CN117909855B - Data monitoring and auxiliary governance method based on electric power model - Google Patents
Data monitoring and auxiliary governance method based on electric power model Download PDFInfo
- Publication number
- CN117909855B CN117909855B CN202410317710.0A CN202410317710A CN117909855B CN 117909855 B CN117909855 B CN 117909855B CN 202410317710 A CN202410317710 A CN 202410317710A CN 117909855 B CN117909855 B CN 117909855B
- Authority
- CN
- China
- Prior art keywords
- model
- power
- data
- unit
- adaptive
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 66
- 238000000034 method Methods 0.000 title claims abstract description 46
- 239000002245 particle Substances 0.000 claims abstract description 57
- 238000005457 optimization Methods 0.000 claims abstract description 53
- 238000004088 simulation Methods 0.000 claims abstract description 49
- 230000008859 change Effects 0.000 claims abstract description 17
- 238000005094 computer simulation Methods 0.000 claims abstract description 6
- 238000013480 data collection Methods 0.000 claims abstract description 4
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 238000004458 analytical method Methods 0.000 claims description 43
- 238000012545 processing Methods 0.000 claims description 42
- 230000003044 adaptive effect Effects 0.000 claims description 33
- 238000004364 calculation method Methods 0.000 claims description 27
- 238000012549 training Methods 0.000 claims description 27
- 238000004891 communication Methods 0.000 claims description 22
- 238000000605 extraction Methods 0.000 claims description 22
- 238000004422 calculation algorithm Methods 0.000 claims description 18
- 230000002159 abnormal effect Effects 0.000 claims description 17
- 238000011156 evaluation Methods 0.000 claims description 17
- 230000006870 function Effects 0.000 claims description 17
- 238000012795 verification Methods 0.000 claims description 17
- 238000012937 correction Methods 0.000 claims description 16
- 238000005070 sampling Methods 0.000 claims description 11
- 230000004927 fusion Effects 0.000 claims description 9
- 230000005540 biological transmission Effects 0.000 claims description 8
- 230000007246 mechanism Effects 0.000 claims description 8
- 230000005856 abnormality Effects 0.000 claims description 7
- 230000035945 sensitivity Effects 0.000 claims description 7
- 230000002457 bidirectional effect Effects 0.000 claims description 6
- 238000012706 support-vector machine Methods 0.000 claims description 6
- 230000003750 conditioning effect Effects 0.000 claims description 5
- 230000006872 improvement Effects 0.000 claims description 5
- 238000003908 quality control method Methods 0.000 claims description 5
- 238000012360 testing method Methods 0.000 claims description 5
- 238000011217 control strategy Methods 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 4
- 230000000739 chaotic effect Effects 0.000 claims description 3
- 238000002790 cross-validation Methods 0.000 claims description 3
- 238000007418 data mining Methods 0.000 claims description 3
- 238000009826 distribution Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 238000003860 storage Methods 0.000 claims description 3
- 230000002194 synthesizing effect Effects 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000013500 data storage Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 238000010223 real-time analysis Methods 0.000 description 2
- 206010033799 Paralysis Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000004146 energy storage Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000003631 expected effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 238000005067 remediation Methods 0.000 description 1
- 238000009420 retrofitting Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000007480 spreading Effects 0.000 description 1
- 238000003892 spreading Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a data monitoring and auxiliary governance method based on an electric power model, which relates to the technical field of data monitoring and solves the problems that the monitoring and auxiliary governance method of an electric power system in the prior art has low efficiency and lagged monitoring capability, and the adopted method is as follows: firstly, data collection and preprocessing are carried out, then a cloud server is used for constructing an electric power model, then the electric power system is monitored in real time based on the electric power model, then a simulation dynamic simulation method is adopted for carrying out simulation and prediction on the change trend of electric power parameters through the electric power model, the stability and reliability of the electric power system are evaluated according to preset rules, and an auxiliary treatment decision of the electric power system is carried out through a self-adaptive particle swarm optimization decision model.
Description
Technical Field
The invention relates to the technical field of data monitoring, in particular to a data monitoring and auxiliary governance method based on a power model.
Background
The power system is an important support for economic operation in modern society, and stable operation of the power system has important significance for guaranteeing power safety and daily production and life. With the ever-increasing scale of power systems and ever-increasing power loads, grid operation is facing increasingly complex challenges. Meanwhile, new energy sources, energy storage technologies and other emerging technologies are applied to the power system to bring new changes and challenges. In this case, how to ensure safe and stable operation of the power grid becomes a problem to be solved. The power system is a complex dynamic system involving numerous electrical devices, sensors and complex control systems. During the operation of the power system, numerous faults and problems, such as overload, equipment faults, power grid faults and the like, can occur, and the problems can lead to instability and even paralysis of the power system, thereby affecting the operation of the whole power grid. In order to ensure safe, stable and reliable operation of the power system, real-time monitoring, prediction and auxiliary management of the power system are required.
However, in the traditional power monitoring method, the problems of low data acquisition and processing efficiency, poor accuracy, limited information quantity and the like are solved, in the traditional power monitoring method, only certain parameters of a power system can be monitored, the running condition of the power system can not be comprehensively reflected, the traditional power system monitoring method generally only depends on discrete sampling data and experience judgment, the operation difficulty is high, and the accuracy and the instantaneity are limited. Existing power models typically rely solely on historical data and fail to effectively combine real-time data for model training and optimization. Conventional power system governance decisions typically rely on expert experience and subjective judgment, and the decision process is susceptible to interference and misleading.
Therefore, the invention discloses a data monitoring and auxiliary governance method based on a power model, which is used for realizing real-time monitoring and auxiliary governance of a power system by collecting, storing and processing a large amount of power parameters and equipment operation state data and constructing a complete power model, thereby improving the safety, stability and operation efficiency of the power system.
Disclosure of Invention
Aiming at the defects of the prior art, the invention discloses a data monitoring and auxiliary governance method based on a power model, which is used for realizing real-time monitoring and auxiliary governance of a power system by collecting, storing and processing a large amount of power parameters and equipment operation state data and constructing a complete power model, thereby improving the safety, stability and operation efficiency of the power system; the data are screened, standardized and quality controlled by a quality control strategy and a sensor signal conditioning circuit. Therefore, the data acquisition and processing efficiency can be greatly improved, and the data accuracy and the information quantity are improved; the combination of the distributed sensor network, the cloud server and the power model is adopted, so that the real-time monitoring of the power system is realized, and the monitoring accuracy and the real-time performance are improved; the model is trained and adjusted in real time through the increment updating unit and the self-adaptive adjusting module, so that the accuracy and the practicability of the model are improved; the self-adaptive particle swarm optimization decision model is adopted, and an optimal decision scheme can be obtained through real-time data monitoring and simulation prediction, so that the treatment efficiency and accuracy are improved; by constructing a complete electric power model, the model comprises a plurality of components such as a feature extraction unit, a data balance unit, a model training unit, a self-adaptive adjustment module, an increment updating unit, a processing analysis unit and the like, so that the running condition of the electric power system can be comprehensively reflected, and abnormal conditions can be monitored in real time; and the automation degree and the intelligent degree are high.
The invention adopts the following technical scheme:
A data monitoring and auxiliary governance method based on a power model comprises the following steps:
Step one, data collection and preprocessing, namely collecting power parameters and equipment operation state data in real time through a distributed sensor network and monitoring nodes, transmitting the collected data to a cloud end for storage and processing by adopting an Ethernet, wherein the power parameters and the equipment operation state data screen, normalize and control the quality of the collected data through a quality control strategy and a sensor signal conditioning circuit, and the power parameters and the equipment operation state data at least comprise operation voltage, current, output power, a switch state and a topological structure of a power system;
Constructing a power model, namely constructing the power model through a cloud server, wherein the cloud server constructs the power model based on historical data and real-time data, the power model comprises a feature extraction unit, a data balance unit, a model training unit, an adaptive adjustment module, an increment updating unit and a processing analysis unit, the output end of the feature extraction unit is connected with the input end of the data balance unit, the output end of the data balance unit is connected with the input end of the model training unit, the model training unit is in bidirectional connection with the adaptive adjustment module, the output end of the model training unit is connected with the input end of the processing analysis unit, and the output end of the increment updating unit is connected with the input ends of the model training unit and the processing analysis unit;
Step three, real-time data monitoring, namely, monitoring the power system in real time based on the power model, comparing and analyzing actually collected data with a simulation calculation result through a power abnormality iterative optimization model by the power model, and judging the abnormal condition of the current power parameter according to the comparison result;
Step four, simulation prediction, wherein the power model adopts a simulation dynamic simulation method to perform simulation prediction on the power parameter change trend, and evaluates the stability and reliability of the power system according to preset rules;
Step five, auxiliary governance decision making, namely carrying out auxiliary governance decision making of the electric power system through a self-adaptive particle swarm optimization decision making model, wherein the self-adaptive particle swarm optimization decision making model obtains an optimal decision scheme of the electric power system according to real-time data monitoring results and simulation prediction results;
and step six, model improvement and optimization, wherein the electric power model is optimized and improved through a real-time feedback verification mechanism.
As a further technical scheme of the invention, the feature extraction unit maps high-dimensional data to a low-dimensional space through linear discriminant analysis to realize extraction of power parameter features, the data balance unit performs balance processing on the extracted features through synthesizing minority oversampling, and performs training sample proportion division by adopting a method combining cross validation and random division, the model training unit performs training on the balanced features through a support vector machine, the adaptive adjustment module performs adaptive adjustment on parameters of the model according to real-time data, the increment update unit performs increment learning and updating on historical data and real-time data through a time window, and the processing analysis unit performs processing and analysis by adopting a data rule engine and a data mining algorithm.
As a further technical scheme of the invention, the power anomaly iterative optimization model analyzes anomaly nodes of the power system through power parameter time sequence sampling, the power anomaly iterative optimization model monitors the anomaly nodes in the power system in real time through phase space reconstruction and time sequence iteration, the phase space reconstruction extracts operation characteristics of the power system through mapping power parameters onto a phase space, and a phase space orbit intermediate vector distance calculation formula of the power parameters is as follows:
（1）
in the case of the formula (1), Phase space orbit intermediate vector spacing representing power parameters,/>Distribution cross term coefficient representing power parameter,/>Representing the power load intensity of the power system,/>And/>Representing equalization coefficients of the data samples, the time series iterations monitoring the power anomaly parameter time series quantities through power parameter time series scalar analysis and feature extraction, formulated as:
（2）
In the formula (2) of the present invention, Represents the nth iteration sequence quantity of the power parameter,/>Multi-element quantitative value function representing observed power anomaly parameter scalar,/>A multivariate magnitude function representing a time series scalar of the power parameter,Representing measurement errors,/>Representing the initial sampling time,/>And (3) representing a sampling interval, wherein n represents the iteration times, the power anomaly iteration optimization model obtains an information fusion result based on phase space reconstruction and time-frequency feature extraction, and a calculation formula is as follows:
（3）
In the formula (3) of the present invention, Representing the fusion result of phase space reconstruction and time-frequency characteristic extraction information,/>Representing the embedding dimension of a power anomaly load sequence in a time frequency,/>Time-frequency characteristics representing power parameters; comparing and analyzing the information fusion result with the simulation calculation result, wherein the calculation formula is as follows:
（4）
in the formula (4) of the present invention, Representing the actual collected data,/>Representing simulation calculation results,/>The difference degree between the actually collected data and the simulation calculation result is represented, when the actual error exceeds the error threshold, the abnormal node of the current power parameter is judged, and the error threshold calculation formula is as follows:
（5）
in the formula (5) of the present invention, Representing an error threshold value,/>Representing standard deviation of data,/>Representing a multiple of the error threshold for controlling the sensitivity of the error threshold.
According to the further technical scheme, the cloud database and the monitoring computer are used as simulation control units of the power system, the cloud database and the monitoring computer simulate dynamic operation of the power equipment according to real-time monitoring data and adaptively adjust a control strategy according to preset rules, the cloud database and the monitoring computer simulate power parameter change trend through the component object model COM and transmit simulation control instructions through a wireless communication bridge, and the component object model COM transmits and receives the simulation control instructions through a serial communication protocol.
As a further technical scheme of the invention, the adaptive particle swarm optimization decision model comprises an input layer, an adaptive parameter adjusting layer, a particle updating layer, an adaptive weight layer, a multi-neighborhood searching layer and an output layer, and the adaptive particle swarm optimization decision model comprises the following steps:
S1, initializing input, namely setting initial population size, maximum iteration times, inertia weight and neighborhood search parameters through the input layer, and inputting an initial particle swarm, an objective function to be optimized and a decision variable into the adaptive particle swarm optimization decision model;
s2, self-adaptive parameter adjustment, wherein the self-adaptive parameter adjustment layer is used for setting control parameters, objective functions and self-adaptive algorithms according to optimal solutions of particle swarms and historical optimal solutions, and dynamically adjusting the neighborhood searching range based on different neighborhood searching strategies;
S3, particle updating, namely calculating the fitness value of the particle and updating the speed and the position of the particle through the particle updating layer, wherein the particle updating layer calculates the fitness value of the particle according to the current position and the speed and updates the position and the speed of the particle;
S4, self-adaptive inertia weight adjustment, wherein the inertia weight is dynamically adjusted through the self-adaptive weight layer, the self-adaptive weight layer self-adaptively updates the inertia weight according to the current iteration times and the global history optimal fitness value, and the self-adaptive weight layer adjusts the change trend of the inertia weight according to the change condition of the objective function;
S5, multi-neighborhood searching, wherein a plurality of neighborhood searching strategies are formulated through the multi-neighborhood searching layer, and the plurality of neighborhood searching strategies comprise global neighborhood searching, local neighborhood searching and chaotic searching;
and S6, outputting an optimal decision scheme, and outputting an optimal solution and an optimal position of particles through the output layer when the maximum iteration number is reached or the termination condition is met, otherwise, repeating the S2.
As a further technical scheme of the invention, the real-time feedback verification mechanism comprises a model evaluation unit, a feedback analysis unit, a model correction unit and a cross verification unit, wherein the model evaluation unit evaluates the electric power model by adopting a mean square error, an average absolute error and an R-square coefficient, the feedback analysis unit carries out real-time analysis on an evaluation result through a real-time stream processing engine, the model correction unit corrects and optimizes the electric power model according to the analysis result, the cross verification unit verifies and optimizes the generalization capability and robustness of the model through a test data set, the output end of the model evaluation unit is connected with the input end of the feedback analysis unit, the output end of the feedback analysis unit is connected with the input end of the model correction unit, and the model correction unit is in bidirectional connection with the cross verification unit.
As a further technical scheme of the invention, the ethernet adopts a wireless anti-interference communication network to transmit the power parameter and the equipment running state data to the cloud for processing, the wireless anti-interference communication network performs noise reduction processing through a frequency selective filter, and improves the modulation performance and anti-interference capability of carrier communication through wide amplitude modulation, high speed modulation, code division multiplexing and error correction coding, and the wireless anti-interference communication network enhances the signal receiving capability through a sensitive radio frequency amplifier, a low noise index mixer and a gain antenna, and adopts a low-distortion high speed analog-to-digital converter to improve the signal receiving sensitivity.
Has the positive beneficial effects that:
The invention discloses a data monitoring and auxiliary management method based on a power model, which can effectively improve the accuracy and precision of power data processing, feature classification and data identification; the data are screened, standardized and quality controlled by a quality control strategy and a sensor signal conditioning circuit. Therefore, the data acquisition and processing efficiency can be greatly improved, and the data accuracy and the information quantity are improved; the combination of the distributed sensor network, the cloud server and the power model is adopted, so that the real-time monitoring of the power system is realized, and the monitoring accuracy and the real-time performance are improved; the model is trained and adjusted in real time through the increment updating unit and the self-adaptive adjusting module, so that the accuracy and the practicability of the model are improved; the self-adaptive particle swarm optimization decision model is adopted, and an optimal decision scheme can be obtained through real-time data monitoring and simulation prediction, so that the treatment efficiency and accuracy are improved; by constructing a complete electric power model, the model comprises a plurality of components such as a feature extraction unit, a data balance unit, a model training unit, a self-adaptive adjustment module, an increment updating unit, a processing analysis unit and the like, so that the running condition of the electric power system can be comprehensively reflected, and abnormal conditions can be monitored in real time; and the automation degree and the intelligent degree are high.
