A monitoring method and system based on electric power energy management system
Through time-domain reflectometer and non-invasive load monitoring technology combined with machine learning algorithms, the problem of low efficiency in load node position recognition in power energy systems is solved, and higher recognition accuracy and real-time monitoring of system status is achieved, and power is enhanced. The risk resistance of the energy system.
Patent Information
- Application Number
- CN202411087125.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-08
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-08-08
AI Technical Summary
In power energy systems, traditional methods are used to determine the location of load nodes inefficiently and prone to errors, and are susceptible to noise interference when processing complex load data, resulting in low recognition accuracy.
Time domain reflectometer and non-invasive load monitoring technology are used to determine the location of the load node, analyze the status of the power energy system through machine learning algorithms, and optimize and adjust the power energy system using a multi-layer perceptron neural network to build a user interactive interface to display power data in real time.
It improves the accuracy and flexibility of load identification, improves the correlation and accuracy of load characteristics, and enhances the risk resistance of the power and energy system.
Smart Images

Figure CN118611274B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power system monitoring, and in particular to a monitoring method and system based on an electric power energy management system. Background Art
[0002] With the continuous growth of global energy demand and the increasing application of renewable energy, power energy management systems have gradually become an important tool for optimizing energy use and improving energy efficiency. In recent years, the development of smart grid and Internet of Things technologies has injected new vitality into power energy management systems, enabling them to monitor and manage electricity more intelligently, automatically and efficiently. Traditional power management methods mainly rely on wired monitoring and manual control, which lack real-time and accuracy and are difficult to meet the needs of modern energy management. With the introduction of machine learning algorithms, new solutions have been provided for power energy management systems, but the existing technologies still have shortcomings. In determining the location of load nodes in power energy systems, traditional methods are inefficient and prone to errors. When processing complex load data, they are easily interfered by noise, resulting in low recognition accuracy. Summary of the invention
[0003] In view of the problems existing in the above-mentioned existing monitoring methods and systems based on electric power energy management systems, the present invention is proposed.
[0004] Therefore, the problem to be solved by the present invention is that the traditional method for determining the location of load nodes in the electric energy system is inefficient and prone to errors. When processing complex load data, it is easily interfered by noise, resulting in low recognition accuracy.
[0005] To solve the above technical problems, the present invention provides the following technical solutions: a monitoring method based on an electric power energy management system, which includes collecting power data and determining the location of load nodes in the electric power energy system through a time domain reflectometer; using non-invasive load monitoring technology to monitor load nodes and extract load characteristics; constructing a machine learning algorithm based on load characteristics to analyze the status of the electric power system for early warning, and optimizing and adjusting the electric power system based on a multi-layer perceptron neural network; and constructing a user interaction interface to display power data in real time.
[0006] As a preferred embodiment of the monitoring method based on the electric energy management system described in the present invention, wherein: the collecting of electric power data and determining the location of load nodes in the electric power system by time domain reflectometry refers to deploying sensors in the electric power system to collect electric power data and collect power line parameters in the electric power system, and calculating the first proportional factor k and the first adjustment factor k according to the power line parameters. , :
[0007] ;
[0008] Where s is the length of the power line, b is the total width of the power line;
[0009] Calculate the second scale factor k1 and the second adjustment factor :
[0010] ;
[0011] Where h is the thickness of the power line;
[0012] Calculate the first scaling factor k and the first adjustment factor k , , the second scale factor k1 and the second adjustment factor Elliptic integral of the first kind:
[0013] ;
[0014] Where x is the input data, including the first scale factor k and the first adjustment factor k , and the second scaling factor k1 and the second adjustment factor , is the integral variable, and its region ranges from 0 to π / 2;
[0015] Calculation of effective dielectric constant of power lines based on elliptic integral of the first kind :
[0016] ;
[0017] in is the relative dielectric constant of the power line material;
[0018] According to the effective dielectric constant Calculate the characteristic impedance Z0 of the power line:
[0019] ;
[0020] In the no-load state of the electric energy system, a TDR device is used to inject a high-frequency pulse signal into the power line as the incident voltage signal, and the reflected voltage signal is collected. The ratio of the reflected voltage signal to the incident voltage signal is used as the no-load reflection coefficient. The no-load impedance of the electric energy system is calculated based on the no-load reflection coefficient:
[0021] ;
[0022] Where Z r is the no-load impedance, is the no-load reflection coefficient;
[0023] Repeat the operation to obtain the load impedance Z of the power energy system when the power energy system is connected to the load y, and calculate the impedance variation coefficient S based on the load impedance and no-load impedance:
[0024] ;
[0025] In the electric power energy system, power lines at different locations are selected and the impedance variation coefficient is calculated using TDR equipment. The location of each test and the time difference t between the incident voltage signal and the reflected voltage signal are recorded. The impedance variation coefficients of all locations are formed into a line graph according to the distance from small to large. The peak in the line graph is identified, and the test location of the impedance variation coefficient at the peak location is extracted. The distance between the load node and the test location is calculated based on the time difference t and the signal speed to further determine the load node location.
[0026] As a preferred embodiment of the monitoring method based on the power energy management system of the present invention, wherein: the use of non-intrusive load monitoring technology to monitor the load node and extract the load characteristics refers to determining the position of the load node and collecting the power data of the load node in real time through the sensor to obtain the real-time sampled current I k and voltage V k , calculate the current I k and voltage V k The Pearson correlation coefficient of and the scaling factor are calculated:
[0027] ;
[0028] in is the scaling factor, cor(I k , V k ) is the current I k and voltage V k Pearson correlation coefficient of
[0029] Construct an adaptive scaling recursive model A according to the scaling factor:
[0030] ;
[0031] Among them I k (i) and I k (j) are the i-th and k-th currents respectively, and L is the total number of current samples;
[0032] The adaptive scaling recursive graph is obtained through the adaptive scaling recursive model A, and the deep visual processing network Swin-Transformer is used to identify the adaptive scaling recursive graph, and the load characteristics of the load nodes therein are extracted, including the average power consumption, average running time and number of daily operations.
[0033] As a preferred solution of the monitoring method based on the electric energy management system described in the present invention, wherein: the construction of a machine learning algorithm based on load characteristics to analyze the state of the electric energy system for early warning refers to the construction of a convolutional neural network, and the input format is defined as a load characteristic format;
[0034] Collect historical power data in the power energy system to extract historical load characteristics, input the historical load characteristics into the convolutional neural network as a training set for training, and optimize the convolutional neural network parameters;
[0035] The load characteristics extracted from the real-time data of the load nodes are input into the optimized convolutional neural network to output the power energy system status, including normal and abnormal;
[0036] If the power energy system status is normal, then keep monitoring;
[0037] If the power energy system status is abnormal, an early warning will be issued.
[0038] As a preferred solution of the monitoring method based on the electric energy management system of the present invention, wherein: the optimization and adjustment of the electric energy system based on the multi-layer perceptron neural network refers to constructing a multi-layer perceptron neural network,
[0039] Collect historical load characteristics of all load nodes in the power energy network to train the multi-layer perceptron neural network. The optimization goal is to minimize the operating energy consumption. The parameters of the multi-layer perceptron neural network are optimized based on forward propagation, back propagation and minimization of loss function.
[0040] The load characteristics of all load nodes in the electric energy network are collected and input into the trained multi-layer perceptron neural network, the optimal operating time and optimal operating load of each load node are output, and the electric energy system is adjusted based on the optimal operating time and optimal operating load.
[0041] As a preferred solution of the monitoring method based on the electric power energy management system described in the present invention, the sensors include current sensors, voltage sensors, energy consumption sensors and GPS sensors, and the current sensors, voltage sensors, energy consumption sensors and GPS sensors are each assigned into a group and deployed equidistantly in the electric power energy system.
[0042] As a preferred scheme of the monitoring method based on the electric power energy management system described in the present invention, wherein: the construction of a user interactive interface to display power data in real time refers to constructing a user interface to display the power data collected by the sensor, adding interactive functions to the user interface, allowing the user to customize the power data display content, and highlighting the load nodes in the electric power system and synchronously displaying them in the user interface.
[0043] Another object of the present invention is to provide a monitoring system based on an electric power energy management system, which comprises a data collection module for deploying sensors in the electric power energy system to collect power data;
[0044] A load identification module, used to identify load nodes and calculate load node locations based on power data;
[0045] A feature extraction module, used for extracting load features according to power data of load nodes;
[0046] The early warning optimization module is used to analyze and warn the status of the power energy system according to the load characteristics, and simultaneously optimize and adjust the power energy system;
[0047] The display module is used to display power data and load nodes to users.
[0048] A computer device comprises: a memory and a processor; the memory stores a computer program, and the processor implements the steps of the above-mentioned monitoring method based on the electric power energy management system when executing the computer program.
[0049] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the monitoring method based on the electric power energy management system.
[0050] The beneficial effects of the present invention are as follows: the present invention collects power data and calculates the positions of load nodes in the electric energy system based on the time domain reflectometer, thereby improving the accuracy and flexibility of load identification, and uses recursive graphs and deep visual processing networks to extract load characteristics of load nodes, thereby improving the relevance and accuracy of the characteristics, and simultaneously performs early warning and adjustment of the electric energy system based on the load characteristics, thereby enhancing the risk resistance of the electric energy system. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.
[0052] Figure 1 It is a flow chart of a monitoring method based on an electric power energy management system.
[0053] Figure 2 It is a structural diagram of the monitoring system based on the power energy management system. DETAILED DESCRIPTION
[0054] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.
[0055] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.
[0056] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments. Example
[0057] Reference Figure 1 , which is the first embodiment of the present invention, and which provides a monitoring method based on an electric power energy management system. The monitoring method based on an electric power energy management system includes:
[0058] S1, collect power data and determine the location of load nodes in the power energy system through time domain reflectometry;
[0059] Specifically, collecting power data and determining the location of load nodes in the power energy system through a time domain reflectometer refers to deploying sensors in the power energy system to collect power data and collect power line parameters in the power energy system, and calculating the first proportionality factor k and the first adjustment factor k according to the power line parameters. , :
[0060] ;
[0061] Where s is the length of the power line, b is the total width of the power line;
[0062] Calculate the second scale factor k1 and the second adjustment factor :
[0063] ;
[0064] Where h is the thickness of the power line;
[0065] Calculate the first scale factor k and the first adjustment factor k , , the second scale factor k1 and the second adjustment factor Elliptic integral of the first kind:
[0066] ;
[0067] Where x is the input data, including the first scale factor k and the first adjustment factor k , and the second scaling factor k1 and the second adjustment factor , is the integral variable, and its region ranges from 0 to π / 2;
[0068] Calculation of effective dielectric constant of power lines based on elliptic integral of the first kind :
[0069] ;
[0070] in is the relative dielectric constant of the power line material;
[0071] According to the effective dielectric constant Calculate the characteristic impedance Z0 of the power line:
[0072] ;
[0073] In the no-load state of the electric energy system, a TDR device is used to inject a high-frequency pulse signal into the power line as the incident voltage signal, and the reflected voltage signal is collected. The ratio of the reflected voltage signal to the incident voltage signal is used as the no-load reflection coefficient. The no-load impedance of the electric energy system is calculated based on the no-load reflection coefficient:
[0074] ;
[0075] Where Z r is the no-load impedance, is the no-load reflection coefficient;
[0076] Repeat the operation to obtain the load impedance Z of the power energy system when the power energy system is connected to the load y , and calculate the impedance variation coefficient S based on the load impedance and no-load impedance:
[0077] ;
[0078] In the electric energy system, power lines at different locations are selected and the impedance variation coefficient is calculated using TDR equipment. The location of each test and the time difference t between the incident voltage signal and the reflected voltage signal are recorded. The impedance variation coefficients of all locations are formed into a line graph according to the distance from small to large. The peak in the line graph is identified, and the test location of the impedance variation coefficient at the peak location is extracted. The distance between the load node and the test location is calculated based on the time difference t and the signal speed to further determine the load node location.
[0079] By deploying sensors in the electric energy system to collect power data and power line parameters in real time, the real-time and accuracy of the data is ensured, providing a reliable basis for subsequent calculations. By calculating the first proportional factor (k and k') and the second proportional factor (k1 and k1'), and performing the first-class elliptic integral calculation, the effective dielectric constant of the power line can be accurately determined. This step greatly improves the calculation accuracy of the electrical characteristics of the power line. In the no-load and load states, the TDR device is used to inject high-frequency pulse signals and measure the reflected signals, so that the no-load impedance and load impedance of the power line can be accurately obtained, thereby calculating the impedance variation coefficient. By analyzing the reflected signals, the discontinuities and load nodes in the power line can be accurately identified, providing a basis for the subsequent load position. The impedance variation coefficient reflects the impedance variation of the power line under different load conditions. By measuring at different positions and forming a line graph of the impedance variation coefficient, the load node position can be intuitively identified. The peak position in the line graph corresponds to the load node position. By calculating the time difference and signal speed, the distance between the load node and the test position can be accurately determined, thereby improving the accuracy and flexibility of load node identification. By comprehensively analyzing the impedance variation coefficient and the time difference of the reflected signal, the position of the load node can be accurately calculated, avoiding the errors and inefficiencies in the traditional method. Accurately determining the load node position is of great significance for optimizing the energy efficiency management of the power system, improving the system's risk resistance and realizing intelligent regulation.
[0080] It should also be noted that the sensors include current sensors, voltage sensors, energy consumption sensors and GPS sensors. The current sensors, voltage sensors, energy consumption sensors and GPS sensors are each allocated into groups and deployed equidistantly in the electric power system.
[0081] The equidistant deployment of sensor groups in the electric energy system ensures the comprehensiveness and uniformity of data collection and avoids blind spots and deviations in data collection. The equidistant deployment enables the system to evenly cover the entire monitoring area and improves the representativeness of the data. The current sensor and voltage sensor collect the current and voltage data in the power line in real time and provide high-precision instantaneous power parameters. These data can be used to calculate instantaneous power, load conditions and power quality. The energy consumption data provided by the energy consumption sensor, including active power, reactive power and total energy consumption, can comprehensively evaluate the energy efficiency of the system. Through these data, energy consumption analysis and energy allocation optimization can be carried out to reduce energy waste. The geographic location information provided by the GPS sensor ensures the precise deployment of the sensor group and the geographic tagging of the data. The geographic information helps to locate the fault point, optimize the sensor layout, and improve the monitoring accuracy and response speed of the system.
[0082] S2, use non-intrusive load monitoring technology to monitor load nodes and extract load characteristics;
[0083] Specifically, the non-intrusive load monitoring technology is used to monitor the load node and extract the load characteristics, which means that after determining the location of the load node, the power data of the load node is collected in real time by the sensor to obtain the real-time sampled current I k and voltage V k , construct the recursive model R:
[0084] ;
[0085] Where L is the total number of current samples, I k (i) and I k (j) are the i-th and k-th currents, respectively, is the recursive distance threshold, is the Heaviside function, which has a value of 1 when x is greater than or equal to 0, and 0 otherwise;
[0086] To make the features more obvious, use the unthresholded recursive model T:
[0087] ;
[0088] Introducing a scaling factor The scaled recursive model S is obtained by exponentially scaling the threshold-free recursive model T:
[0089] ;
[0090] Finally, the adaptive scaling recursive model is obtained by adaptively adjusting the scaling coefficient and combining the phase difference information between current and voltage.
[0091] Calculate the current I k and voltage V k The Pearson correlation coefficient of and the scaling factor are calculated:
[0092] ;
[0093] in is the scaling factor, cor(I k , V k ) is the current I k and voltage V k Pearson correlation coefficient of
[0094] Construct an adaptive scaling recursive model A according to the scaling factor:
[0095] ;
[0096] Among them I k (i) and I k (j) are the i-th and k-th currents respectively, and L is the total number of current samples;
[0097] The adaptive scaling recursive graph is obtained through the adaptive scaling recursive model A, and the deep visual processing network Swin-Transformer is used to identify the adaptive scaling recursive graph, and the load characteristics of the load nodes therein are extracted, including the average power consumption, average running time and number of daily operations.
[0098] The current and voltage data of the load nodes are collected in real time by sensors, ensuring the timeliness and accuracy of the data. These data are the basis for subsequent feature extraction and analysis. Real-time data collection is the core of the monitoring system, which can reflect the instantaneous operating status of the load nodes and provide a reliable basis for system optimization. The recursive graph reveals the dynamic characteristics of the system by showing the recursive behavior of the current and voltage data points in the phase space. The recursive graph makes the characteristics of the load node clearer. The threshold-free recursive graph eliminates the influence of the threshold and makes the characteristics more obvious. The scaled recursive graph further emphasizes the characteristics of the data through exponential scaling and improves the accuracy of feature extraction. By adaptively adjusting the scaling factor and combining the phase difference information between current and voltage, the characteristics of the power data are dynamically reflected. The adaptive scaling recursive graph improves the accuracy of load feature extraction. The Pearson correlation coefficient measures the linear correlation between current and voltage. The scaling factor is used to adjust the scaling ratio of the recursive graph to make feature extraction more targeted. Through the calculation of the correlation coefficient and the scaling factor, the feature extraction of the recursive graph can be dynamically adjusted to further improve the accuracy and robustness of feature extraction. Swin-Transformer uses its multi-level window self-attention mechanism to effectively extract load features in the adaptive scaling recursive graph, including average power consumption, average operating time, and number of daily operations. These features reflect the operating status and energy consumption pattern of the load node.
[0099] S3. Build a machine learning algorithm based on load characteristics to analyze the status of the power energy system for early warning, and optimize and adjust the power energy system based on a multi-layer perceptron neural network;
[0100] Specifically, constructing a machine learning algorithm to analyze the state of the electric energy system for early warning based on load characteristics refers to constructing a convolutional neural network and defining the input format as a load feature format;
[0101] Collect historical power data in the power energy system to extract historical load characteristics, input the historical load characteristics into the convolutional neural network as a training set for training, and optimize the convolutional neural network parameters;
[0102] The load characteristics extracted from the real-time data of the load nodes are input into the optimized convolutional neural network to output the power energy system status, including normal and abnormal;
[0103] If the power energy system status is normal, then keep monitoring;
[0104] If the power energy system status is abnormal, an early warning will be issued.
[0105] The convolutional neural network extracts local features of the data through convolution operations, and gradually extracts higher-level features through a multi-layer network structure, thereby improving the ability to process complex data. The input format is defined as a load feature format to ensure that the model can accurately receive and process power data. By collecting and analyzing historical power data, historical load features are extracted to form a rich training data set. These data reflect the operating characteristics of the power system under different states and provide sufficient samples for model training. By inputting historical load features into the convolutional neural network for training, the network parameters are continuously optimized so that the model can accurately identify the operating state of the power system. The training process includes forward propagation and back propagation. The parameters are adjusted by minimizing the loss function to improve the prediction accuracy of the model. The load features extracted from the real-time data of the load node are input into the optimized convolutional neural network to analyze the system state in real time. The model outputs the power energy system state according to the real-time data, including normal and abnormal, to ensure the immediate response of the system. If the power energy system state is normal, monitoring is maintained to ensure that the system operates within the normal range. If the system state is abnormal, an early warning is immediately issued to prompt maintenance personnel to take necessary measures to avoid the expansion of faults and increased losses.
[0106] Furthermore, optimizing and adjusting the electric energy system based on a multi-layer perceptron neural network refers to constructing a multi-layer perceptron neural network.
[0107] Collect historical load characteristics of all load nodes in the power energy network to train the multi-layer perceptron neural network. The optimization goal is to minimize the operating energy consumption. The parameters of the multi-layer perceptron neural network are optimized based on forward propagation, back propagation and minimization of loss function.
[0108] The load characteristics of all load nodes in the electric energy network are collected and input into the trained multi-layer perceptron neural network, the optimal operating time and optimal operating load of each load node are output, and the electric energy system is adjusted based on the optimal operating time and optimal operating load.
[0109] An MLP neural network is constructed. By setting the input layer, hidden layer and output layer, it is ensured that the network can handle complex nonlinear relationships. The input layer receives load characteristic data, the hidden layer extracts the deep characteristics of the data through the activation function, and the output layer generates the optimal operating time and the optimal operating load. The historical load characteristics of all load nodes in the power energy network are collected to form a training data set. These data reflect the load characteristics of the system under different operating conditions and provide rich samples for model training. The MLP neural network is trained based on the optimization method of forward propagation, back propagation and minimization of loss function. The forward propagation calculates the output result, the back propagation calculates the gradient and updates the network parameters, so that the loss function value is gradually reduced, and the real-time load characteristics of all load nodes in the power energy network are input into the trained MLP neural network. The network outputs the optimal operating time and optimal operating load of each load node according to the real-time data.
[0110] S4. Build a user interactive interface to display power data in real time;
[0111] Specifically, constructing a user interactive interface to display power data in real time means constructing a user interface to display power data collected by sensors, adding interactive functions to the user interface, allowing users to customize the power data display content, and synchronously displaying the highlighted labels of load nodes in the power energy system in the user interface.
[0112] By building a user interaction interface and providing an intuitive graphical display method, users can easily view and understand power data. The interactive interface supports multiple display forms, including charts, curves, maps, etc., which improves the diversity and flexibility of data display and realizes real-time display of power data, ensuring that users can see the latest data changes and system status at the first time. This is of great significance for timely discovery and handling of abnormal situations and optimization of system operation. Adding interactive functions to the user interface allows users to interact with the interface through operations such as clicking, dragging, and zooming. Users can customize the display content according to actual needs and choose different data views and display methods, which improves the flexibility and usability of the interface. Example
[0113] Reference Figure 2 , which is the second embodiment of the present invention, and which is different from the previous embodiment, provides a monitoring system based on an electric power energy management system, which includes:
[0114] A data collection module, used to deploy sensors in the electric energy system to collect electric power data;
[0115] A load identification module, used to identify load nodes and calculate load node locations based on power data;
[0116] A feature extraction module, used for extracting load features according to power data of load nodes;
[0117] The early warning optimization module is used to analyze and warn the status of the power energy system according to the load characteristics, and simultaneously optimize and adjust the power energy system;
[0118] The display module is used to display power data and load nodes to users.
[0119] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.
[0120] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.
[0121] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.
[0122] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit with a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit with a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
[0123] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.
Claims
1. A monitoring method based on an electric power energy management system, characterized in that: include, Collect power data and determine the location of load nodes in the power energy system through time domain reflectometry; Use non-intrusive load monitoring technology to monitor load nodes and extract load characteristics; Build a machine learning algorithm based on load characteristics to analyze the status of the power energy system for early warning, and optimize and adjust the power energy system based on a multi-layer perceptron neural network; Build a user interactive interface to display power data in real time; The collecting of power data and determining the location of the load node in the power energy system by time domain reflectometry refers to deploying sensors in the power energy system to collect power data and collect power line parameters in the power energy system, and calculating the first proportionality factor k and the first adjustment factor according to the power line parameters. : ; Where s is the length of the power line, b is the total width of the power line; Calculate the second scale factor k1 and the second adjustment factor ; ; Where h is the thickness of the power line; Calculate the first scaling factor k and the first adjustment factor k , , the second scale factor k1 and the second adjustment factor Elliptic integral of the first kind: ; Where x is the input data, including the first scale factor k and the first adjustment factor k , and the second scaling factor k1 and the second adjustment factor , is the integral variable, and its region ranges from 0 to π / 2; Calculation of effective dielectric constant of power lines based on elliptic integral of the first kind : ; According to the effective dielectric constant Calculate the characteristic impedance Z0 of the power line: ; In the no-load state of the electric energy system, a TDR device is used to inject a high-frequency pulse signal into the power line as the incident voltage signal, and the reflected voltage signal is collected. The ratio of the reflected voltage signal to the incident voltage signal is used as the no-load reflection coefficient. The no-load impedance of the electric energy system is calculated based on the no-load reflection coefficient: ; Where Z r is the no-load impedance, is the no-load reflection coefficient; Repeat the operation to obtain the load impedance Z of the power energy system when the power energy system is connected to the load y , and calculate the impedance variation coefficient S based on the load impedance and no-load impedance: ; Select power lines at different locations in the power energy system and use TDR equipment to calculate the impedance variation coefficient, record the location of each test and the time difference t between the incident voltage signal and the reflected voltage signal, form a line graph of the impedance variation coefficients of all locations from small to large distances, identify the peak in the line graph, extract the test location of the impedance variation coefficient at the peak location, and calculate the distance between the load node and the test location based on the time difference t and the signal speed, and further determine the load node location; The use of non-intrusive load monitoring technology to monitor load nodes and extract load characteristics refers to determining the location of the load node and collecting the power data of the load node in real time through the sensor to obtain the real-time sampled current I k and voltage V k , calculate the current I k and voltage V k The Pearson correlation coefficient of and the scaling factor are calculated: ; in is the scaling factor, cor(I k , V k ) is the current I k and voltage V k Pearson correlation coefficient of Construct an adaptive scaling recursive model A according to the scaling factor: ; Among them I k (i) and I k (j) are the i-th and k-th currents respectively, and L is the total number of current samples; The adaptive scaling recursive graph is obtained through the adaptive scaling recursive model A, and the adaptive scaling recursive graph is recognized using the deep visual processing network Swin-Transformer to extract the load characteristics of the load nodes, including the average power consumption, average running time, and the number of daily operations. The constructing of a machine learning algorithm according to load characteristics to analyze the state of the electric energy system for early warning refers to constructing a convolutional neural network and defining the input format as a load characteristic format; Collect historical power data in the power energy system to extract historical load characteristics, input the historical load characteristics into the convolutional neural network as a training set for training, and optimize the convolutional neural network parameters; The load characteristics extracted from the real-time data of the load nodes are input into the optimized convolutional neural network to output the power energy system status, including normal and abnormal; If the power energy system status is normal, then keep monitoring; If the power energy system status is abnormal, an early warning will be issued; The optimization and adjustment of the electric energy system based on the multi-layer perceptron neural network refers to constructing a multi-layer perceptron neural network, Collect historical load characteristics of all load nodes in the power energy network to train the multi-layer perceptron neural network. The optimization goal is to minimize the operating energy consumption. The parameters of the multi-layer perceptron neural network are optimized based on forward propagation, back propagation and minimization of loss function. The load characteristics of all load nodes in the electric energy network are collected and input into the trained multi-layer perceptron neural network, the optimal operating time and optimal operating load of each load node are output, and the electric energy system is adjusted based on the optimal operating time and optimal operating load.
2. The monitoring method based on the electric energy management system according to claim 1, characterized in that: The sensors include current sensors, voltage sensors, energy consumption sensors and GPS sensors. The current sensors, voltage sensors, energy consumption sensors and GPS sensors are each allocated into a group and deployed equidistantly in the electric power system.
3. The monitoring method based on the electric power energy management system according to claim 2, characterized in that: Constructing a user interactive interface to display power data in real time refers to constructing a user interface to display the power data collected by sensors, adding interactive functions to the user interface, allowing users to customize the power data display content, and synchronously displaying the load nodes in the power energy system with highlighted labels in the user interface.
4. A monitoring system based on an electric energy management system and a monitoring method based on an electric energy management system according to any one of claims 1 to 3, characterized in that: include, A data collection module, used to deploy sensors in the electric energy system to collect electric power data; A load identification module, used to identify load nodes and calculate load node locations based on power data; A feature extraction module, used for extracting load features according to power data of load nodes; The early warning optimization module is used to analyze and warn the status of the power energy system according to the load characteristics, and simultaneously optimize and adjust the power energy system; The display module is used to display power data and load nodes to users.
5. A computer device comprising: Memory and processor; The memory stores a computer program, characterized in that: when the processor executes the computer program, the steps of the monitoring method based on the electric power energy management system described in any one of claims 1 to 3 are implemented.
6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the monitoring method based on the electric power energy management system described in any one of claims 1 to 3 are implemented.
Citation Information
Patent Citations
Distributed cable state detection system based on TDR
CN103487722A
Energy monitoring management system for grid system
CN116937575A
Determining the installation location and orientation of ultrasonic sensors using neural networks
DE102019119585A1
System and method to measure and control power consumption in a residential or commercial building via a wall socket to ensure optimum energy usage therein
US20120265586A1
Ac power over ethernet
US20150115741A1
