Automatic on-duty inspection method for master station dispatching system OCS

Abstract

The invention discloses an automatic on-duty inspection method for a master station dispatching system OCS, and the method comprises the steps: carrying out the clustering analysis of feature vectors according to a preset clustering model, and obtaining a clustering result of equipment; dynamically adjusting a data sampling period according to the clustering result, and shortening or prolonging the sampling period according to the equipment state change rate; establishing a delay prediction model according to historical data of equipment communication delay, and if the predicted delay exceeds a preset threshold value, starting a standby communication channel; aiming at the abnormally fluctuating equipment state data, generating a control instruction execution sequence by adopting a priority scheduling algorithm; generating a control instruction through a real-time control decision engine and issuing the control instruction to the target equipment; adjusting a control strategy according to an execution result of the control instruction; and optimizing a data acquisition and control instruction issuing strategy through a detection result of network fluctuation and hardware performance.

Description

Technical Field

[0001] The invention relates to the field of electric power inspection, in particular to an automatic on-duty inspection method for a master station dispatching system (OCS). Background Art

[0002] In the automated duty inspection of the master station dispatching system OCS, the realization of real-time control capabilities faces a core technical problem: how to find a balance between high-concurrency data collection and low-latency control command issuance. The system needs to process status data from thousands of devices at the same time. These data are updated at a frequency of milliseconds and the data volume is huge. However, the issuance of control instructions requires extremely high timeliness. Analysis, decision-making and execution must be completed in a very short time after the device status changes.

[0003] This leads to a contradiction: in order to ensure the integrity of data collection, a longer data sampling cycle needs to be set; but a longer sampling cycle will delay the issuance of control instructions, affecting the real-time performance of the system. At the same time, the communication delays of different devices vary. Some devices may delay or lose status data upload due to network fluctuations or hardware performance limitations. In this case, the system must ensure both the timeliness of control instructions and the accuracy of instructions to avoid making wrong decisions due to incomplete or delayed data. In addition, when multiple devices are in abnormal conditions at the same time, the system also needs to prioritize the control instructions of which devices, which involves the priority scheduling of control instructions.

[0004] How to ensure the priority control of key equipment while not affecting the normal inspection of other equipment has become another technical difficulty. The solution to these problems is directly related to the stability and reliability of the system, and requires in-depth analysis of the technical details of each link and optimization of the system architecture and algorithm. Summary of the invention

[0005] The purpose of the present invention is to solve the above problems and provide a master station dispatching system OCS automated duty inspection method, which can realize intelligent automatic inspection of the master station dispatching system and improve operation efficiency and reliability.

[0006] The technical solution adopted by the present invention to solve the technical problem is:

[0007] An automatic on-duty inspection method for a master station dispatching system OCS comprises the following steps:

[0008] a. Obtaining device status data in the master station dispatching system, grouping and extracting features from the device status data, and obtaining feature vectors corresponding to multiple devices;

[0009] b. performing cluster analysis on the feature vector according to a preset clustering model to obtain a clustering result of the device;

[0010] c. dynamically adjusting the data sampling period according to the clustering result, and shortening or extending the sampling period according to the rate of change of the device state;

[0011] d. Perform time series analysis on the equipment status data after dynamic sampling to determine the equipment status change trend. If abnormal fluctuations are found, the control instruction generation module is triggered;

[0012] e. Establish a delay prediction model based on the historical data of device communication delay, and activate the backup communication channel if the predicted delay exceeds the preset threshold;

[0013] f. For abnormally fluctuating equipment status data, a priority scheduling algorithm is used to generate the execution order of control instructions;

[0014] g Generate control instructions through real-time control decision engine and send them to target devices;

[0015] h. Adjust the control strategy according to the execution results of the control instructions;

[0016] iOptimize data collection and control instruction delivery strategies through network fluctuation and hardware performance detection results.

[0017] Furthermore, step a performs time series analysis on the dynamically sampled device status data, including the following steps:

[0018] Dynamically sample and obtain the equipment status data in the master station dispatching system OCS, input the data into the time series analysis module, and use the ARIMA algorithm to model and analyze the data change trend;

[0019] If the analysis result exceeds the preset threshold, it is judged as an abnormal fluctuation, triggering the control instruction generation module to generate corresponding control instructions;

[0020] The control instructions are transmitted to the main station dispatching system, which, combined with the automated duty inspection function of the OCS system, performs equipment status monitoring and response operations;

[0021] According to the monitoring results, the equipment status data is updated and fed back to the dynamic sampling module to form a closed-loop control;

[0022] If the device status data remains stable, maintain the current monitoring strategy. If abnormal fluctuations are detected again, repeat the above process.

[0023] Furthermore, step e mainly includes:

[0024] Acquire historical delay data of the communication device, and construct a time series-based delay prediction model according to the historical delay data;

[0025] Setting a delay threshold in the delay prediction model;

[0026] Using the delay prediction model to predict the future delay of the communication device to obtain a delay prediction result;

[0027] Determine whether the delay prediction result exceeds the delay threshold, and if so, trigger a channel switching mechanism;

[0028] According to the channel switching mechanism, starting the backup communication channel while keeping the data integrity verification mechanism running continuously;

[0029] Integrate the delay prediction results with the inspection process of the master station dispatching system to generate an intelligent operation and maintenance plan;

[0030] Acquire real-time status monitoring data of the communication device, and update the parameters of the delay prediction model according to the real-time status monitoring data to improve the delay prediction accuracy;

[0031] Establishing an automated threshold adjustment mechanism in the operation and maintenance system to dynamically adjust the delay threshold according to communication requirements in different scenarios;

[0032] The delay prediction model, channel switching mechanism and data integrity protection mechanism are integrated into the master station scheduling system to form a closed-loop control.

[0033] Furthermore, in step f, a priority scheduling algorithm is used to generate the execution order of control instructions, including:

[0034] Classify the abnormally fluctuating equipment status data to obtain multiple abnormal categories;

[0035] For each abnormality category, a preset scoring model is used to calculate the corresponding importance score of the device, and the abnormality severity level of each device is obtained. The preset comprehensive evaluation rules are used to determine the comprehensive evaluation result of the device based on the importance score and the abnormality severity level. Based on the comprehensive evaluation result, a priority scheduling algorithm is used to generate the initial execution order of the control instructions.

[0036] Further, step g mainly includes:

[0037] Obtain equipment status data from the master station dispatching system OCS and determine preset control conditions;

[0038] Determine whether the device state satisfies the trigger condition, and if so, use a real-time control decision engine to generate a control instruction;

[0039] According to the operating status of the target device, the control instruction is sent to the corresponding device;

[0040] Monitor the device status changes in real time through the automated duty inspection module to determine whether the device status meets the preset conditions;

[0041] If the preset condition is met, dynamically updating the control instruction;

[0042] The updated control instructions are executed by the master station scheduling system OCS.

[0043] The beneficial effects of the present invention are:

[0044] The present invention obtains the real-time status data of thousands of devices through a distributed data acquisition framework and groups them by device type and region. For grouped data, the present invention adopts a dynamic sampling cycle strategy to adjust the sampling frequency according to the rate of change of the device status. The ARIMA time series analysis algorithm is used to judge the trend of device status changes, and the control instruction generation is triggered when abnormal fluctuations are found. The present invention also establishes a communication delay prediction model to enable a backup channel when the predicted delay exceeds the threshold. Through the priority scheduling algorithm and the real-time control decision engine, the control instructions are generated and executed according to the importance of the equipment and the severity of the abnormality. If the execution result does not meet expectations, the present invention will re-analyze the data and adjust the strategy. In addition, the present invention also uses a machine learning algorithm to optimize the data acquisition and instruction issuance strategy to improve system stability. This method can realize the intelligent automatic inspection of the master station scheduling system and improve operation efficiency and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0046] Figure 1 This is a flow chart of the automatic on-duty inspection of the master station dispatching system OCS of the present invention.

[0047] In the figure: DETAILED DESCRIPTION

[0048] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0049] like Figure 1 As shown, a master station dispatching system OCS automated duty inspection method includes the following steps:

[0050] a. Obtaining device status data in the master station dispatching system, grouping and extracting features from the device status data, and obtaining feature vectors corresponding to multiple devices;

[0051] b. performing cluster analysis on the feature vector according to a preset clustering model to obtain a clustering result of the device;

[0052] c. dynamically adjusting the data sampling period according to the clustering result, and shortening or extending the sampling period according to the rate of change of the device state;

[0053] d. Perform time series analysis on the equipment status data after dynamic sampling to determine the equipment status change trend. If abnormal fluctuations are found, the control instruction generation module is triggered;

[0054] e. Establish a delay prediction model based on the historical data of device communication delay, and activate the backup communication channel if the predicted delay exceeds the preset threshold;

[0055] f. For abnormally fluctuating equipment status data, a priority scheduling algorithm is used to generate the execution order of control instructions;

[0056] g Generate control instructions through real-time control decision engine and send them to target devices;

[0057] h. Adjust the control strategy according to the execution results of the control instructions;

[0058] iOptimize data collection and control instruction delivery strategies through network fluctuation and hardware performance detection results.

[0059] Step a mainly includes: obtaining the real-time status data of the device status data in the master station scheduling system OCS, judging whether the data delay of each device exceeds a preset threshold for the real-time status data, and if it exceeds the preset threshold, triggering the data retransmission mechanism to re-acquire the real-time status data of the device; performing feature extraction on the acquired real-time status data of multiple devices to obtain a feature vector corresponding to each device; and performing cluster analysis on the feature vector corresponding to each device according to a pre-established clustering model to obtain a clustering result for each device.

[0060] Step c mainly includes: obtaining the device status data group data in the master station scheduling system OCS, and calculating the device status change rate for each group. If the device status change rate is higher than the preset threshold, shorten the sampling period; if the device status change rate is lower than the preset threshold, extend the sampling period. According to the adjusted sampling period, re-collect the device status data. Calculate the new device status change rate for the re-collected device status data. If the new device status change rate is not within the preset threshold range, adjust the sampling period again. According to the finally determined sampling period, continuously monitor the device status change rate to maintain a dynamic adjustment mechanism.

[0061] Step d mainly includes: dynamically sampling and acquiring the equipment status data in the master station dispatching system OCS, inputting the data into the time series analysis module, and using the ARIMA algorithm to model and analyze the data change trend. If the analysis result exceeds the preset threshold, it is judged as an abnormal fluctuation, triggering the control instruction generation module to generate the corresponding control instruction. The control instruction is transmitted to the master station dispatching system, and combined with the automated duty inspection function of the OCS system, equipment status monitoring and response operations are performed. According to the monitoring results, the equipment status data is updated and fed back to the dynamic sampling module to form a closed-loop control. If the equipment status data remains stable, maintain the current monitoring strategy. If abnormal fluctuations are detected again, repeat the above process.

[0062] The dynamic sampling system continuously monitors the operating status of the equipment and adjusts the data collection frequency in real time. Taking transformer oil temperature monitoring as an example, conventional fixed sampling may collect data once an hour, while dynamic sampling is flexibly adjusted according to the oil temperature change trend. When the oil temperature fluctuates slightly within the normal range, the sampling interval can be extended to two hours. When the oil temperature is detected to rise rapidly, the system automatically shortens the sampling interval to ten minutes to ensure that abnormal changes are captured in time.

[0063] The time series analysis module uses the ARIMA model to model historical data. Taking the temperature monitoring of the switch cabinet in the power distribution room as an example, a prediction model is established by analyzing the temperature change data of the past three months. The model can predict the temperature change trend within the next four hours. When the predicted value exceeds the preset temperature threshold range, the system determines it as an abnormal fluctuation.

[0064] The control instruction generation module formulates corresponding measures according to the abnormal type. For example, in the load monitoring of transmission lines, when it is detected that the load of a certain line continues to rise and is predicted to exceed the rated capacity, the system automatically generates a load transfer instruction. The instruction may include specific operation steps to transfer part of the load to the backup line.

[0065] The main station dispatching system works in conjunction with the OCS system to achieve automated inspections. Taking the substation as an example, when the system receives an abnormal warning of equipment, the OCS system automatically adjusts the angle of the surveillance camera to focus on monitoring the abnormal equipment. At the same time, the system automatically notifies the on-duty personnel and displays the real-time status data of the abnormal equipment on the monitoring screen.

[0066] Closed-loop control optimizes monitoring strategies through continuous data feedback. For example, in cable temperature monitoring, if the temperature continues to rise after the system issues a load reduction instruction, the sampling period will be further shortened to five minutes and the emergency plan will be activated. On the contrary, if the load reduction measures are effective and the temperature gradually drops to the normal range, the system will gradually resume the normal sampling period.

[0067] There are correlations between multiple monitoring objects, and the system needs to conduct a comprehensive analysis. For example, if the current of a bus suddenly increases, the system not only monitors the temperature change of the bus itself, but also pays attention to the status changes of related equipment such as circuit breakers and cables. By establishing a correlation model between devices, the system can more accurately determine the propagation trend of abnormal conditions, thereby formulating more effective control strategies.

[0068] Step e mainly includes: obtaining historical delay data of the communication device, building a delay prediction model based on time series according to the historical delay data; setting a delay threshold in the delay prediction model; using the delay prediction model to predict the future delay of the communication device to obtain a delay prediction result;

[0069]

[0070] P(t) represents the delay prediction value at time t, α represents the attenuation coefficient, w_i represents the weight coefficient, β represents the time attenuation parameter, and t_i represents the historical time point.

[0071] Determine whether the delay prediction result exceeds the delay threshold, and if so, trigger the channel switching mechanism; according to the channel switching mechanism, start the backup communication channel, and keep the data integrity verification mechanism running continuously; integrate the delay prediction result with the inspection process of the master station scheduling system to generate an intelligent operation and maintenance plan; obtain real-time status monitoring data of communication equipment, and update the parameters of the delay prediction model according to the real-time status monitoring data to improve the delay prediction accuracy; establish an automated threshold adjustment mechanism in the operation and maintenance system, and dynamically adjust the delay threshold according to the communication needs in different scenarios; integrate the delay prediction model, channel switching mechanism and data integrity protection mechanism into the master station scheduling system to form a closed-loop control.

[0072] The historical delay data of communication equipment usually contains multiple dimensions, such as end-to-end delay of data transmission, network transmission delay, processing delay, etc. The historical delay data collection of a substation shows that under normal working conditions, the end-to-end delay is maintained at an average of about 50 milliseconds, but it fluctuates when the equipment is heavily loaded.

[0073] The time series prediction model built based on this data can capture the regular characteristics of delay changes. The threshold setting of the delay prediction model needs to take into account actual business needs. For example, in remote control operation scenarios, the delay threshold is usually set to 100 milliseconds. Exceeding this threshold may affect the timeliness of instructions.

[0074] The prediction model can predict possible delay surges 10 to 15 minutes in advance by analyzing the trend of historical data. In a certain area's dispatch automation system, when it is predicted that the communication delay is about to exceed the threshold, the system will automatically start the backup fiber channel. At the same time, the data integrity verification mechanism uses checksum technology to ensure that data will not be lost or erroneous during the channel switching process. This mechanism has increased the system availability from the original 99.9% to 99.99% in actual operation.

[0075] In the process of generating intelligent operation and maintenance solutions, the system combines the predicted delay trend with the inspection plan of the master station scheduling system. For example, when it detects that the delay of a certain fiber channel may increase, the system will automatically adjust the inspection frequency and intensify the monitoring of the channel. Through real-time status monitoring, the system continuously collects equipment operation data to update the parameters of the prediction model.

[0076] In different scenarios, the tolerance for communication delay varies. For example, the transmission of protection information requires a delay of no more than 10 milliseconds, while ordinary monitoring data can accept a delay of 50 milliseconds. The automated threshold adjustment mechanism dynamically sets the threshold according to the business type to ensure that the communication quality meets differentiated needs.

[0077] In the master station dispatching system, delay prediction, channel switching and data integrity protection form an organic whole. For example, when the communication quality of a substation deteriorates, the system first detects the abnormal trend through the prediction model, then automatically switches to the backup channel, and starts the data verification mechanism to ensure that key information is not affected. This closed-loop control enables the system to proactively prevent and respond quickly when facing fluctuations in communication quality, significantly improving the reliability of power grid dispatching.

[0078] Step f mainly includes: obtaining the equipment status data in the master station scheduling system OCS; judging whether there is abnormal fluctuation for the equipment status data; if there is abnormal fluctuation, classifying the equipment status data of the abnormal fluctuation to obtain multiple abnormal categories; for each abnormal category, using a preset scoring model to calculate the importance score corresponding to each device in the abnormal category; obtaining the abnormal severity level of each device; according to the importance score and the abnormal severity level, using a preset comprehensive evaluation rule to determine the comprehensive evaluation result of each device; according to the comprehensive evaluation result, using a priority scheduling algorithm to generate an initial execution order of control instructions; inputting the initial execution order into the automated duty inspection module, verifying the initial execution order through the automated duty inspection module, and obtaining an adjusted control instruction execution order; feeding back the adjusted control instruction execution order to the master station scheduling system OCS to generate a final control instruction execution plan.

[0079] The main station dispatching system OCS is the nerve center of the power system operation, monitoring and managing the operating status of the entire power grid in real time.

[0080] After obtaining the equipment status data, the system needs to quickly determine whether there are abnormal fluctuations. For example, the oil temperature of the main transformer of a substation suddenly rises by 10 degrees, which is an abnormal fluctuation. Classifying the abnormally fluctuating equipment status data is to deal with the problem more targeted. For example, the abnormality can be divided into categories such as voltage abnormality, frequency abnormality, and temperature abnormality. Each category can be further subdivided, such as voltage abnormality can be divided into overvoltage, undervoltage, etc.

[0081] The establishment of the importance scoring model can take into account factors such as the position of the equipment in the system and the scope of influence. For example, the circuit breaker of the trunk line is more important than the distribution transformer. The score can be quantitatively calculated from 1 to 100, and the higher the score, the more important it is. The severity level of the abnormality can usually be divided into several levels such as minor, general, severe, and critical. If the load of a transmission line reaches 95% of the rated capacity, it can be judged as a severe level. The comprehensive evaluation rule needs to weigh the importance score and the severity of the abnormality. For example, the weighted average method can be used to give a 60% weight to the importance score and a 40% weight to the severity of the abnormality to calculate the comprehensive score.

[0082] The priority scheduling algorithm generates the initial execution order of control instructions based on the comprehensive evaluation results. For example, for multiple exceptions that need to be handled, they can be sorted from high to low according to the comprehensive scores, and the exceptions with high scores are handled first.

[0083] The automated duty inspection module verifies the initial execution order and may take some human factors into consideration. For example, if a line is less abnormal but is located in a densely populated area, it may be given a higher priority. After the final control instruction execution plan is formed, the system will automatically issue control instructions. For example, for an overheated transformer, an instruction to reduce the load may be issued; for a line with abnormal voltage, the voltage regulating equipment may be adjusted.

[0084] Real-time monitoring of equipment status changes is to adjust the execution plan in a timely manner. For example, if the load of a certain line quickly returns to normal after executing a load reduction instruction, the system can dynamically adjust the plan and transfer resources to other anomalies that need to be handled.

[0085] The establishment of this process will help improve the intelligence level of power grid operation, achieve rapid and accurate fault handling, and improve the reliability and stability of the power grid. Through the data-driven decision-making process, human judgment errors are reduced, and scientific decision-making support is provided for operation and maintenance personnel.

[0086] Step g mainly includes: obtaining the equipment status data in the master station dispatching system OCS and determining the preset control conditions; judging whether the equipment status meets the trigger conditions, and if so, using the real-time control decision engine to generate control instructions; issuing the control instructions to the corresponding equipment according to the operating status of the target equipment; monitoring the equipment status changes in real time through the automated duty inspection module to judge whether the equipment status meets the preset conditions; if so, dynamically updating the control instructions; and executing the updated control instructions through the master station dispatching system OCS.

[0087] The equipment status data in the master station dispatching system includes key information such as equipment operating parameters, alarm information, historical trends, etc. Taking the substation as an example, the transformer's oil temperature, load rate, voltage and other operating parameters are collected in real time into the system.

[0088] The preset control conditions may include specific indicators such as temperature threshold, load upper limit, voltage deviation range, etc. These indicators are set based on the equipment's safe operating procedures and historical experience values.

[0089] The judgment mechanism of the triggering condition can adopt a multi-dimensional evaluation method. For example, if the transformer oil temperature exceeds 75 degrees and lasts for more than 30 minutes, and the load rate is higher than 90%, an overload warning will be triggered.

[0090] The real-time control decision engine will generate load reduction instructions based on the current status. Possible control measures include cutting off some non-important loads, starting backup equipment, etc.

[0091] The evaluation of the operating status of the target equipment involves multiple dimensions. Taking the distribution line as an example, it is necessary to consider indicators such as line load rate, power factor, and voltage quality. When it is detected that the load rate of a line exceeds 85%, the system will analyze the status of the upstream and downstream equipment of the line, and generate a reasonable load transfer plan based on the power flow distribution.

[0092] Step h mainly includes: obtaining equipment status data, and determining the degree of matching between the status data and the preset target; if the matching degree is lower than the preset threshold, starting the status data analysis process, and determining the adjustment direction of the control strategy according to the analysis results; using the strategy optimization algorithm to generate the adjusted control strategy, and applying the adjusted control strategy to the main station dispatching system; monitoring the execution result of the adjusted control strategy through automated duty inspection; if the execution result does not meet expectations, returning to the status data analysis process.

[0093] In the electric power dispatching system, equipment status data acquisition covers key indicators such as voltage level, active power, reactive power, etc.

[0094] The data matching degree usually adopts the similarity calculation method to compare the measured value with the target value. For example, if the transformer load rate target value is set at 70%, and the actual operating load rate is 85%, the matching degree is lower than the preset threshold of 90%, and the analysis process needs to be started.

[0095] Status data analysis focuses on equipment operating characteristics and external environmental impacts. Taking the distribution transformer as an example, analyze its load characteristics, temperature changes, loss levels and other operating parameters. According to the daily load curve, if the load rate of a transformer exceeds 85% during the working day, and the ambient temperature is high, the heat dissipation of the equipment is limited, then the control strategy direction needs to be adjusted.

[0096] The strategy optimization algorithm comprehensively considers equipment capacity, operating efficiency and system stability. Taking the substation bus voltage control as an example, through the reactive power optimization algorithm, the operating voltage can be appropriately reduced while ensuring the voltage qualification rate, which can reduce equipment losses.

[0097] The optimized control strategy includes specific parameters such as voltage regulation target value, regulation step, response time, etc. The automated duty inspection system monitors the execution effect of the strategy in real time. Taking the operation of switchgear as an example, the system monitors key indicators such as state change signal, operation time, and current transformer data. If an abnormal operation is found or the execution result deviates from expectations, such as exceeding the opening and closing time limit or unbalanced load transfer, the system will trigger the feedback mechanism.

[0098] The practical application of a substation shows that this method can effectively improve the level of equipment management. Originally, the operating personnel needed to inspect the equipment operating status every two hours. After adopting this method, the system can automatically complete status monitoring and strategy optimization, greatly improving the operating efficiency. When the equipment status is abnormal, the system can quickly identify and take corresponding measures to avoid potential risks such as equipment overload and heating. In the sag control of transmission line conductors, the system collects meteorological data and load data to establish a model of the relationship between conductor temperature and sag. When there is a large load or high temperature weather, the system automatically adjusts the line power distribution to ensure that the conductor sag meets safety requirements. Practice has proved that this method can effectively prevent line tripping accidents caused by excessive sag.

[0099] Step i mainly includes: using machine learning algorithms to optimize data collection and control instruction delivery strategies based on the detection results of network fluctuations or hardware performance limitations to improve system stability

[0100] The detection results of network fluctuations and hardware performance are obtained, and the detection results are judged by using a preset threshold to obtain abnormal data features; for the abnormal data features, data acquisition optimization rules are constructed through a decision tree model; according to the data acquisition optimization rules, a control instruction issuance strategy is generated, and the control instruction issuance strategy is screened by using a support vector machine model to obtain a screened control instruction issuance strategy; the screened control instruction issuance strategy is applied to the main dispatching system to obtain the execution result of the automated duty inspection; for the execution result, the change trend of the system stable state is analyzed through a random forest model; according to the stable state change trend, the parameter configuration of the data acquisition optimization rules and the control instruction issuance strategy is adjusted to obtain the optimized data acquisition rules and control instruction issuance strategy; the optimized data acquisition rules and control instruction issuance strategy are reapplied to the main dispatching system, the execution result of the automated duty inspection is updated, and the system stability is improved.

[0101] In the power IoT master station dispatching system, abnormal data characteristics of network fluctuations and hardware performance can be identified by monitoring key indicators such as equipment communication delay and packet loss rate. For example, when the communication delay exceeds the preset threshold or the packet loss rate is greater than 5%, it is determined to be an abnormal data characteristic.

[0102] When the decision tree model constructs data collection optimization rules, the collection method can be set according to the various operating conditions and load parameters of the power equipment. For example, for high-voltage line fault detection devices, when the load rate is at peak times, the data collection frequency needs to be increased to once per minute; when it is at low times, it can be reduced to once every five minutes, thereby balancing data real-time performance and system resource consumption.

[0103] In the screening of control command issuance strategies, the support vector machine model can evaluate factors such as command type and execution timing. Specifically, it can intelligently match the timing of command issuance according to the equipment type and operating conditions. For example, for the distribution transformer temperature control system, when a temperature rise trend is detected, the priority of issuing control commands can be intelligently adjusted to ensure that cooling measures are initiated in a timely manner.

[0104] When analyzing the trend of system stability through the random forest model, multiple dimensional indicators such as grid frequency fluctuation and voltage stability can be comprehensively considered. For example, when it is found that the voltage fluctuation amplitude in a certain area is gradually increasing, the model will predict the potential stability risk and give early warning information.

[0105] In terms of parameter configuration optimization, the system will dynamically adjust data collection rules and control strategies based on the analysis results of historical operation data. For example, for distribution automation terminals, when the data quality decreases during a period of time, the system will automatically improve the sampling accuracy and adjust the data filtering threshold accordingly. At the same time, the execution order of control instructions will also be optimized according to the response characteristics of the equipment to ensure that the control instructions of key equipment are executed first.

[0106] In actual application scenarios, this optimization mechanism can effectively deal with various emergencies in power grid operation. For example, when a short circuit fault is detected in a feeder, the system can quickly adjust the data collection frequency of related equipment and optimize the action strategy of the protection device to improve the efficiency of fault handling. In this way, the stability of system operation is guaranteed and the accuracy of dispatch control is improved.

[0107] In the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "left", "right", "up", "down", etc. are based on the directions or positional relationships shown in the accompanying drawings, and are 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 direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention.

[0108] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection, it can be a direct connection, it can be an indirect connection through an intermediate medium, and it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

Claims

1. A master station dispatching system OCS automated duty inspection method, characterized in that: The following steps are involved: a. Obtaining device status data in the master station dispatching system, grouping and extracting features from the device status data, and obtaining feature vectors corresponding to multiple devices; b. performing cluster analysis on the feature vector according to a preset clustering model to obtain a clustering result of the device; c. dynamically adjusting the data sampling period according to the clustering result, and shortening or extending the sampling period according to the rate of change of the device state; d. Perform time series analysis on the equipment status data after dynamic sampling to determine the equipment status change trend. If abnormal fluctuations are found, the control instruction generation module is triggered; e. Establish a delay prediction model based on the historical data of device communication delay, and activate the backup communication channel if the predicted delay exceeds the preset threshold; f. For abnormally fluctuating equipment status data, a priority scheduling algorithm is used to generate the execution order of control instructions; g Generate control instructions through real-time control decision engine and send them to target devices; h. Adjust the control strategy according to the execution results of the control instructions; iOptimize data collection and control instruction delivery strategies through network fluctuation and hardware performance detection results.

2. The method for automatic on-duty inspection of a master station dispatching system OCS according to claim 1, characterized in that: Step d performs time series analysis on the dynamically sampled device status data, including the following steps: Dynamically sample and obtain the equipment status data in the master station dispatching system OCS, input the data into the time series analysis module, and use the ARIMA algorithm to model and analyze the data change trend; If the analysis result exceeds the preset threshold, it is judged as an abnormal fluctuation, triggering the control instruction generation module to generate corresponding control instructions; The control instructions are transmitted to the main station dispatching system, which, combined with the automated duty inspection function of the OCS system, performs equipment status monitoring and response operations; According to the monitoring results, the equipment status data is updated and fed back to the dynamic sampling module to form a closed-loop control; If the device status data remains stable, maintain the current monitoring strategy. If abnormal fluctuations are detected again, repeat the above process.

3. The method for automatic on-duty inspection of a master station dispatching system OCS according to claim 1, characterized in that: Step e mainly includes: Acquire historical delay data of the communication device, and construct a time series-based delay prediction model according to the historical delay data; Setting a delay threshold in the delay prediction model; Using the delay prediction model to predict the future delay of the communication device to obtain a delay prediction result; Determine whether the delay prediction result exceeds the delay threshold, and if so, trigger a channel switching mechanism; According to the channel switching mechanism, starting the backup communication channel while keeping the data integrity verification mechanism running continuously; Integrate the delay prediction results with the inspection process of the master station dispatching system to generate an intelligent operation and maintenance plan; Acquire real-time status monitoring data of the communication device, and update the parameters of the delay prediction model according to the real-time status monitoring data to improve the delay prediction accuracy; Establishing an automated threshold adjustment mechanism in the operation and maintenance system to dynamically adjust the delay threshold according to communication requirements in different scenarios; The delay prediction model, channel switching mechanism and data integrity protection mechanism are integrated into the master station scheduling system to form a closed-loop control.

4. The method for automatic on-duty inspection of a master station dispatching system OCS according to claim 1, characterized in that: In step f, a priority scheduling algorithm is used to generate the execution order of control instructions, including: Classify the abnormally fluctuating equipment status data to obtain multiple abnormal categories; For each abnormality category, a preset scoring model is used to calculate the corresponding importance score of the device, and the abnormality severity level of each device is obtained. The preset comprehensive evaluation rules are used to determine the comprehensive evaluation result of the device based on the importance score and the abnormality severity level. Based on the comprehensive evaluation result, a priority scheduling algorithm is used to generate the initial execution order of the control instructions.

5. The method for automatic on-duty inspection of a master station dispatching system OCS according to claim 1, characterized in that: Step g mainly includes: Obtain equipment status data from the master station dispatching system OCS and determine preset control conditions; Determine whether the device state satisfies the trigger condition, and if so, use a real-time control decision engine to generate a control instruction; According to the operating status of the target device, the control instruction is sent to the corresponding device; Monitor the device status changes in real time through the automated duty inspection module to determine whether the device status meets the preset conditions; If the preset condition is met, dynamically updating the control instruction; The updated control instructions are executed by the master station scheduling system OCS.

Images