Drawings
FIG. 1 is a schematic flow chart of a data monitoring and auxiliary governance method based on a power model according to the present invention;
FIG. 2 is a block diagram of a power model in a data monitoring and auxiliary governance method based on the power model according to the present invention;
FIG. 3 is a schematic diagram of a real-time feedback verification mechanism in a data monitoring and auxiliary governance method based on a power model according to the present invention;
FIG. 4 is a schematic flow chart of an adaptive particle swarm optimization decision model in a data monitoring and auxiliary governance method based on an electric power model;
FIG. 5 is a schematic diagram of a model architecture of an adaptive particle swarm optimization decision model in a data monitoring and assisted remediation method based on a power model according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
A data monitoring and auxiliary governance method based on a power model comprises the following steps:
Step one, data collection and preprocessing, namely collecting power parameters and equipment operation state data in real time through a distributed sensor network and monitoring nodes, transmitting the collected data to a cloud end for storage and processing by adopting an Ethernet, wherein the power parameters and the equipment operation state data screen, normalize and control the quality of the collected data through a quality control strategy and a sensor signal conditioning circuit, and the power parameters and the equipment operation state data at least comprise operation voltage, current, output power, a switch state and a topological structure of a power system;
Constructing a power model, namely constructing the power model through a cloud server, wherein the cloud server constructs the power model based on historical data and real-time data, the power model comprises a feature extraction unit, a data balance unit, a model training unit, an adaptive adjustment module, an increment updating unit and a processing analysis unit, the output end of the feature extraction unit is connected with the input end of the data balance unit, the output end of the data balance unit is connected with the input end of the model training unit, the model training unit is in bidirectional connection with the adaptive adjustment module, the output end of the model training unit is connected with the input end of the processing analysis unit, and the output end of the increment updating unit is connected with the input ends of the model training unit and the processing analysis unit;
Step three, real-time data monitoring, namely, monitoring the power system in real time based on the power model, comparing and analyzing actually collected data with a simulation calculation result through a power abnormality iterative optimization model by the power model, and judging the abnormal condition of the current power parameter according to the comparison result;
Step four, simulation prediction, wherein the power model adopts a simulation dynamic simulation method to perform simulation prediction on the power parameter change trend, and evaluates the stability and reliability of the power system according to preset rules;
Step five, auxiliary governance decision making, namely carrying out auxiliary governance decision making of the electric power system through a self-adaptive particle swarm optimization decision making model, wherein the self-adaptive particle swarm optimization decision making model obtains an optimal decision scheme of the electric power system according to real-time data monitoring results and simulation prediction results;
and step six, model improvement and optimization, wherein the electric power model is optimized and improved through a real-time feedback verification mechanism.
In the above embodiment, the feature extraction unit maps the high-dimensional data to the low-dimensional space through linear discriminant analysis to achieve extraction of the power parameter features, the data balancing unit performs equalization processing on the extracted features through synthesizing minority oversampling, and performs training sample proportion division by adopting a method combining cross validation and random division, the model training unit performs training on the equalized features through a support vector machine, the adaptive adjustment module performs adaptive adjustment on parameters of the model according to real-time data, the incremental update unit performs incremental learning and updating on historical data and real-time data through a time window, and the processing analysis unit performs processing and analysis by adopting a data rule engine and a data mining algorithm.
In the above embodiment, the power anomaly iterative optimization model analyzes the anomaly node of the power system through power parameter time sequence sampling, the power anomaly iterative optimization model monitors the anomaly node in the power system in real time through phase space reconstruction and time sequence iteration, the phase space reconstruction extracts the operation feature of the power system through mapping power parameters onto a phase space, and a phase space orbit intermediate vector distance calculation formula of the power parameters is as follows:
（1）
in the case of the formula (1), Phase space orbit intermediate vector spacing representing power parameters,/>Distribution cross term coefficient representing power parameter,/>Representing the power load intensity of the power system,/>And/>Representing equalization coefficients of the data samples, the time series iterations monitoring the power anomaly parameter time series quantities through power parameter time series scalar analysis and feature extraction, formulated as:
（2）
In the formula (2) of the present invention, Represents the nth iteration sequence quantity of the power parameter,/>Multi-element quantitative value function representing observed power anomaly parameter scalar,/>A multivariate magnitude function representing a time series scalar of the power parameter,Representing measurement errors,/>Representing the initial sampling time,/>And (3) representing a sampling interval, wherein n represents the iteration times, the power anomaly iteration optimization model obtains an information fusion result based on phase space reconstruction and time-frequency feature extraction, and a calculation formula is as follows:
（3）
In the formula (3) of the present invention, Representing the fusion result of phase space reconstruction and time-frequency characteristic extraction information,/>Representing the embedding dimension of a power anomaly load sequence in a time frequency,/>Time-frequency characteristics representing power parameters; comparing and analyzing the information fusion result with the simulation calculation result, wherein the calculation formula is as follows:
（4）
in the formula (4) of the present invention, Representing the actual collected data,/>Representing simulation calculation results,/>The difference degree between the actually collected data and the simulation calculation result is represented, when the actual error exceeds the error threshold, the abnormal node of the current power parameter is judged, and the error threshold calculation formula is as follows:
（5）
in the formula (5) of the present invention, Representing an error threshold value,/>Representing standard deviation of data,/>Representing a multiple of the error threshold for controlling the sensitivity of the error threshold.
In a specific embodiment, the power anomaly iterative optimization model is a model for monitoring anomaly nodes in a power system in real time through phase space reconstruction and time sequence iteration by sampling a power parameter time sequence. Phase space reconstruction is the mapping of power parameters onto phase space to extract operational characteristics of the power system for better analysis and handling of anomalies. The time sequence iteration means that the prediction accuracy and the robustness of the model are improved by predicting and analyzing the power parameter time sequence data and continuously optimizing and adjusting the model according to the actual data. By the method, the abnormal nodes in the power system can be monitored in real time, and the abnormal nodes can be processed and repaired in time. The power anomaly iterative optimization model comprises several hardware working environments:
Data acquisition unit: the real-time data acquisition device is used for acquiring real-time data of each node in the power system, including parameters such as voltage, current and power. The data collector needs to have the characteristics of high precision, high stability, high reliability and the like.
A data storage device: and the device is used for storing the acquired real-time data and simulation calculation results. The data storage device needs to have the characteristics of large capacity, fast reading and writing, high reliability and the like.
And (3) a server: the power anomaly iterative optimization method is used for processing and analyzing the acquired data and running a power anomaly iterative optimization model. The server needs to have the characteristics of high performance, high stability, high reliability and the like.
Network equipment: and the data transmission channel is used for connecting all hardware devices and constructing a data transmission channel. The network device needs to have low latency, high bandwidth, high security, and the like.
An external sensor: is used for collecting environmental parameters such as temperature, humidity, wind speed and the like. The external sensor is required to have the characteristics of high precision, high stability, high reliability and the like.
In summary, the hardware working environment of the power anomaly iterative optimization model needs to include a data collector, a data storage device, a server, a network device, an external sensor and other components, so as to realize real-time monitoring and optimization of the power system.
And (3) respectively adopting an electric power abnormal iteration optimization model (A group) and a support vector machine (B group) to carry out comparison experiments, respectively installing the electric power abnormal iteration optimization model (A group) and the support vector machine (B group) on the same electric power system test platform, debugging and optimizing, designing the working conditions of the electric power system, including abnormal conditions such as load change, power grid short circuit and the like, acquiring time sequence data of electric power parameters by utilizing simulation data acquisition, respectively running the electric power abnormal iteration optimization model and other selected algorithms to monitor the electric power parameters in real time, analyzing abnormal conditions of electric power nodes, comparing experimental data with simulation calculation results, calculating the accuracy of the model, comparing the diagnosis speeds of the models, and recording in a table 1.
Table 1 results statistics table
The experimental result shows that the accuracy of the power abnormality detection by adopting the power abnormality iterative optimization model is higher than that by adopting the support vector machine, and the diagnosis speed is faster, so that the power abnormality iterative optimization model has better effect in the aspect of power abnormality detection, can detect the abnormal condition in the power system more quickly and accurately, and can process and maintain in time. The model effect can be evaluated more stably by repeating the experiment for a plurality of times, and the result fully proves the superiority of the power anomaly iterative optimization model.
In the above embodiment, the simulation dynamic simulation method uses a cloud database and a monitoring computer as a simulation control unit of the power system, the cloud database and the monitoring computer simulate dynamic operation of the power equipment according to real-time monitoring data and adaptively adjust a control strategy according to a preset rule, the cloud database and the monitoring computer simulate a power parameter variation trend through a component object model COM and transmit a simulation control instruction through a wireless communication bridge, and the component object model COM realizes transmission and reception of the simulation control instruction through a serial communication protocol.
In a specific embodiment, on a cloud database server, an actual power system model is established according to actual power system conditions. The model comprises parameters such as voltage, current, power and the like of each node, and the influence of factors such as line loss, load fluctuation and the like on the power system is considered. On the basis of the model, the operation conditions of the simulation are set, including the initial state, the load change condition, the fault condition and the like of the power system, and the time range and the precision of the simulation are set. Uploading the power model to a cloud computing server, performing simulation computation on the cloud computing server, simulating the change trend of the power parameters, and generating a simulation result of the power system. In the simulation calculation process, simulation parameters can be adjusted according to actual needs so as to further verify the stability and reliability of the power system. And analyzing the obtained simulation result, evaluating the stability and reliability of the power system, and formulating a reasonable improvement scheme according to a preset rule. Various aspects of the power system, such as voltage stability, frequency stability, power load balancing, fault tolerance, etc., need to be considered in assessing the stability and reliability of the power system. Based on analysis results and expected effects, an improvement scheme is formulated, power system optimization is performed on a cloud database server, stability and reliability of the power system are improved through the aspects of optimizing the structure, control strategy, fault treatment and the like of the power system, and the purpose of optimizing operation of the power system is achieved. And uploading the optimized power system model to a cloud computing server again for testing and verifying, checking the performance and stability of the system, and adjusting and optimizing according to actual results so as to ensure that the power system can run stably and reliably. In the process, the monitoring computer can monitor the running state of the power system in real time and discover problems and abnormal conditions in time.
In the above embodiment, the adaptive particle swarm optimization decision model includes an input layer, an adaptive parameter adjustment layer, a particle update layer, an adaptive weight layer, a multi-neighborhood search layer, and an output layer, and the adaptive particle swarm optimization decision model includes the following steps:
S1, initializing input, namely setting initial population size, maximum iteration times, inertia weight and neighborhood search parameters through the input layer, and inputting an initial particle swarm, an objective function to be optimized and a decision variable into the adaptive particle swarm optimization decision model;
s2, self-adaptive parameter adjustment, wherein the self-adaptive parameter adjustment layer is used for setting control parameters, objective functions and self-adaptive algorithms according to optimal solutions of particle swarms and historical optimal solutions, and dynamically adjusting the neighborhood searching range based on different neighborhood searching strategies;
S3, particle updating, namely calculating the fitness value of the particle and updating the speed and the position of the particle through the particle updating layer, wherein the particle updating layer calculates the fitness value of the particle according to the current position and the speed and updates the position and the speed of the particle;
S4, self-adaptive inertia weight adjustment, wherein the inertia weight is dynamically adjusted through the self-adaptive weight layer, the self-adaptive weight layer self-adaptively updates the inertia weight according to the current iteration times and the global history optimal fitness value, and the self-adaptive weight layer adjusts the change trend of the inertia weight according to the change condition of the objective function;
S5, multi-neighborhood searching, wherein a plurality of neighborhood searching strategies are formulated through the multi-neighborhood searching layer, and the plurality of neighborhood searching strategies comprise global neighborhood searching, local neighborhood searching and chaotic searching;
and S6, outputting an optimal decision scheme, and outputting an optimal solution and an optimal position of particles through the output layer when the maximum iteration number is reached or the termination condition is met, otherwise, repeating the S2.
In a specific embodiment, the adaptive particle swarm optimization decision model can find a global optimal solution faster through strategies such as adaptive adjustment of inertia weight, multi-neighborhood search and the like, so that the adaptive particle swarm optimization decision model is widely applied to various fields. Compared with the standard particle swarm algorithm, the main advantages of the algorithm are as follows: adaptive inertial weights: the algorithm introduces the idea of self-adaptive inertia weight, and can adaptively adjust the size and the change trend of the inertia weight according to the change condition of an objective function so as to achieve global and local search balance and improve the convergence rate of the algorithm. Multi-neighborhood search: the algorithm adopts a multi-neighborhood search strategy, including global neighborhood search, local neighborhood search, chaos search and the like, so that the algorithm can jump out a local optimal solution, and the global search performance is effectively improved. Adaptive population update: the algorithm adjusts various parameters according to the fitness value of each particle and the historical optimal position of the individual through an adaptive population updating mechanism so as to realize better algorithm performance. The application of algorithms is very widespread, with one of the most typical areas being power system optimization. For example, in grid planning, algorithms can implement grid optimization designs, including creating, expanding, or retrofitting grids to meet future power demands. In the aspect of power generation scheduling, the algorithm can reasonably arrange the power generation plans of all the generator sets according to market demands and real-time power demands so as to achieve the optimal balance of economy and reliability.
In the above embodiment, the real-time feedback verification mechanism includes a model evaluation unit, a feedback analysis unit, a model correction unit and a cross verification unit, where the model evaluation unit evaluates the power model by using a mean square error, an average absolute error and an R-square coefficient, the feedback analysis unit performs real-time analysis on an evaluation result by using a real-time stream processing engine, the model correction unit corrects and optimizes the power model according to the analysis result, the cross verification unit verifies and optimizes generalization capability and robustness of the model by using a test dataset, an output end of the model evaluation unit is connected with an input end of the feedback analysis unit, an output end of the feedback analysis unit is connected with an input end of the model correction unit, and the model correction unit is connected with the cross verification unit in a bidirectional manner.
In a particular embodiment, mean Square Error (MSE) and Mean Absolute Error (MAE) are common model evaluation metrics. The MSE measures the average quadratic error between the actual and predicted values, and the MAE measures the average absolute error between the actual and predicted values. The smaller their values, the better the predictive power of the model. The R-square coefficient is another commonly used model evaluation index, which is an index for measuring the fitting degree of a model, and the closer to 1, the better the model is fitted. The R-side coefficient typically has a value between 0 and 1, the closer to 1, indicating a stronger predictive power of the model. By adopting the indexes, the model evaluation module can evaluate and analyze the model to identify problems and defects of the model, and the feedback analysis unit analyzes the evaluation result so as to further correct and optimize the model. Meanwhile, through the cross verification module, the generalization capability and the robustness of the model can be verified, so that the optimal model parameters and structure can be determined. Through continuous model evaluation, analysis and correction, the accuracy and reliability of the power model are improved, and more accurate and effective decision support is provided for stable operation and sustainable development of the power system.
In the above embodiment, the ethernet uses a wireless anti-interference communication network to transmit the power parameter and the device running state data to the cloud for processing, where the wireless anti-interference communication network performs noise reduction processing through a frequency selective filter, and improves the modulation performance and anti-interference capability of carrier communication through wide amplitude modulation, high speed modulation, code division multiplexing and error correction coding, and the wireless anti-interference communication network enhances the signal receiving capability through a sensitive radio frequency amplifier, a low noise index mixer and a gain antenna, and improves the signal receiving sensitivity through a low-distortion high speed analog-to-digital converter.
In a specific embodiment, the ethernet uses a lorewan wireless network to transmit the power parameters and the device running state data to the cloud for processing. LoRaWAN is a wireless communication technology with low power consumption, long distance and wide area, is suitable for large-scale low-power consumption application scenes of the Internet of things such as a distributed sensor network, and has good anti-interference performance and coverage capability. In the data transmission process, the wireless anti-interference communication network adopts a plurality of techniques such as a frequency spectrum spreading technique, a error division coding technique and the like, so that the modulation performance and the anti-interference capability of carrier communication are improved, and the reliability and the stability of data transmission are ensured. Meanwhile, the wireless anti-interference communication network improves the signal receiving capacity through the technologies of a sensitive radio frequency amplifier, a low noise index mixer, a high gain antenna and the like, and improves the signal receiving sensitivity by adopting a low-distortion high-speed analog-to-digital converter, so that the integrity and the accuracy of data in the transmission process are ensured. In the aspect of cloud data processing, methods such as real-time data stream processing technology and machine learning can be adopted to rapidly analyze and process mass data, and the running state characteristics of the power system are extracted so as to support applications such as real-time monitoring, fault prediction and intelligent optimization of the power system.
While specific embodiments of the present invention have been described above, it will be understood by those skilled in the art that these specific embodiments are by way of example only, and that various omissions, substitutions, and changes in the form and details of the methods and systems described above may be made by those skilled in the art without departing from the spirit and scope of the invention. For example, it is within the scope of the present invention to combine the above-described method steps to perform substantially the same function in substantially the same way to achieve substantially the same result. Accordingly, the scope of the invention is limited only by the following claims.
Claims (5)
1. The utility model provides a data monitoring and auxiliary governance method based on electric power model which characterized in that: the method comprises the following steps:
Step one, data collection and preprocessing, namely collecting power parameters and equipment operation state data in real time through a distributed sensor network and monitoring nodes, transmitting the collected data to a cloud end for storage and processing by adopting an Ethernet, wherein the power parameters and the equipment operation state data screen, normalize and control the quality of the collected data through a quality control strategy and a sensor signal conditioning circuit, and the power parameters and the equipment operation state data at least comprise operation voltage, current, output power, a switch state and a topological structure of a power system;
Constructing a power model, namely constructing the power model through a cloud server, wherein the cloud server constructs the power model based on historical data and real-time data, the power model comprises a feature extraction unit, a data balance unit, a model training unit, an adaptive adjustment module, an increment updating unit and a processing analysis unit, the output end of the feature extraction unit is connected with the input end of the data balance unit, the output end of the data balance unit is connected with the input end of the model training unit, the model training unit is in bidirectional connection with the adaptive adjustment module, the output end of the model training unit is connected with the input end of the processing analysis unit, and the output end of the increment updating unit is connected with the input ends of the model training unit and the processing analysis unit;
Step three, real-time data monitoring, namely, monitoring the power system in real time based on the power model, comparing and analyzing actually collected data with a simulation calculation result through a power abnormality iterative optimization model by the power model, and judging the abnormal condition of the current power parameter according to the comparison result;
Step four, simulation prediction, wherein the power model adopts a simulation dynamic simulation method to perform simulation prediction on the power parameter change trend, and evaluates the stability and reliability of the power system according to preset rules;
Step five, auxiliary governance decision making, namely carrying out auxiliary governance decision making of the electric power system through a self-adaptive particle swarm optimization decision making model, wherein the self-adaptive particle swarm optimization decision making model obtains an optimal decision scheme of the electric power system according to real-time data monitoring results and simulation prediction results;
Step six, model improvement and optimization, wherein the electric power model is optimized and improved through a real-time feedback verification mechanism;
The power anomaly iterative optimization model analyzes anomaly nodes of the power system through power parameter time sequence sampling, monitors the anomaly nodes in the power system in real time through phase space reconstruction and time sequence iteration, the phase space reconstruction extracts operation characteristics of the power system through mapping power parameters onto a phase space, and a phase space orbit intermediate vector distance calculation formula of the power parameters is as follows:
（1）
in the case of the formula (1), Phase space orbit intermediate vector spacing representing power parameters,/>Distribution cross term coefficient representing power parameter,/>Representing the power load intensity of the power system,/>And/>Representing equalization coefficients of the data samples, the time series iterations monitoring the power anomaly parameter time series quantities through power parameter time series scalar analysis and feature extraction, formulated as:
) （2）
In the formula (2) of the present invention, Represents the nth iteration sequence quantity of the power parameter,/>Multi-element quantitative value function representing observed power anomaly parameter scalar,/>Multi-element magnitude function representing a power parameter time series scalar,/>Representing measurement errors,/>Representing the initial sampling time,/>And (3) representing a sampling interval, wherein n represents the iteration times, the power anomaly iteration optimization model obtains an information fusion result based on phase space reconstruction and time-frequency feature extraction, and a calculation formula is as follows:
（3）
In the formula (3) of the present invention, Representing the fusion result of phase space reconstruction and time-frequency characteristic extraction information,/>Representing the embedding dimension of a power anomaly load sequence in a time frequency,/>Time-frequency characteristics representing power parameters; comparing and analyzing the information fusion result with the simulation calculation result, wherein the calculation formula is as follows:
（4）
in the formula (4) of the present invention, Representing the actual collected data,/>Representing simulation calculation results,/>The difference degree between the actually collected data and the simulation calculation result is represented, when the actual error exceeds the error threshold, the abnormal node of the current power parameter is judged, and the error threshold calculation formula is as follows:
（5）
in the formula (5) of the present invention, Representing an error threshold value,/>Representing standard deviation of data,/>Representing a multiple of the error threshold for controlling the sensitivity of the error threshold;
The adaptive particle swarm optimization decision model comprises an input layer, an adaptive parameter adjusting layer, a particle updating layer, an adaptive weight layer, a multi-neighborhood searching layer and an output layer, and the adaptive particle swarm optimization decision model comprises the following steps:
S1, initializing input, namely setting initial population size, maximum iteration times, inertia weight and neighborhood search parameters through the input layer, and inputting an initial particle swarm, an objective function to be optimized and a decision variable into the adaptive particle swarm optimization decision model;
s2, self-adaptive parameter adjustment, wherein the self-adaptive parameter adjustment layer is used for setting control parameters, objective functions and self-adaptive algorithms according to optimal solutions of particle swarms and historical optimal solutions, and dynamically adjusting the neighborhood searching range based on different neighborhood searching strategies;
S3, particle updating, namely calculating the fitness value of the particle and updating the speed and the position of the particle through the particle updating layer, wherein the particle updating layer calculates the fitness value of the particle according to the current position and the speed and updates the position and the speed of the particle;
S4, self-adaptive inertia weight adjustment, wherein the inertia weight is dynamically adjusted through the self-adaptive weight layer, the self-adaptive weight layer self-adaptively updates the inertia weight according to the current iteration times and the global history optimal fitness value, and the self-adaptive weight layer adjusts the change trend of the inertia weight according to the change condition of the objective function;
S5, multi-neighborhood searching, wherein a plurality of neighborhood searching strategies are formulated through the multi-neighborhood searching layer, and the plurality of neighborhood searching strategies comprise global neighborhood searching, local neighborhood searching and chaotic searching;
and S6, outputting an optimal decision scheme, and outputting an optimal solution and an optimal position of particles through the output layer when the maximum iteration number is reached or the termination condition is met, otherwise, repeating the S2.
2. The method for monitoring and assisting in managing data based on a power model according to claim 1, wherein: the feature extraction unit maps high-dimensional data to a low-dimensional space through linear discriminant analysis to extract power parameter features, the data balance unit performs balance processing on the extracted features through synthesizing minority oversampling, and performs training sample proportion division by adopting a method combining cross validation and random division, the model training unit performs training on the balanced features through a support vector machine, the self-adaptive adjustment module performs self-adaptive adjustment on parameters of the model according to real-time data, the increment updating unit performs increment learning and updating on historical data and real-time data through a time window, and the processing analysis unit performs processing and analysis by adopting a data rule engine and a data mining algorithm.
3. The method for monitoring and assisting in managing data based on a power model according to claim 1, wherein: the simulation dynamic simulation method adopts a cloud database and a monitoring computer as a simulation control unit of the power system, the cloud database and the monitoring computer simulate dynamic operation of power equipment according to real-time monitoring data and carry out self-adaptive adjustment on a control strategy according to preset rules, the cloud database and the monitoring computer simulate power parameter change trend through a component object model COM and carry out transmission of simulation control instructions through a wireless communication bridge, and the component object model COM realizes transmission and reception of the simulation control instructions through a serial communication protocol.
4. The method for monitoring and assisting in managing data based on a power model according to claim 1, wherein: the real-time feedback verification mechanism comprises a model evaluation unit, a feedback analysis unit, a model correction unit and a cross verification unit, wherein the model evaluation unit evaluates the electric model by adopting mean square error, average absolute error and R-square coefficient, the feedback analysis unit carries out instant analysis on an evaluation result through a real-time stream processing engine, the model correction unit corrects and optimizes the electric model according to the analysis result, the cross verification unit verifies and optimizes the generalization capability and robustness of the model through a test data set, the output end of the model evaluation unit is connected with the input end of the feedback analysis unit, the output end of the feedback analysis unit is connected with the input end of the model correction unit, and the model correction unit is in bidirectional connection with the cross verification unit.
5. The method for monitoring and assisting in managing data based on a power model according to claim 1, wherein: the Ethernet adopts a wireless anti-interference communication network to transmit the power parameters and the equipment running state data to the cloud for processing, the wireless anti-interference communication network carries out noise reduction processing through a frequency selective filter and improves the modulation performance and the anti-interference capability of carrier communication through wide amplitude modulation, high speed modulation, code division multiplexing and error correction coding, and the wireless anti-interference communication network enhances the signal receiving capability through a sensitive radio frequency amplifier, a low noise index mixer and a gain antenna and improves the signal receiving sensitivity through a low-distortion high speed analog-to-digital converter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410317710.0A CN117909855B (en) | 2024-03-20 | 2024-03-20 | Data monitoring and auxiliary governance method based on electric power model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410317710.0A CN117909855B (en) | 2024-03-20 | 2024-03-20 | Data monitoring and auxiliary governance method based on electric power model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117909855A CN117909855A (en) | 2024-04-19 |
| CN117909855B true CN117909855B (en) | 2024-05-10 |
Family
ID=90682352
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410317710.0A Active CN117909855B (en) | 2024-03-20 | 2024-03-20 | Data monitoring and auxiliary governance method based on electric power model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117909855B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116523640A (en) * | 2023-04-08 | 2023-08-01 | 上海圈讯科技股份有限公司 | Financial information management system based on scheduling feedback algorithm |
| CN117495210A (en) * | 2024-01-02 | 2024-02-02 | 衡水烨通建设工程有限公司 | Highway concrete construction quality management system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109871259A (en) * | 2013-08-16 | 2019-06-11 | 运软网络科技(上海)有限公司 | A kind of imitative brain calculates the method and system of virtualization |
-
2024
- 2024-03-20 CN CN202410317710.0A patent/CN117909855B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116523640A (en) * | 2023-04-08 | 2023-08-01 | 上海圈讯科技股份有限公司 | Financial information management system based on scheduling feedback algorithm |
| CN117495210A (en) * | 2024-01-02 | 2024-02-02 | 衡水烨通建设工程有限公司 | Highway concrete construction quality management system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117909855A (en) | 2024-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200327435A1 (en) | Systems and methods for sequential power system model parameter estimation | |
| Liu et al. | Wind power plant prediction by using neural networks | |
| CN110530650B (en) | Method for monitoring performance state of heavy-duty gas turbine based on generalized regression neural network and box diagram analysis | |
| CN116992399B (en) | Power equipment operation and maintenance assessment method based on power data analysis | |
| CN112187528B (en) | Industrial control system communication flow online monitoring method based on SARIMA | |
| CN112734128A (en) | 7-day power load peak value prediction method based on optimized RBF | |
| CN117354652B (en) | High-speed multimode optical module performance optimization method based on variation self-encoder | |
| CN112232386B (en) | Voltage sag severity prediction method based on support vector machine | |
| CN114583767B (en) | Data-driven wind power plant frequency modulation response characteristic modeling method and system | |
| CN114994537A (en) | EKF-LSTM-based lithium battery state of charge joint prediction measurement method | |
| Zhu et al. | Research on PSO‐ARMA‐SVR Short‐Term Electricity Consumption Forecast Based on the Particle Swarm Algorithm | |
| CN117909855B (en) | Data monitoring and auxiliary governance method based on electric power model | |
| Zhao et al. | Lithium‐Ion Battery State‐of‐Health Estimation Method Using Isobaric Energy Analysis and PSO‐LSTM | |
| CN114169118A (en) | Power distribution network topological structure identification method considering distributed power supply output correlation | |
| CN114552653A (en) | Grid-connected inverter system control parameter multi-objective optimization method based on intelligent algorithm | |
| Li et al. | Extraction of Abnormal Points from On-line Operation Data of Intelligent Meter Based on LSTM | |
| CN113537338A (en) | Robust line parameter identification method based on LSTM neural network and improved SCADA data | |
| Luo et al. | Fine-grained bandwidth estimation for smart grid communication network | |
| CN108009686B (en) | Photovoltaic power generation power prediction method, device and system | |
| Hong et al. | Deep-Learning-Assisted Topology Identification and Sensor Placement for Active Distribution Network | |
| CN117878925B (en) | Power transmission data control method and system of smart power grid | |
| Bai et al. | Abnormal Detection Scheme of Substation Equipment based on Intelligent Fusion Terminal | |
| CN117834455B (en) | Electric power Internet of things data transmission simulation method and system | |
| CN117829822B (en) | Power transformer fault early warning method and system | |
| CN111950853B (en) | Electric power running state white list generation method based on information physical bilateral data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |