required tool type of the abnormal device, and obtains the operation and maintenance work order list; The operation attribute similarity calculation sub-module, based on the operation and maintenance work order list, extracts the operation attributes of each work order, including tool requirements, voltage level, and operation risk level, using the formula: ； Calculate the similarity of the operation attributes between multiple work orders; Among them, represents the work order and the similarity value of the work order ; represents the value of the work order on the th job attribute, represents the value of the work order on the th job attribute, is the mean value of the work order on all job attributes, is the mean value of the work order on all job attributes, is the total number of job attributes, is the index of the job attribute; The task cluster path planning sub-module, based on the operation attribute similarity, combines the geographical coordinates of the device, the priority of the task, and the resource requirements, plans the task cluster path, and obtains the grid operation and maintenance scheduling path parameters.

Images