CN117709587A - Digital management and control method and system based on quantitative fee prediction and analysis - Google Patents
Digital management and control method and system based on quantitative fee prediction and analysis Download PDFInfo
- Publication number
- CN117709587A CN117709587A CN202311715085.7A CN202311715085A CN117709587A CN 117709587 A CN117709587 A CN 117709587A CN 202311715085 A CN202311715085 A CN 202311715085A CN 117709587 A CN117709587 A CN 117709587A
- Authority
- CN
- China
- Prior art keywords
- data
- load
- analysis
- prediction
- rpa
- 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.)
- Pending
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 94
- 238000000034 method Methods 0.000 title claims abstract description 89
- 230000002159 abnormal effect Effects 0.000 claims abstract description 50
- 238000007726 management method Methods 0.000 claims abstract description 31
- 238000012545 processing Methods 0.000 claims abstract description 27
- 238000005516 engineering process Methods 0.000 claims abstract description 25
- 238000012544 monitoring process Methods 0.000 claims abstract description 18
- 230000004044 response Effects 0.000 claims abstract description 18
- 238000013439 planning Methods 0.000 claims abstract description 15
- 238000004220 aggregation Methods 0.000 claims abstract description 7
- 230000002776 aggregation Effects 0.000 claims abstract description 7
- 230000008569 process Effects 0.000 claims description 35
- 238000001514 detection method Methods 0.000 claims description 29
- 230000006399 behavior Effects 0.000 claims description 25
- 238000004422 calculation algorithm Methods 0.000 claims description 25
- 230000006870 function Effects 0.000 claims description 15
- 238000012417 linear regression Methods 0.000 claims description 14
- 238000007405 data analysis Methods 0.000 claims description 12
- 238000013136 deep learning model Methods 0.000 claims description 11
- 230000005856 abnormality Effects 0.000 claims description 10
- 238000009826 distribution Methods 0.000 claims description 10
- 230000015654 memory Effects 0.000 claims description 10
- 238000010801 machine learning Methods 0.000 claims description 9
- 238000012423 maintenance Methods 0.000 claims description 9
- 230000000737 periodic effect Effects 0.000 claims description 9
- 238000012731 temporal analysis Methods 0.000 claims description 9
- 238000000700 time series analysis Methods 0.000 claims description 9
- 230000009466 transformation Effects 0.000 claims description 9
- 230000009467 reduction Effects 0.000 claims description 8
- 238000003860 storage Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 7
- 238000012300 Sequence Analysis Methods 0.000 claims description 6
- 230000007246 mechanism Effects 0.000 claims description 6
- 238000005065 mining Methods 0.000 claims description 6
- 238000013486 operation strategy Methods 0.000 claims description 6
- 238000011161 development Methods 0.000 claims description 5
- 230000004927 fusion Effects 0.000 claims description 5
- 238000000513 principal component analysis Methods 0.000 claims description 5
- 238000013527 convolutional neural network Methods 0.000 claims description 4
- 230000001965 increasing effect Effects 0.000 claims description 4
- 239000000203 mixture Substances 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 230000003595 spectral effect Effects 0.000 claims description 4
- 206010000117 Abnormal behaviour Diseases 0.000 claims description 3
- 238000004140 cleaning Methods 0.000 claims description 3
- 239000002131 composite material Substances 0.000 claims description 3
- 230000010485 coping Effects 0.000 claims description 3
- 238000013480 data collection Methods 0.000 claims description 3
- 238000013461 design Methods 0.000 claims description 3
- 238000004146 energy storage Methods 0.000 claims description 3
- 230000002708 enhancing effect Effects 0.000 claims description 3
- 230000010354 integration Effects 0.000 claims description 3
- 230000007787 long-term memory Effects 0.000 claims description 3
- 238000003909 pattern recognition Methods 0.000 claims description 3
- 230000006403 short-term memory Effects 0.000 claims description 3
- 238000007619 statistical method Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 230000008901 benefit Effects 0.000 abstract description 4
- 238000004801 process automation Methods 0.000 abstract description 4
- 238000013468 resource allocation Methods 0.000 abstract description 2
- 230000002354 daily effect Effects 0.000 description 10
- 238000005457 optimization Methods 0.000 description 3
- 238000003491 array Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000007796 conventional method Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 208000012661 Dyskinesia Diseases 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000007727 cost benefit analysis Methods 0.000 description 1
- 238000013479 data entry Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a digital management and control method and a digital management and control system based on quantitative fee prediction and analysis, which relate to the technical field of RPA robot process automation and comprise the following steps: developing an RPA flow; collecting load data and daily electric quantity data, automatically analyzing the regularity of the data, and inquiring and counting report data according to business rules by utilizing a rule matching technology; automatically monitoring load fluctuation, early warning in advance, and carrying out data aggregation and processing by utilizing a rule matching technology; and the RPA robot performs power grid planning analysis and load prediction according to the processed and aggregated results. The digital control method based on quantitative fee prediction and analysis provided by the invention can obviously improve the accuracy and response speed of power prediction. The method has the advantages of more effectively managing the load of the power grid, optimizing the resource allocation, improving the response capability to abnormal conditions, realizing automatic monitoring of load fluctuation, early warning in advance and facilitating planning analysis and load prediction of the power grid.
Description
Technical Field
The invention relates to the technical field of RPA robot process automation, in particular to a digital management and control method and system based on quantitative fee prediction and analysis.
Background
Robot Process Automation (RPA) is a business process automation technology that uses software robots to simulate the operation of human users to implement automated execution of various business processes. These tasks include large-scale, repeatable, rule-based work such as data entry, report generation, and email. The RPA can be regarded as a digital employee by an enterprise, so that the enterprise or the employee can complete repeated and monotonous flow work, the manual error is reduced, the operation efficiency is improved, and the operation cost is reduced.
With the high-speed development of social economy, power load control and prediction become important guarantee links for meeting power requirements, ensuring power grid safety, improving power utilization efficiency, optimizing power equipment configuration and promoting cross-energy development.
At present, special people are required to count electric quantity and industry classification load every day, time and labor are consumed, and the data accuracy is low. The load trend cannot be intuitively embodied, the requirement on the accuracy of industry load prediction is higher because of the increasing supply and sales expansion requirements, the distribution and the increment of industry load and regional load cannot be intuitively embodied in the existing effective mode, and the production and operation development trend of the industry and the region and the electricity consumption cost condition of each enterprise cannot be intuitively calculated.
By using RPA, the accuracy and efficiency of power load prediction can be improved, thereby better managing power resources.
Disclosure of Invention
The present invention has been made in view of the above-described problems.
Therefore, the technical problems solved by the invention are as follows: the existing traditional power prediction method has the problems of insufficient accuracy, inflexible response and optimization of how to efficiently process complex data and abnormal conditions.
In order to solve the technical problems, the invention provides the following technical scheme: a digital management and control method based on quantitative fee prediction and analysis comprises the following steps:
developing an RPA process through an RPA process editor, wherein the RPA robot performs the RPA process according to a preset task execution period, task execution parameters and the RPA process;
collecting load data and daily electric quantity data, automatically analyzing the regularity of the data, and inquiring and counting report data according to business rules by utilizing a rule matching technology;
automatically monitoring load fluctuation, early warning in advance, and carrying out data aggregation and processing by utilizing a rule matching technology;
and the RPA robot performs power grid planning analysis and load prediction according to the processed and aggregated results.
As a preferable scheme of the digital management and control method based on quantitative fee prediction and analysis, the invention comprises the following steps: the development of the RPA process comprises the steps of selecting a proper RPA process editor according to data processing requirements, integration capacity and user friendliness, configuring an RPA editor environment, interfacing with a database and an analysis tool of a power system, and setting task execution period and execution parameters of the RPA robot according to power grid data processing requirements.
As a preferable scheme of the digital management and control method based on quantitative fee prediction and analysis, the invention comprises the following steps: the method comprises the steps of collecting load data and daily electric quantity data, wherein the data collection of an electric power system is automated by using an RPA tool, the required load and electric quantity data are extracted at fixed time by a design script, real-time data are obtained by using an API if the electric power system supports an API interface, manual operation is simulated by using the RPA tool if the electric power system does not support the API system, and a form is automatically filled in and a report is downloaded.
As a preferable scheme of the digital management and control method based on quantitative fee prediction and analysis, the invention comprises the following steps: the automatic data regularity analysis comprises the steps of performing preliminary data cleaning and processing on the collected data, and performing deep regularity analysis by applying linear regression analysis and a machine learning algorithm; carrying out time sequence analysis on the data set, and identifying periodic changes and trends of the load and the electric quantity; identifying different modes of power consumption by using a clustering algorithm;
the different modes comprise peak period load identification, abnormal consumption behavior analysis, periodic consumption mode mining and low load period detection;
the utilization rule matching technology comprises the steps of monitoring power load data when load identification is carried out in a peak period, analyzing current load conditions in real time, predicting a future load mode by utilizing historical data and a time sequence prediction model based on machine learning, adjusting power supply according to a prediction result, and implementing demand response measures;
when the abnormal consumption behavior analysis is performed, the power consumption data is analyzed in real time by applying an isolated forest abnormality detection algorithm, behavior which is obviously different from the conventional mode is identified, if the abnormal behavior is detected, a GIS system is integrated, a problem area is rapidly positioned and maintenance personnel are dispatched after an abnormal alarm is received, and a coping strategy is formulated by combining historical maintenance data and current operation information;
when the method is in regular consumption mode mining, a regular consumption mode is identified from historical load data by using a clustering and time sequence analysis method, the consumption mode is compared with calendar information, the relevance between the load mode and date is determined, and a power grid operation strategy is adjusted according to the identified consumption mode;
when the detection is carried out in the low-load period, the low-load period is continuously monitored and predicted by a mobile average line or autoregressive model statistical analysis method, the regularity of the period is analyzed, the maintenance and upgrading work of the power grid is carried out by utilizing the low-load period, the influence on daily operation is minimized, an energy storage system is activated in the low-load period, and redundant energy is stored for use in the high-load period;
and inquiring and counting report data according to the business rules, wherein the step of automatically summarizing the analysis result by using an RPA tool, and generating a business report.
As a preferable scheme of the digital management and control method based on quantitative fee prediction and analysis, the invention comprises the following steps: the automatic regular analysis of the data is represented by collecting load data and daily electric quantity data to form an original data set as X= { X 1 ,x 2 ,…,x n Data preprocessing and feature transformation, denoted,
X′＝φ(X)
wherein phi represents a composite function, including data normalization, principal component analysis, and higher order polynomial feature transformation, to extract nonlinear features and reduce dimensionality of the data;
linear regression analysis of the preprocessed dataset X' with multimodal fusion, expressed as,
wherein beta is ij Regression coefficient, alpha, representing the j-th model j Representing model weights, e representing error terms;
based on the results of the linear regression, the residual is further analyzed, and deeper data patterns are mined using a deep learning model, denoted,
R＝X′-Y
Z＝ψ(R)
wherein ψ represents a deep learning model;
a dynamic time series analysis and a nonlinear predictive model are introduced to process Z, denoted as,
T＝g(Z,T past ,Θ)
where g is a time series model representing a combination of autoregressive and nonlinear characteristics, Θ is a time series data T taking into account history past Model parameter sets of (2);
clustering the results of the time series analysis, and introducing high-dimensional data analysis techniques, expressed as,
C＝h(T)
P＝θ(C,Ξ)
wherein h represents a high-level clustering algorithm, which comprises a spectral clustering or Gaussian mixture model, θ represents a high-dimensional data analysis function, and the combination of the dimension reduction technology comprises t-SNE or UMAP, and xi represents related parameters;
the different modes further comprise analyzing a time sequence T of the power load data to identify the overall trend and the periodic variation of the load, determining standard thresholds of the peak and the low load, and judging a high load mode and a low load mode;
classifying the power load data C by using a clustering algorithm, identifying different load modes, marking a high load mode and a low load mode with obvious characteristics, entering a peak period load identification or low load period detection analysis flow if the trend of T and the clustering result of C are both directed to the high load or the low load, and entering an abnormal consumption behavior analysis or abnormal consumption behavior analysis flow if the trend of T and the clustering result of C are not directed to the high load or the low load.
As a preferable scheme of the digital management and control method based on quantitative fee prediction and analysis, the invention comprises the following steps: the load identification of entering the peak period or the detection and analysis flow of the low load period comprises analyzing data points in peak clusters, checking the distribution of the data points in a time sequence T, confirming the peak period if the load values of a plurality of continuous time points exceed a peak threshold value, analyzing the distribution of the data points in the low load clusters in the T if the load values of a plurality of continuous time points do not exceed the peak threshold value, and confirming the low load period if the load values of a plurality of continuous time points are lower than the low load threshold value;
the abnormal consumption behavior analysis or the abnormal consumption behavior analysis flow comprises the steps of utilizing the output Z of the deep learning model psi to identify an abnormal mode, defining the distance from a clustering center or the size of a prediction error as an abnormal index, and marking the abnormal consumption behavior if the Z shows a characteristic which is obviously different from the normal mode;
if the features are the same, the pattern recognition result P is used for analyzing the regularity of the consumption pattern, whether the P reveals the repeated consumption pattern is checked, and if the P shows obvious periodicity or regularity pattern, the regular consumption pattern is confirmed.
As a preferable scheme of the digital management and control method based on quantitative fee prediction and analysis, the invention comprises the following steps: identifying different modes of power consumption comprises identifying data points which cannot be classified into any pre-mode, tracking the variation condition of the data points by using a data monitoring technology, carrying out deep analysis on unclassified data, and detecting the stability of a power grid by using an anomaly detection algorithm in combination with real-time data and historical data;
the anomaly detection algorithm comprises the steps of constructing an anomaly detection model by utilizing the combination of a convolutional neural network and a long-term and short-term memory network, training the characteristics of unclassified data in a power system, applying the model to comprehensive analysis of real-time power load data and historical consumption modes, and adjusting a power grid operation strategy through an automatic decision system by combining the anomaly detection result of the model;
if the abnormal load detected by the model exceeds 10% of the normal range, triggering a secondary response mechanism, adjusting the power distribution of the area, implementing a slight load reduction measure, if the abnormal load detected by the model exceeds 20% of the normal range, triggering a primary response mechanism, monitoring the abnormal load type, reducing the power supply of a non-critical area when the high load is abnormal, enhancing the power supply of a critical infrastructure, and starting a standby power supply system; when the low load is abnormal, the generated energy is reduced, the power storage is increased, and the energy use is optimized;
and (3) implementing an automatic early warning system, and immediately sending an alarm to the cloud platform if the abnormality is detected, wherein the alarm comprises the abnormality type, the expected influence and the suggested emergency measures.
It is another object of the present invention to provide a digital management and control system based on quantitative fee prediction and analysis, which can solve the problems of inaccurate power prediction and inflexible response through advanced data analysis technology and machine learning algorithm.
In order to solve the technical problems, the invention provides the following technical scheme: a digital management and control system based on quantitative fee prediction and analysis, comprising: the system comprises an RPA flow editing module, a data acquisition module, a data processing module and a power grid planning module; the RPA process editing module is used for developing an RPA process through an RPA process editor, and the RPA robot is used for executing the task according to a preset task execution period, task execution parameters and the RPA process; the data acquisition module is used for acquiring load data and daily electric quantity data, automatically analyzing the regularity of the data, and inquiring and counting report data according to business rules by utilizing a rule matching technology; the data processing module is used for automatically monitoring load fluctuation, early warning in advance and carrying out data aggregation and processing by utilizing a rule matching technology; and the power grid planning module is used for carrying out power grid planning analysis and load prediction by the RPA robot according to the processed and aggregated results.
A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of a digital management method based on quantitative fee prediction and analysis as described above.
A computer readable storage medium having stored thereon a computer program, characterized in that the computer program when executed by a processor implements the steps of a digital management method based on royalty prediction and analysis as described above.
The invention has the beneficial effects that: the digital control method based on quantitative fee prediction and analysis provided by the invention can obviously improve the accuracy and response speed of power prediction. The method has the advantages of more effectively managing the load of the power grid, optimizing the resource allocation, improving the response capability to abnormal conditions, realizing automatic monitoring of load fluctuation, early warning in advance and facilitating planning analysis and load prediction of the power grid.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a digital management and control method based on quantitative fee prediction and analysis according to an embodiment of the present invention.
Fig. 2 is a diagram showing the same overall structure of digital management and control based on quantitative fee prediction and analysis according to a second embodiment of the present invention.
Detailed Description
So that the manner in which the above recited objects, features and advantages of the present invention can be understood in detail, a more particular description of the invention, briefly summarized above, may be had by reference to the embodiments, some of which are illustrated in the appended drawings. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways other than those described herein, and persons skilled in the art will readily appreciate that the present invention is not limited to the specific embodiments disclosed below.
Example 1
Referring to fig. 1, for one embodiment of the present invention, a digital management and control method based on quantitative fee prediction and analysis is provided, including:
and developing an RPA process by an RPA process editor, wherein the RPA robot performs the RPA process according to a preset task execution period, task execution parameters and the RPA process.
And collecting load data and daily electric quantity data, automatically analyzing the regularity of the data, and inquiring and counting report data according to business rules by utilizing a rule matching technology.
And automatically monitoring load fluctuation, early warning in advance, and carrying out data aggregation and processing by utilizing a rule matching technology.
And the RPA robot performs power grid planning analysis and load prediction according to the processed and aggregated results.
Developing the RPA flow comprises selecting a proper RPA flow editor according to data processing requirements, integration capability and user friendliness, configuring an RPA editor environment, interfacing with a database and an analysis tool of a power system, and setting task execution period and execution parameters of the RPA robot according to power grid data processing requirements.
The method comprises the steps of collecting load data and daily electric quantity data, wherein the data collection of an electric power system is automated by using an RPA tool, the required load and electric quantity data are extracted at regular time by a design script, real-time data are obtained by using an API if the electric power system supports an API interface, manual operation is simulated by using the RPA tool if the electric power system does not support the API system, and a form is automatically filled in and a report is downloaded.
Automatically performing regularity analysis on the data comprises the steps of performing preliminary data cleaning and processing on the acquired data, and performing deep regularity analysis by applying linear regression analysis and a machine learning algorithm; carrying out time sequence analysis on the data set, and identifying periodic changes and trends of the load and the electric quantity; different modes of power consumption are identified using a clustering algorithm.
The different modes comprise peak period load identification, abnormal consumption behavior analysis, periodic consumption mode mining and low load period detection.
The method comprises the steps of identifying load in a peak period, monitoring power load data, analyzing current load conditions in real time, predicting a future load mode by using historical data and a time sequence prediction model based on machine learning, adjusting power supply according to a prediction result, such as improving power supply to a critical area or reducing power supply to a non-critical area, and implementing demand response measures.
When the abnormal consumption behavior analysis is performed, the power consumption data is analyzed in real time by applying an isolated forest abnormality detection algorithm, behaviors which are obviously different from the conventional modes are identified, if the abnormal behaviors are detected, a GIS system is integrated, a problem area is rapidly positioned after an abnormal alarm is received, maintenance personnel are dispatched, and a coping strategy is formulated by combining historical maintenance data and current operation information.
When the method is in regular consumption mode mining, a regular consumption mode is identified from historical load data by using a clustering and time sequence analysis method, the consumption mode is compared with calendar information, the relevance between the load mode and date is determined, and a power grid operation strategy is adjusted according to the identified consumption mode.
When the low-load period is detected, the low-load period is continuously monitored and predicted by a mobile average line or autoregressive model statistical analysis method, the regularity of the period is analyzed, the low-load period is utilized for carrying out maintenance and upgrading work of the power grid, the influence on daily operation is minimized, an energy storage system is activated in the low-load period, and redundant energy is stored for use in the high-load period.
Inquiring and counting report data according to the business rules comprises automatically summarizing analysis results by using an RPA tool to generate a business report.
The automatic regular analysis of the data is represented by collecting load data and daily electric quantity data to form an original data set as X= { X 1 ,x 2 ,…,x n Data preprocessing and feature transformation, denoted,
X′＝φ(X)
wherein phi represents a composite function including data normalization, principal component analysis, and higher order polynomial feature transformation to extract nonlinear features and reduce dimensionality of the data.
Linear regression analysis of the preprocessed dataset X' with multimodal fusion, expressed as,
wherein beta is ij Regression coefficient, alpha, representing the j-th model j Representing model weights, e represents error terms.
Based on the results of the linear regression, the residual is further analyzed, and deeper data patterns are mined using a deep learning model, denoted,
R＝X′-Y
Z＝ψ(R)
where ψ represents the deep learning model.
A dynamic time series analysis and a nonlinear predictive model are introduced to process Z, denoted as,
T＝g(Z,T past ,Θ)
where g is a time series model representing a combination of autoregressive and nonlinear characteristics, Θ is a time series data T taking into account history past Is described.
Clustering the results of the time series analysis, and introducing high-dimensional data analysis techniques, expressed as,
C＝h(T)
P＝θ(C,Ξ)
wherein h represents a high-level clustering algorithm comprising a spectral clustering or Gaussian mixture model, θ represents a high-dimensional data analysis function, and the combined dimension reduction technology comprises t-SNE or UMAP, and xi represents related parameters.
X represents the original dataset in advanced data preprocessing and feature transformation, including all observed data points, X i Representing individual data points in the original dataset, phi represents a feature transformation function, including data normalization, principal Component Analysis (PCA) and higher order polynomial feature transformation, and X' represents the new dataset after phi function processing, including normalized and transformed features.
Y represents the output of the linear regression model in the linear regression analysis of the multimodal fusion, representing the predicted value, β 0j ,β ij Represents the intercept and regression coefficient, alpha, of the jth linear regression model j Representing the weight of the j-th model,for multi-model fusion, E represents the error term of the linear regression model.
In the residual analysis and the deep learning, R represents the residual, i.e. the difference between the original feature X' and the linear regression output Y, ψ represents a deep learning model function, such as a deep neural network or a convolutional neural network, and Z represents the output processed by the deep learning model ψ.
In dynamic time series analysis and nonlinear prediction, g represents a time series analysis function, and autoregressive and nonlinear characteristics are combined, T past Represents past time series data, Θ represents a parameter set of the time series model, and T represents an output of the time series analysis.
In the clustering and high-dimensional data analysis, h represents an advanced clustering algorithm function, such as spectral clustering or Gaussian mixture model, C represents a clustering result, theta represents a high-dimensional data analysis function, and in combination with a dimension reduction technology, xi represents a relevant parameter of the high-dimensional data analysis, and P represents the output of the high-dimensional data analysis and is used for identifying complex modes and relations in data.
The different modes further comprise analyzing the time sequence T of the power load data to identify the overall trend and the periodic variation of the load, determining the standard threshold values of the peak and the low load, and judging the high load mode and the low load mode;
classifying the power load data C by using a clustering algorithm, identifying different load modes, marking a high load mode and a low load mode with obvious characteristics, entering a peak period load identification or low load period detection analysis flow if the trend of T and the clustering result of C are both directed to the high load or the low load, and entering an abnormal consumption behavior analysis or abnormal consumption behavior analysis flow if the trend of T and the clustering result of C are not directed to the high load or the low load.
The process of load identification in peak time or detection and analysis in low load time comprises analyzing data points in peak clusters, checking distribution of the data points in time sequence T, confirming the peak time if load values of a plurality of continuous time points exceed a peak threshold value, analyzing the distribution of the data points in the low load clusters in T if the load values of the plurality of continuous time points do not exceed the peak threshold value, and confirming the low load time if the load values of the plurality of continuous time points are lower than the low load threshold value.
The abnormal consumption behavior analysis or the abnormal consumption behavior analysis flow comprises the steps of utilizing the output Z of the deep learning model psi to identify an abnormal mode, defining the distance from a clustering center or the size of a prediction error as an abnormal index, and marking the abnormal consumption behavior if the Z shows a characteristic which is significantly different from the normal mode.
If the features are the same, the pattern recognition result P is used for analyzing the regularity of the consumption pattern, whether the P reveals the repeated consumption pattern is checked, and if the P shows obvious periodicity or regularity pattern, the regular consumption pattern is confirmed.
Identifying different modes of power consumption comprises identifying data points which cannot be classified into any pre-mode, tracking the change condition of the data points by using a data monitoring technology, carrying out deep analysis on unclassified data, and detecting the stability of a power grid by using an anomaly detection algorithm in combination with real-time data and historical data.
The anomaly detection algorithm comprises the steps of constructing an anomaly detection model by utilizing the combination of a convolutional neural network and a long-term and short-term memory network, training the characteristics of unclassified data in the power system, applying the model to comprehensive analysis of real-time power load data and historical consumption modes, and adjusting a power grid operation strategy through an automatic decision system by combining the anomaly detection result of the model.
If the abnormal load detected by the model exceeds 10% of the normal range, triggering a secondary response mechanism, adjusting the power distribution of the area, implementing a slight load reduction measure, if the abnormal load detected by the model exceeds 20% of the normal range, triggering a primary response mechanism, monitoring the abnormal load type, reducing the power supply of a non-critical area when the high load is abnormal, enhancing the power supply of a critical infrastructure, and starting a standby power supply system; and when the load is abnormal, the generated energy is reduced, the power storage is increased, and the energy use is optimized.
And (3) implementing an automatic early warning system, and immediately sending an alarm to the cloud platform if the abnormality is detected, wherein the alarm comprises the abnormality type, the expected influence and the suggested emergency measures.
Example 2
Referring to fig. 2, for one embodiment of the present invention, a digital management and control system based on quantitative fee prediction and analysis is provided, including:
the system comprises an RPA flow editing module, a data acquisition module, a data processing module and a power grid planning module.
The RPA process editing module is used for developing an RPA process through the RPA process editor, and the RPA robot is used for executing the task according to the preset task execution period, the task execution parameters and the RPA process.
The data acquisition module is used for acquiring load data and daily electric quantity data, automatically analyzing the regularity of the data, and inquiring and counting report data according to business rules by utilizing a rule matching technology.
The data processing module is used for automatically monitoring load abnormal movement, early warning in advance and carrying out data aggregation and processing by utilizing a rule matching technology.
And the power grid planning module is used for carrying out power grid planning analysis and load prediction by the RPA robot according to the processed and aggregated results.
Example 3
One embodiment of the present invention, which is different from the first two embodiments, is:
the functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-only memory (ROM), a random access memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: electrical connection (electronic device), portable computer disk cartridge (magnetic device), random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (eeprom) with one or more wiring
(EPROM or flash memory), fiber optic means, and portable Compact Disc Read Only Memory (CDROM). In addition, the computer readable medium may even be paper or other suitable medium on which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.
It is to be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, may be implemented using any one or combination of the following techniques, as is well known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like.
Example 4
For one embodiment of the invention, a digital management and control method based on quantitative fee prediction and analysis is provided, and in order to verify the beneficial effects of the invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.
Grid scale: a medium-sized urban power grid with diversified load characteristics is selected. Data type: including historical power load data, climate change data, consumer behavior data, and the like. Time span: the historical data covers at least the power consumption of the past two years. Historical power data and related operational data of the selected grid are collected, while real-time data over a short period of time is collected. For the traditional method: and carrying out load prediction by applying a basic statistical model and carrying out abnormality detection by a simple threshold method. For the method of the invention: advanced data processing and machine learning algorithms are applied to load prediction and anomaly detection in complex modes. Performance evaluation: and comparing the performances of the two methods in the aspects of prediction accuracy, anomaly detection, data processing speed and the like. Cost-benefit analysis: and evaluating the reduction of the operation cost and the resource optimization efficiency. The experimental results are shown in table 1.
Table 1 comparative experiment table
As can be seen from table 1, conventional power system management generally relies on underlying data processing and predictive models, which do not perform well in handling complex power data and rapidly changing market demands. Conventional methods often fail to efficiently process large-scale data sets, and lack the ability to predict and handle grid load anomalies. Furthermore, the identification of peaks, valleys and abnormal patterns of power consumption is often slow to react, which constitutes an important obstacle in fast response and optimizing grid operation.
Limitations of existing methods in terms of data processing and prediction result in inefficient power grid operation and insufficient response during peak hours and emergency situations. Furthermore, the lack of application of advanced data analysis and machine learning techniques makes conventional methods unable to accurately predict and analyze complex patterns of electrical loads.
The method of the invention obviously improves the prediction accuracy through a high-level algorithm. The abnormal pattern is more accurately identified. The time consumption is obviously reduced, and the response speed is improved. Greatly reduces the operation cost and simultaneously realizes higher energy optimization efficiency.
It should be noted that the above embodiments are only for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention may be modified or substituted without departing from the spirit and scope of the technical solution of the present invention, which is intended to be covered in the scope of the claims of the present invention.
Claims (10)
1. A digital management and control method based on quantitative fee prediction and analysis is characterized by comprising the following steps:
developing an RPA process through an RPA process editor, wherein the RPA robot performs the RPA process according to a preset task execution period, task execution parameters and the RPA process;
collecting load data and daily electric quantity data, automatically analyzing the regularity of the data, and inquiring and counting report data according to business rules by utilizing a rule matching technology;
automatically monitoring load fluctuation, early warning in advance, and carrying out data aggregation and processing by utilizing a rule matching technology;
and the RPA robot performs power grid planning analysis and load prediction according to the processed and aggregated results.
2. The digital management and control method based on quantitative fee prediction and analysis according to claim 1, wherein: the development of the RPA process comprises the steps of selecting a proper RPA process editor according to data processing requirements, integration capacity and user friendliness, configuring an RPA editor environment, interfacing with a database and an analysis tool of a power system, and setting task execution period and execution parameters of the RPA robot according to power grid data processing requirements.
3. The digital management and control method based on quantitative fee prediction and analysis as set forth in claim 2, wherein: the method comprises the steps of collecting load data and daily electric quantity data, wherein the data collection of an electric power system is automated by using an RPA tool, the required load and electric quantity data are extracted at fixed time by a design script, real-time data are obtained by using an API if the electric power system supports an API interface, manual operation is simulated by using the RPA tool if the electric power system does not support the API system, and a form is automatically filled in and a report is downloaded.
4. A digital management and control method based on quantitative fee prediction and analysis as set forth in claim 3, wherein: the automatic data regularity analysis comprises the steps of performing preliminary data cleaning and processing on the collected data, and performing deep regularity analysis by applying linear regression analysis and a machine learning algorithm; carrying out time sequence analysis on the data set, and identifying periodic changes and trends of the load and the electric quantity; identifying different modes of power consumption by using a clustering algorithm;
the different modes comprise peak period load identification, abnormal consumption behavior analysis, periodic consumption mode mining and low load period detection;
the utilization rule matching technology comprises the steps of monitoring power load data when load identification is carried out in a peak period, analyzing current load conditions in real time, predicting a future load mode by utilizing historical data and a time sequence prediction model based on machine learning, adjusting power supply according to a prediction result, and implementing demand response measures;
when the abnormal consumption behavior analysis is performed, the power consumption data is analyzed in real time by applying an isolated forest abnormality detection algorithm, behavior which is obviously different from the conventional mode is identified, if the abnormal behavior is detected, a GIS system is integrated, a problem area is rapidly positioned and maintenance personnel are dispatched after an abnormal alarm is received, and a coping strategy is formulated by combining historical maintenance data and current operation information;
when the method is in regular consumption mode mining, a regular consumption mode is identified from historical load data by using a clustering and time sequence analysis method, the consumption mode is compared with calendar information, the relevance between the load mode and date is determined, and a power grid operation strategy is adjusted according to the identified consumption mode;
when the detection is carried out in the low-load period, the low-load period is continuously monitored and predicted by a mobile average line or autoregressive model statistical analysis method, the regularity of the period is analyzed, the maintenance and upgrading work of the power grid is carried out by utilizing the low-load period, the influence on daily operation is minimized, an energy storage system is activated in the low-load period, and redundant energy is stored for use in the high-load period;
and inquiring and counting report data according to the business rules, wherein the step of automatically summarizing the analysis result by using an RPA tool, and generating a business report.
5. The digital management and control method based on quantitative fee prediction and analysis according to claim 4, wherein: the automatic regular analysis of the data is represented by collecting load data and daily electric quantity data to form an original data set as X= { X 1 ,x 2 ,…,x n Data preprocessing and feature transformation, denoted,
X′＝φ(X)
wherein phi represents a composite function, including data normalization, principal component analysis, and higher order polynomial feature transformation, to extract nonlinear features and reduce dimensionality of the data;
linear regression analysis of the preprocessed dataset X' with multimodal fusion, expressed as,
wherein beta is ij Regression coefficient, alpha, representing the j-th model j Representing model weights, e representing error terms;
based on the results of the linear regression, the residual is further analyzed, and deeper data patterns are mined using a deep learning model, denoted,
R＝X′-Y
Z＝ψ(R)
wherein ψ represents a deep learning model;
a dynamic time series analysis and a nonlinear predictive model are introduced to process Z, denoted as,
T＝g(Z,T past ,Θ)
where g is a time series model representing a combination of autoregressive and nonlinear characteristics, Θ is a time series data T taking into account history past Model parameter sets of (2);
clustering the results of the time series analysis, and introducing high-dimensional data analysis techniques, expressed as,
C＝h(T)
P＝θ(C,Ξ)
wherein h represents a high-level clustering algorithm, which comprises a spectral clustering or Gaussian mixture model, θ represents a high-dimensional data analysis function, and the combination of the dimension reduction technology comprises t-SNE or UMAP, and xi represents related parameters;
the different modes further comprise analyzing a time sequence T of the power load data to identify the overall trend and the periodic variation of the load, determining standard thresholds of the peak and the low load, and judging a high load mode and a low load mode;
classifying the power load data C by using a clustering algorithm, identifying different load modes, marking a high load mode and a low load mode with obvious characteristics, entering a peak period load identification or low load period detection analysis flow if the trend of T and the clustering result of C are both directed to the high load or the low load, and entering an abnormal consumption behavior analysis or abnormal consumption behavior analysis flow if the trend of T and the clustering result of C are not directed to the high load or the low load.
6. The digital management and control method based on quantitative fee prediction and analysis according to claim 5, wherein: the load identification of entering the peak period or the detection and analysis flow of the low load period comprises analyzing data points in peak clusters, checking the distribution of the data points in a time sequence T, confirming the peak period if the load values of a plurality of continuous time points exceed a peak threshold value, analyzing the distribution of the data points in the low load clusters in the T if the load values of a plurality of continuous time points do not exceed the peak threshold value, and confirming the low load period if the load values of a plurality of continuous time points are lower than the low load threshold value;
the abnormal consumption behavior analysis or the abnormal consumption behavior analysis flow comprises the steps of utilizing the output Z of the deep learning model psi to identify an abnormal mode, defining the distance from a clustering center or the size of a prediction error as an abnormal index, and marking the abnormal consumption behavior if the Z shows a characteristic which is obviously different from the normal mode;
if the features are the same, the pattern recognition result P is used for analyzing the regularity of the consumption pattern, whether the P reveals the repeated consumption pattern is checked, and if the P shows obvious periodicity or regularity pattern, the regular consumption pattern is confirmed.
7. The digital management and control method based on quantitative fee prediction and analysis according to claim 6, wherein: identifying different modes of power consumption comprises identifying data points which cannot be classified into any pre-mode, tracking the variation condition of the data points by using a data monitoring technology, carrying out deep analysis on unclassified data, and detecting the stability of a power grid by using an anomaly detection algorithm in combination with real-time data and historical data;
the anomaly detection algorithm comprises the steps of constructing an anomaly detection model by utilizing the combination of a convolutional neural network and a long-term and short-term memory network, training the characteristics of unclassified data in a power system, applying the model to comprehensive analysis of real-time power load data and historical consumption modes, and adjusting a power grid operation strategy through an automatic decision system by combining the anomaly detection result of the model;
if the abnormal load detected by the model exceeds 10% of the normal range, triggering a secondary response mechanism, adjusting the power distribution of the area, implementing a slight load reduction measure, if the abnormal load detected by the model exceeds 20% of the normal range, triggering a primary response mechanism, monitoring the abnormal load type, reducing the power supply of a non-critical area when the high load is abnormal, enhancing the power supply of a critical infrastructure, and starting a standby power supply system; when the low load is abnormal, the generated energy is reduced, the power storage is increased, and the energy use is optimized;
and (3) implementing an automatic early warning system, and immediately sending an alarm to the cloud platform if the abnormality is detected, wherein the alarm comprises the abnormality type, the expected influence and the suggested emergency measures.
8. A system employing the digital management and control method based on quantitative fee prediction and analysis as set forth in any one of claims 1 to 7, comprising: the system comprises an RPA flow editing module, a data acquisition module, a data processing module and a power grid planning module;
the RPA process editing module is used for developing an RPA process through an RPA process editor, and the RPA robot is used for executing the task according to a preset task execution period, task execution parameters and the RPA process;
the data acquisition module is used for acquiring load data and daily electric quantity data, automatically analyzing the regularity of the data, and inquiring and counting report data according to business rules by utilizing a rule matching technology;
the data processing module is used for automatically monitoring load fluctuation, early warning in advance and carrying out data aggregation and processing by utilizing a rule matching technology;
and the power grid planning module is used for carrying out power grid planning analysis and load prediction by the RPA robot according to the processed and aggregated results.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the digital management and control method based on royalty prediction and analysis of any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium having stored thereon a computer program, which when executed by a processor, implements the steps of a digital management method based on royalty prediction and analysis according to any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311715085.7A CN117709587A (en) | 2023-12-14 | 2023-12-14 | Digital management and control method and system based on quantitative fee prediction and analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311715085.7A CN117709587A (en) | 2023-12-14 | 2023-12-14 | Digital management and control method and system based on quantitative fee prediction and analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117709587A true CN117709587A (en) | 2024-03-15 |
Family
ID=90156447
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311715085.7A Pending CN117709587A (en) | 2023-12-14 | 2023-12-14 | Digital management and control method and system based on quantitative fee prediction and analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117709587A (en) |
-
2023
- 2023-12-14 CN CN202311715085.7A patent/CN117709587A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Khanlari et al. | Prioritizing equipments for preventive maintenance (PM) activities using fuzzy rules | |
| CN111624931B (en) | Industrial park electricity utilization internet intelligent operation and maintenance management and control system and method | |
| CN117494292B (en) | Engineering progress management method and system based on BIM and AI large model | |
| US20230213895A1 (en) | Method for Predicting Benchmark Value of Unit Equipment Based on XGBoost Algorithm and System thereof | |
| CN117575334A (en) | Highway construction safety monitoring system | |
| CN117439256A (en) | Power station equipment management method and system based on Internet of things | |
| CN113807751B (en) | Safety risk grade assessment method and system based on knowledge graph | |
| CN117670378B (en) | Food safety monitoring method and system based on big data | |
| Alvarez Quiñones et al. | Machine learning for predictive maintenance scheduling of distribution transformers | |
| CN117236665B (en) | Material production scheduling optimization method and system | |
| Ouda et al. | Machine Learning and Optimization for Predictive Maintenance based on Predicting Failure in the Next Five Days. | |
| Kohli | Using machine learning algorithms on data residing in SAP ERP application to predict equipment failures | |
| Ahmed et al. | Enhancing stock portfolios for enterprise management and investment in energy industry | |
| Petruschke et al. | MACHINE LEARNING BASED IDENTIFICATION OF ENERGY EFFICIENCY MEASURES FOR MACHINE TOOLS USING LOAD PROFILES AND MACHINE SPECIFIC META DATA. | |
| Rizvi | Powering Efficiency: Exploring Artificial Intelligence for Real-time Energy Management in Buildings | |
| CN116780530A (en) | Large-scale new energy access power grid weak link identification prediction method and system | |
| CN117709587A (en) | Digital management and control method and system based on quantitative fee prediction and analysis | |
| CN114168409A (en) | Service system running state monitoring and early warning method and system | |
| Kulbovskyi et al. | Model and methods of management of efficiency of use of production and technological potential of rolling stock of rail transport | |
| Li et al. | An application on building information model to procurement strategy of copper raw material with big data analytics | |
| Hao et al. | Demand forecasting for rush repair spare parts of power equipment using fuzzy C-means clustering and the fuzzy decision tree | |
| CN116542380B (en) | Power plant supply chain carbon footprint optimization method and device based on natural language | |
| Dong et al. | A multi-stage risk-adjusted control chart for monitoring and early-warningof products sold with two-dimensional warranty | |
| Gupta et al. | An Overview of Predictive Maintenance and Load Forecasting | |
| Liu et al. | Application of data mining in industrial engineering |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication |