Power grid system monitoring method and system based on Internet of Things, and electronic equipment

Abstract

The invention discloses a power grid system monitoring method and system based on the Internet of Things, and electronic equipment, and belongs to the technical field of the Internet of Things. The method comprises the following steps: acquiring current data, temperature data and user emotion data of a power grid system, and determining a current regression coefficient, a temperature regression coefficient and a user emotion data regression coefficient; calculating a power grid state prediction observation value; calculating a power grid state actual observation value; and acquiring a preset power grid state observation threshold, and if the actual observation value exceeds the preset power grid state observation threshold, sending alarm information to a control center. According to the scheme, by obtaining the real-time current, temperature and user emotion data, the power grid state can be dynamically monitored, and the capability of responding to potential problems is improved. And a regression coefficient and a prediction formula are used, so that the power grid state can be predicted more accurately, and managers are helped to make timely decisions. Potential risks are found and processed in time, the reliability and safety of a power grid system can be enhanced, and the continuity of power supply is guaranteed.

Description

Technical Field

[0001] The present application belongs to the technical field of Internet of Things, and specifically relates to a power grid system monitoring method, system and electronic equipment based on Internet of Things. Background Art

[0002] With the rapid development of the power industry and the continuous expansion of the power grid, the power grid system has become one of the most important infrastructures in modern society. However, the complexity, interconnection and dynamics of the power grid system have also brought many challenges to its safe and stable operation. In order to ensure the reliable power supply of the power grid system, optimize resource allocation and reduce failure losses, it is particularly important to monitor the power grid system.

[0003] When monitoring the power grid system today, maintenance personnel need to regularly inspect the power grid equipment and record the equipment's operating status and parameters. At the same time, they also need to use professional monitoring equipment and systems to view key data such as the grid's current, voltage, and temperature in real time. Once data anomalies or equipment failures are discovered, maintenance personnel need to take immediate measures, such as adjusting the operating mode and removing faulty equipment, to ensure the safe and stable operation of the power grid.

[0004] However, manual monitoring requires maintenance personnel to regularly inspect power grid equipment, record data, and manually analyze and judge. This process is time-consuming, labor-intensive, and inefficient, and it is difficult to adapt to the monitoring needs of large-scale power grid systems. In addition, during manual monitoring, maintenance personnel may record data incorrectly or miss data due to fatigue, negligence, etc., which affects the accurate judgment of the power grid status. Summary of the invention

[0005] The embodiments of the present application provide a power grid system monitoring method, system and electronic device based on the Internet of Things, which solves the problem that manual monitoring of the power grid system in the prior art requires operation and maintenance personnel to regularly patrol the power grid equipment, record data, and manually analyze and judge. This process is time-consuming and labor-intensive, inefficient, and difficult to adapt to the monitoring needs of large-scale power grid systems. In addition, during the manual monitoring process, the operation and maintenance personnel may cause data recording errors or omissions due to fatigue, negligence, etc., thereby affecting the problem of accurate judgment of the power grid status.

[0006] In a first aspect, an embodiment of the present application provides a method for monitoring a power grid system based on the Internet of Things, the method comprising: Acquire current data, temperature data and user emotion data of the power grid system, input the current data, temperature data and user emotion data into a preset regression coefficient determination model, and obtain a current regression coefficient, a temperature regression coefficient and a user emotion data regression coefficient; Obtain a preset grid state prediction formula, and calculate a grid state prediction observation value according to the current data, temperature data, user emotion data, current regression coefficient, temperature regression coefficient, user emotion data regression coefficient, and the preset grid state prediction formula; Determine a multidimensional vector observation value according to the current data, temperature data and user emotion data, obtain a preset power grid state update formula, and calculate the actual power grid state observation value according to the multidimensional vector observation value, the power grid state prediction observation value and the preset power grid state update formula; wherein the preset power grid state update formula is: ; in, is the actual observed value of the power grid status; is the pre-calculated Kalman gain; Predict observations for power grid status; is the multidimensional vector observation value; H is the pre-measured observation matrix; A preset power grid state observation threshold is obtained, and if the actual observation value exceeds the preset power grid state observation threshold, an alarm message is sent to a control center.

[0007] Furthermore, the preset grid state prediction formula is: ; in, Predict observations for power grid status; is a preset bias item; is the current regression coefficient; is the current data; is the temperature regression coefficient; is the temperature data; is the regression coefficient of user sentiment data; User sentiment data.

[0008] Further, after obtaining the preset grid state observation threshold, if the actual observation value exceeds the preset grid state observation threshold, after sending an alarm message to the control center, the method further includes: Inputting the current data, temperature data and user emotion data into a preset load prediction model to obtain predicted load data of the power grid; The real-time output power data, minimum output power, maximum output power and preset dispatching power calculation formula of each distributed power source of the power grid are obtained, and the dispatching power of each distributed power source is determined according to the real-time output power data, minimum output power, maximum output power and preset dispatching power calculation formula; wherein the preset dispatching power calculation formula is: in, is the dispatching power of each distributed power source; is the minimum output power of each distributed power source; is the maximum output power of each distributed power source; To predict load data; is the total output power of other distributed power sources j, which represents the sum of the real-time output powers of all other power sources different from distributed power source i.

[0009] Further, after obtaining the preset grid state observation threshold, if the actual observation value exceeds the preset grid state observation threshold, after sending an alarm message to the control center, the method further includes: Acquire voltage data, frequency data, and device status data of the power grid system, input the current data, temperature data, voltage data, frequency data, and device status data into a preset fault detection model to obtain a fault score, and determine whether a power grid fault exists according to the fault score and a preset fault threshold; If the fault score exceeds the preset fault threshold, it is determined that a power grid fault exists, and the fault cause is determined by a preset fault detection model, and the fault cause is sent to a control center.

[0010] Furthermore, the training process of the preset regression coefficient determination model includes: Obtaining first historical current data, first historical temperature data, historical user emotion data, historical current regression coefficient, historical temperature regression coefficient, and historical user emotion data regression coefficient; labeling a current regression coefficient label of the first historical current data according to the historical current regression coefficient, labeling a temperature regression coefficient label of the first historical temperature data according to the historical temperature regression coefficient, and labeling a user emotion data regression coefficient label of the historical user emotion data according to the historical user emotion data regression coefficient; Creating a first data set based on the first historical current data and the current regression coefficient label, creating a second data set based on the first historical temperature data and the temperature regression coefficient label, and creating a third data set based on the historical user emotion data and the user emotion data regression coefficient label; A regression coefficient determination model is constructed, and the regression coefficient determination model is trained according to the first data set, the second data set, and the third data set until the regression coefficient determination model reaches a preset regression coefficient determination model training standard.

[0011] Furthermore, the training process of the preset load forecasting model includes: Obtaining historical grid loads corresponding to first historical current data, first historical temperature data, and historical user emotion data; Creating a fourth data set according to the first historical current data, the first historical temperature data, and the historical user emotion data, and labeling a grid load label of the fourth data set according to the historical grid load; A load prediction model is constructed, and the load prediction model is trained according to the fourth data set and the power grid load label until the load prediction model reaches a preset load prediction model training standard.

[0012] Furthermore, the training process of the preset fault detection model includes: Acquire first historical voltage data, first historical frequency data, and first historical device status data, and acquire historical fault scores corresponding to the first historical current data, first historical temperature data, first historical voltage data, first historical frequency data, and first historical device status data; Creating a fifth data set according to the first historical current data, the first historical temperature data, the first historical voltage data, the first historical frequency data, and the first historical device status data, and labeling a fault score label of the fifth data set according to the historical fault score; Obtain historical fault records, and determine historical fault causes and operating parameter sets corresponding to each historical fault cause according to the historical fault records; wherein the operating parameter set includes second historical current data, second historical temperature data, second historical voltage data, second historical frequency data, and second historical device status data; Marking a fault cause label of the operating parameter set according to the historical fault cause; A fault detection model is constructed, and the fault detection model is trained according to the fifth data set, the fault score label, the operating parameter set, and the fault cause label until the fault detection model reaches a preset fault detection model training standard.

[0013] In a second aspect, an embodiment of the present application provides a power grid system monitoring system based on the Internet of Things, the system comprising: The acquisition module is used to obtain the current data, temperature data and user emotion data of the power grid system, and input the current data, temperature data and user emotion data into a preset regression coefficient determination model to obtain the current regression coefficient, temperature regression coefficient and user emotion data regression coefficient; A prediction module, used to obtain a preset grid state prediction formula, and calculate a grid state prediction observation value according to the current data, temperature data, user emotion data, current regression coefficient, temperature regression coefficient, user emotion data regression coefficient and the preset grid state prediction formula; A state update module is used to determine a multidimensional vector observation value according to the current data, temperature data and user emotion data, obtain a preset power grid state update formula, and calculate the actual power grid state observation value according to the multidimensional vector observation value, the power grid state prediction observation value and the preset power grid state update formula; wherein the preset power grid state update formula is: ; in, is the actual observed value of the power grid status; is the Kalman gain; Predict observations for power grid status; is the multidimensional vector observation value; H is the pre-measured observation matrix; The alarm module is used to obtain a preset grid state observation threshold, and if the actual observation value exceeds the preset grid state observation threshold, an alarm message is sent to the control center.

[0014] In a third aspect, an embodiment of the present application provides an electronic device, which includes a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the method described in the first aspect.

[0015] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0016] In an embodiment of the present application, current data, temperature data and user emotion data of the power grid system are obtained, and the current data, temperature data and user emotion data are input into a preset regression coefficient determination model to obtain a current regression coefficient, a temperature regression coefficient and a user emotion data regression coefficient; a preset power grid state prediction formula is obtained, and a power grid state prediction observation value is calculated according to the current data, temperature data, user emotion data, current regression coefficient, temperature regression coefficient, user emotion data regression coefficient and a preset power grid state prediction formula; a multidimensional vector observation value is determined according to the current data, temperature data and user emotion data, and a preset power grid state update formula is obtained, and an actual power grid state observation value is calculated according to the multidimensional vector observation value, the power grid state prediction observation value and the preset power grid state update formula; wherein the preset power grid state update formula is: ; in, is the actual observed value of the power grid status; is the pre-calculated Kalman gain; Predict observations for power grid status; is a multidimensional vector observation value; H is a pre-measured observation matrix. A preset grid state observation threshold is obtained. If the actual observation value exceeds the preset grid state observation threshold, an alarm message is sent to the control center. Through the above-mentioned Internet of Things-based grid system monitoring method, by obtaining real-time current, temperature and user emotion data, the grid state can be dynamically monitored and the response capability to potential problems can be improved. The use of regression coefficients and prediction formulas can more accurately predict the grid state and help managers make timely decisions. Timely discovery and handling of potential risks can enhance the reliability and security of the grid system and ensure the continuity of power supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flow chart of a method for monitoring a power grid system based on the Internet of Things provided in Example 1 of the present application; Figure 2 It is a flow chart of a method for monitoring a power grid system based on the Internet of Things provided in Embodiment 2 of the present application; Figure 3 It is a structural diagram of a power grid system monitoring system based on the Internet of Things provided in Example 3 of the present application; Figure 4 It is a schematic diagram of the structure of the electronic device provided in Example 4 of the present application. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical scheme and advantages of the present application clearer, the specific embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. It is understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for the convenience of description, only part of the present application is shown in the accompanying drawings, but not all of the content. Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow chart describes each operation (or step) as a sequential process, many of the operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but it can also have additional steps not included in the accompanying drawings. The process can correspond to a method, a function, a procedure, a subroutine, a subprogram, etc.

[0019] The following will be combined with the drawings in the embodiments of the present application to clearly describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application belong to the scope of protection of this application.

[0020] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0021] In combination with the accompanying drawings, an RSMC chip, a chip multi-stage startup method and a Beidou communication navigation device provided in the embodiments of the present application are described in detail through specific embodiments and their application scenarios.

[0022] Embodiment 1 Figure 1 1 is a flow chart of a method for monitoring a power grid system based on the Internet of Things provided in the first embodiment of the present application. Figure 1 As shown, the specific steps include: S101, obtaining current data, temperature data and user emotion data of the power grid system, inputting the current data, temperature data and user emotion data into a preset regression coefficient determination model, and obtaining a current regression coefficient, a temperature regression coefficient and a user emotion data regression coefficient.

[0023] First, the usage scenario of this solution can be to obtain the operating data of the power grid system, determine whether the power grid status is out of the controllable range based on the operating data of the power grid system, and if it is out of the controllable range, send an alarm message to the control center.

[0024] Based on the above usage scenarios, it can be understood that the executor of the present application can be a power grid system monitoring system based on the Internet of Things, and no excessive limitations are made here.

[0025] In this solution, the power grid system can be the overall architecture of the power supply network, including power generation facilities, transmission lines, substations and distribution networks, etc., which are responsible for the production, transmission and distribution of electricity.

[0026] Current data can be the current intensity data measured at each node in the power grid, which reflects the state of power flow and helps monitor load conditions and identify anomalies.

[0027] Temperature data can be temperature measurement data of power grid equipment (such as transformers, switches, etc.), which is used to evaluate the operating status and overheating risk of the equipment.

[0028] User sentiment data can be information from online platforms, reflecting the public's emotional fluctuations on the power supply situation (such as satisfaction, etc.), which can be used as supplementary information for power demand forecasting.

[0029] The preset regression coefficient determination model can be a statistical model trained based on historical data to predict the state or demand of the power grid. The model generates corresponding regression coefficients by learning the relationship between input variables (such as current, temperature, user emotional data) and target variables to help quantify the impact of these inputs on the state of the power grid.

[0030] The current regression coefficient may be a coefficient reflecting the impact of current changes on grid demand prediction.

[0031] The temperature regression coefficient may be a coefficient reflecting the impact of temperature changes on grid demand prediction.

[0032] The user emotion data regression coefficient may be a coefficient reflecting the impact of user emotion data fluctuations on power grid demand prediction.

[0033] Current usage data can be collected in real time through smart meters installed at users or substations. Temperature sensors are installed at key locations of the power grid to monitor temperature in real time. Use natural language processing (NLP) tools to perform sentiment analysis on real-time comments on the network platform and extract sentiment scores, or crawl network data on related topics through APIs for sentiment analysis. The acquired current data, temperature data, and user sentiment data are organized into a unified format (such as CSV or database table), ensuring that each data point is marked with a timestamp, and then the organized data is input into the preset regression coefficient determination model. The model can be used to determine the corresponding regression coefficients based on new current data, temperature data, and user sentiment data, that is, when new data is input, the model will output predicted regression coefficients, which reflect the current impact of each factor on the power grid system or a specific variable. Specifically, it can include extracting features from the input data that are useful for model prediction. Use the model's parameters (which are learned during the training phase) to calculate the output regression coefficients. The model will output predicted regression coefficients, which reflect the current impact of each factor on the power grid system or a specific variable.

[0034] Based on the above technical solution, optionally, the training process of the preset regression coefficient determination model includes: Obtaining first historical current data, first historical temperature data, historical user emotion data, historical current regression coefficient, historical temperature regression coefficient, and historical user emotion data regression coefficient; labeling a current regression coefficient label of the first historical current data according to the historical current regression coefficient, labeling a temperature regression coefficient label of the first historical temperature data according to the historical temperature regression coefficient, and labeling a user emotion data regression coefficient label of the historical user emotion data according to the historical user emotion data regression coefficient; Creating a first data set based on the first historical current data and the current regression coefficient label, creating a second data set based on the first historical temperature data and the temperature regression coefficient label, and creating a third data set based on the historical user emotion data and the user emotion data regression coefficient label; A regression coefficient determination model is constructed, and the regression coefficient determination model is trained according to the first data set, the second data set, and the third data set until the regression coefficient determination model reaches a preset regression coefficient determination model training standard.

[0035] In this solution, the first historical current data may be the current data of the power grid recorded in a certain period of time in the past, which is used to analyze the changing trend and characteristics of the current.

[0036] The first historical temperature data may be temperature data of power grid-related equipment or environment recorded in a certain period of time in the past, and is also used to analyze the impact of temperature on power grid performance.

[0037] Historical user sentiment data can be historical data on public sentiment obtained through network analysis, which usually reflects the society's feelings and attitudes towards electricity supply, prices, etc.

[0038] The historical current regression coefficient may be a regression coefficient calculated based on historical current data, and is used to establish a mathematical relationship between current and grid load.

[0039] The historical temperature regression coefficient may be a regression coefficient calculated based on historical temperature data, and is used to establish a mathematical relationship between temperature and grid load.

[0040] The historical user emotion data regression coefficient may be a regression coefficient calculated based on the historical user emotion data, and is used to establish a relationship between the user emotion data and the power grid load.

[0041] The current regression coefficient label may be a label annotated on the first historical current data, indicating the regression coefficient corresponding to the data point.

[0042] The temperature regression coefficient label may be a label annotated on the first historical temperature data, indicating the regression coefficient corresponding to the data point.

[0043] The user emotion data regression coefficient label may be a label annotated on the historical user emotion data, indicating the regression coefficient corresponding to the data point.

[0044] The first data set may be composed of first historical current data and its corresponding current regression coefficient labels, which are used to train a model to determine the impact of current.

[0045] The second data set may be composed of the first historical temperature data and its corresponding temperature regression coefficient labels, and is used to train the model to determine the impact of temperature.

[0046] The third data set may be composed of historical user emotion data and its corresponding user emotion data regression coefficient labels, and is used to train a model to determine the impact of the user emotion data.

[0047] The preset regression coefficient determination model training standard can refer to the performance indicators set in the model training process, such as the accuracy, mean square error or other statistical indicators of the model. The training process is considered complete only when the model reaches these preset standards.

[0048] Current data for a period of time in the past can be extracted from the power grid monitoring system. This data can be obtained through sensors, smart meters, or power grid management systems. Relevant historical temperature data is collected from the environmental monitoring system of the power grid equipment. Using network analysis tools, emotional data related to power supply, price, etc. are extracted. This usually involves natural language processing technology to analyze network content. The current regression coefficient obtained by regression analysis of historical current data reflects the impact of current on load. Similarly, the historical temperature data is analyzed to obtain the temperature regression coefficient. Regression analysis is performed on historical user emotion data to determine the impact of emotion on power grid load. The historical current regression coefficient is then applied to the first historical current data to form a current regression coefficient label. Similarly, the historical temperature regression coefficient is applied to the first historical temperature data to form a temperature regression coefficient label. The historical user emotion data regression coefficient is applied to the historical user emotion data to form an emotion regression coefficient label. The first historical current data is then combined with its corresponding current regression coefficient label to form a first data set. The first historical temperature data is combined with its corresponding temperature regression coefficient label to form a second data set. The historical user emotion data is combined with its corresponding user emotion data regression coefficient label to form a third data set. Then, based on the first, second, and third data sets created, the regression coefficients are constructed to determine the model, and an appropriate algorithm (such as linear regression, decision tree, etc.) is selected. The model is trained using the training set, and the model parameters are adjusted to optimize performance. The model performance is continuously evaluated to ensure that it meets the preset training standards (such as accuracy, mean square error, etc.). Once the standards are met, the training process ends.

[0049] In this solution, by analyzing historical data, the grid load can be predicted more accurately, thereby improving the reliability of power supply. Accurate load forecasting can help power companies allocate resources more efficiently, reduce energy waste, and optimize power generation and distribution strategies.

[0050] S102, obtaining a preset grid state prediction formula, and calculating a grid state prediction observation value according to the current data, temperature data, user emotion data, current regression coefficient, temperature regression coefficient, user emotion data regression coefficient and the preset grid state prediction formula.

[0051] The preset grid state prediction formula may be a formula for predicting the state of the grid based on input features (such as current data, temperature data, and user emotion data). This formula usually includes regression coefficients for quantifying the impact of each input variable on the grid state.

[0052] The grid state prediction observation value can be the result calculated by the preset grid state prediction formula, which represents the prediction value of the grid system's operating state monitoring and evaluation. This observation value can be the load demand, voltage level, fault possibility, etc. of the grid, which is used to evaluate the grid's operating state. Specifically, it can represent the predicted health state of the grid system.

[0053] The preset grid state prediction formula can be read from the database, and the current data, temperature data, user emotion data, current regression coefficient, temperature regression coefficient, and user emotion data regression coefficient can be substituted into the formula to calculate the grid state prediction observation value.

[0054] On the basis of the above technical solution, an optional, preset grid state prediction formula is: ; in, Predict observations for power grid status; is a preset bias item; is the current regression coefficient; is the current data; is the temperature regression coefficient; is the temperature data; is the regression coefficient of user sentiment data; User sentiment data.

[0055] In this scheme, It represents the predicted grid state value when all independent variables (such as current, temperature, and user emotion data) are zero. It reflects the basic state of the model when there are no influencing factors. You can collect relevant historical data, including current, temperature, and user emotion data, as well as the corresponding grid state observations. Use a linear regression model to fit the data using methods such as the least squares method. The model will automatically calculate the best Then substitute the required parameters in the formula into the formula to calculate the predicted observation value of the power grid state.

[0056] S103, determining a multidimensional vector observation value according to the current data, temperature data and user emotion data, obtaining a preset power grid state update formula, and calculating the actual power grid state observation value according to the multidimensional vector observation value, the power grid state prediction observation value and the preset power grid state update formula; wherein the preset power grid state update formula is: ; in, is the actual observed value of the power grid status; is the pre-calculated Kalman gain; Predict observations for power grid status; is the multidimensional vector observation value; H is the pre-measured observation matrix.

[0057] A multidimensional vector observation is a vector containing multiple input variables, reflecting various data related to the power grid status. These data usually include current, temperature, and user emotion data. It can be expressed as: in, is the current data; is the temperature data; User sentiment data.

[0058] The actual observation value of the power grid state can be the real value of the operation status monitoring and evaluation of the power grid system obtained by measurement or calculation. This is the real state of the power grid at the current moment. Compared with the predicted observation value, the actual observation value will be more real-time and accurate. Specifically, the actual observation value of the power grid state can represent the actual health status of the entire power grid system.

[0059] The observation matrix can be a mathematical tool to describe how to map state variables to observable data. It plays a key role in Kalman filtering and other state estimation algorithms.

[0060] The Kalman gain can be used to adjust the relationship between the predicted grid state and the actual observed grid state. By calculating the Kalman gain, the real-time estimate of the grid state can be optimized to be more accurate. In grid monitoring, the Kalman gain determines the importance of the observations of current data, temperature data, and user emotion data when updating the grid state. If the Kalman gain is high, it means that the current observation data is considered more reliable and the model will rely more on these real-time observations; conversely, if the gain is low, it will rely more on the model's predictions.

[0061] The multidimensional vector observations can be determined using the current data, temperature data, and user emotion data according to the formula for constructing multidimensional vector observations. When the observation matrix is ​​determined, the observations associated with the state variables can be determined, such as the actual current and temperature data collected by sensors, and the collected user emotion data. According to the dynamic characteristics of the system, the relationship between the state and the observation is constructed. Usually H is a matrix where the number of rows corresponds to the number of observations and the number of columns corresponds to the number of state variables. For example, if there are three state variables (current, temperature, user emotion data) and one observation (actual observation of the power grid state), for example: Indicates that current, temperature, and emotion data directly correspond to state variables. Then use historical data for parameter estimation, and perform linear regression or other statistical analysis methods to determine the specific value of each element in the observation matrix, verify the validity of the observation matrix with actual measured data, and adjust it as needed. When calculating the Kalman gain, you can first calculate the predicted covariance matrix at the current moment based on the system dynamic model and the covariance matrix of the previous time step , specifically, this matrix describes the uncertainty of state estimation and can be calculated according to the following formula: Among them, P is the covariance matrix of the previous time step; A is the state transfer matrix; Q is the process noise covariance.

[0062] The state transfer matrix A describes the dynamic characteristics of the system state over time. It maps the state of the previous moment to the current moment and is usually used to predict the next state. For example, in a power grid model, the state may include current, temperature, and user emotion data, and the A matrix can represent how to calculate the current state based on the state of the previous moment.

[0063] The process noise covariance Q describes the noise introduced by model uncertainty or external interference in the state prediction process. It reflects the deviation between the system dynamic model and the actual system. For example, in power grid monitoring, process noise may come from equipment failure or external environmental changes. The elements of the Q matrix can be obtained through historical data analysis.

[0064] Then, the observation noise covariance R is estimated based on the characteristics of the sensor or historical data. Specifically, the observation data of the sensor over a period of time can be obtained, and then the residual is calculated: in, is the observed value measured by the sensor; is the corresponding true value. And through the collected residuals, statistical analysis is performed to calculate its variance. Specifically, it can be achieved by calculating the average value of the sum of squares of the residuals: Where N is the number of samples.

[0065] Then substitute it into the following formula: in, is the Kalman gain; is the covariance matrix of state prediction, which reflects the accuracy of state prediction. H is the pre-measured observation matrix; R is the covariance matrix of observation noise, which reflects the uncertainty of observation data; is the transpose of the observation matrix H; is the inverse matrix of the weighted observation covariance, which is used to calculate the gain.

[0066] Finally , , Substitute H into the formula to calculate the actual observed value of the power grid state.

[0067] S104, obtaining a preset grid state observation threshold, and if the actual observation value exceeds the preset grid state observation threshold, sending an alarm message to a control center.

[0068] The preset grid state observation threshold may refer to an upper limit or a lower limit of a grid operation parameter set according to historical data and system requirements, and is usually used to determine whether the grid is in a normal operating state.

[0069] Alarm information can be a notification generated when the power grid status exceeds a preset threshold, usually including the abnormality type, specific parameter value, timestamp and recommended response measures.

[0070] A control center can be an organization or system responsible for monitoring and managing grid operations, usually equipped with professionals and technical means to deal with abnormal grid conditions.

[0071] The preset grid status observation threshold can be read from the database, and then the actual observation value is compared with the preset threshold. If the actual observation value exceeds the threshold, an alarm message is generated. The alarm message is sent to the control center through an automated system or manually for subsequent processing.

[0072] In an embodiment of the present application, current data, temperature data and user emotion data of the power grid system are obtained, and the current data, temperature data and user emotion data are input into a preset regression coefficient determination model to obtain a current regression coefficient, a temperature regression coefficient and a user emotion data regression coefficient; a preset power grid state prediction formula is obtained, and a power grid state prediction observation value is calculated according to the current data, temperature data, user emotion data, current regression coefficient, temperature regression coefficient, user emotion data regression coefficient and a preset power grid state prediction formula; a multidimensional vector observation value is determined according to the current data, temperature data and user emotion data, and a preset power grid state update formula is obtained, and a power grid state actual observation value is calculated according to the multidimensional vector observation value, the power grid state prediction observation value and the preset power grid state update formula; wherein the preset power grid state update formula is: ; in, is the actual observed value of the power grid status; is the pre-calculated Kalman gain; Predict observations for power grid status; is a multidimensional vector observation value; H is a pre-measured observation matrix. A preset grid state observation threshold is obtained. If the actual observation value exceeds the preset grid state observation threshold, an alarm message is sent to the control center. Through the above-mentioned Internet of Things-based grid system monitoring method, by obtaining real-time current, temperature and user emotion data, the grid state can be dynamically monitored and the response capability to potential problems can be improved. The use of regression coefficients and prediction formulas can more accurately predict the grid state and help managers make timely decisions. Timely discovery and handling of potential risks can enhance the reliability and security of the grid system and ensure the continuity of power supply.

[0073] On the basis of the above technical solution, optionally, after obtaining a preset grid state observation threshold, if the actual observation value exceeds the preset grid state observation threshold, sending an alarm message to the control center, the method further includes: Acquire voltage data, frequency data, and device status data of the power grid system, input the current data, temperature data, voltage data, frequency data, and device status data into a preset fault detection model to obtain a fault score, and determine whether a power grid fault exists according to the fault score and a preset fault threshold; If the fault score exceeds the preset fault threshold, it is determined that a power grid fault exists, and the fault cause is determined by a preset fault detection model, and the fault cause is sent to a control center.

[0074] In this solution, voltage data can be the voltage level of power transmission in the measured power grid, usually in volts (V). Stable voltage level is an important factor to ensure the normal operation of power equipment. Too high or too low voltage may cause equipment damage or performance degradation.

[0075] Frequency data can represent the periodic changes of current in the power grid, usually in Hertz (Hz). The standard frequency of the power grid is usually 50Hz or 60Hz. The stability of frequency directly affects the operating efficiency and safety of power equipment. Frequency fluctuations may cause equipment failure or system instability.

[0076] Equipment status data can describe the operating conditions of various types of equipment in the power grid (such as transformers, circuit breakers and switches), usually including status identification (normal, faulty, waiting for maintenance, etc.), operating temperature, load conditions, etc. These data help operation and maintenance personnel monitor the health of equipment in real time and identify and handle faults in a timely manner.

[0077] The preset fault detection model can be an algorithmic model, usually based on machine learning or statistical methods, which is used to analyze input data (such as current, temperature, voltage, etc.) and identify potential faults. The model is trained on historical data to improve the accuracy of fault detection.

[0078] The fault score can be a numerical value output by the model to measure whether there is a fault in the power grid. The higher the score, the greater the possibility of a fault. It is obtained by analyzing the input data and comparing it with the normal operating mode learned by the model.

[0079] The preset fault threshold may be a predefined value used to determine the presence of a fault. When the fault score exceeds the threshold, the system will consider that a fault exists.

[0080] The cause of the fault can be the specific cause of the fault identified by the model, which may include equipment failure, overload, environmental factors, etc. Determining the cause of the fault helps to quickly locate the problem and take appropriate repair measures.

[0081] The acquired current data, temperature data, voltage data, frequency data and equipment status data can be organized into a specific format and input into a preset fault detection model. The model can be based on machine learning or statistical analysis methods. The model analyzes the input data and calculates the fault score. A high score indicates that the system is at risk of failure, and a low score indicates that the system is operating normally. The fault score is compared with the preset fault threshold. If the fault score exceeds the threshold, the system determines that there is a fault. The fault score is then analyzed using the preset fault detection model to determine the specific cause of the fault, such as equipment overload, abnormal temperature or voltage fluctuation. The identified fault cause is sent to the control center through a real-time communication system so that the operation and maintenance personnel can take timely measures to deal with it.

[0082] In this solution, through data input and model analysis, the power grid status can be monitored in real time and faults can be discovered in time. Once a fault is discovered, the system can quickly notify the control center to reduce the losses caused by the fault.

[0083] Based on the above technical solution, the optional, preset fault detection model training process includes: Acquire first historical voltage data, first historical frequency data, and first historical device status data, and acquire historical fault scores corresponding to the first historical current data, first historical temperature data, first historical voltage data, first historical frequency data, and first historical device status data; Creating a fifth data set according to the first historical current data, the first historical temperature data, the first historical voltage data, the first historical frequency data, and the first historical device status data, and labeling a fault score label of the fifth data set according to the historical fault score; Obtain historical fault records, and determine historical fault causes and operating parameter sets corresponding to each historical fault cause according to the historical fault records; wherein the operating parameter set includes second historical current data, second historical temperature data, second historical voltage data, second historical frequency data, and second historical device status data; Marking a fault cause label of the operating parameter set according to the historical fault cause; A fault detection model is constructed, and the fault detection model is trained according to the fifth data set, the fault score label, the operating parameter set, and the fault cause label until the fault detection model reaches a preset fault detection model training standard.

[0084] In this solution, the first historical voltage data may be the grid voltage level recorded within a specific time period, reflecting the stability of the grid operation state.

[0085] The first historical frequency data may be a recorded grid frequency value, which is usually around 50 Hz or 60 Hz, reflecting the stability of the grid frequency.

[0086] The first historical device status data may be the operating status of each device (such as a transformer, a switch, etc.) in the power grid, indicating whether the device is operating normally or faulty.

[0087] The historical fault score can be an indicator calculated based on past operating data and fault events to assess the fault risk of the power grid system under specific conditions. The higher the score, the greater the fault risk.

[0088] The historical fault record can be a detailed record of all fault events that occurred in the power grid system, including information such as the time, location, type of the fault and its impact on the system.

[0089] Historical failure causes can be the reasons that led to the historical failure events, such as equipment aging, external factors (such as weather changes), etc. Analyzing these reasons will help prevent subsequent failures.

[0090] The operating parameter set can be a set of operating data associated with a specific fault cause, including current, temperature, voltage, frequency, and device status data. These data are used to analyze the system status when the fault occurs.

[0091] The second historical current / temperature / voltage / frequency / equipment status data may be relevant data of the power grid system recorded in different time periods or under different conditions, and may be associated with different fault causes.

[0092] The fault cause labels can be used to identify and classify the labels of the fault causes in the running parameter set. These labels will be used to train the model so that the model can identify different fault causes.

[0093] The preset fault detection model training standard can be a defined standard for model training completion, such as accuracy, recall, or F1 score. The training is considered complete only when the model meets these standards.

[0094] Historical voltage and frequency data can be extracted from the grid monitoring system or data logger. This can be obtained through database query or data export function. Equipment status data usually comes from the equipment monitoring system and records the operating status of the equipment (normal or faulty). This can also be obtained through database access. Fault scores are usually derived from historical data analysis and may be calculated using fault detection algorithms or preset scoring criteria. Ensure that these data are obtained in the same time period to ensure consistency. Then integrate the above data into a structured data set, ensuring that each record contains a timestamp and all relevant parameters. Associate the historical fault scores with the corresponding timestamps and add a fault score tag to each record. Then extract the fault records from the fault management system, including information such as the time, location and type of the fault. Analyze the root causes of each fault against the fault records and record these causes. The operating parameter set should include the second historical current data, temperature data, voltage data, frequency data and equipment status data related to the fault cause. This data can be extracted from the data store. Match each operating parameter record with the corresponding fault cause and add a fault cause tag. Then, using the fifth data set, fault score labels, operating parameter sets, and fault cause labels, select an appropriate machine learning algorithm (such as decision tree, random forest, support vector machine, etc.). Divide the data into training sets and test sets, and train the model according to the preset training standards (such as accuracy, F1 score, etc.). During the model training process, continuously evaluate the model performance until the preset training standards are met.

[0095] In this solution, through the analysis of historical data and fault causes, the model can more accurately identify potential faults and reduce missed and false alarms. Timely detection of potential risks of faults can provide early warnings to operators, reducing downtime and accident risks.

[0096] Embodiment 2 Figure 2 Schematic diagram of the flow of the power grid system monitoring method based on the Internet of Things provided in the second embodiment of the present application. Figure 2 As shown, the specific steps include: S201, obtaining current data, temperature data and user emotion data of the power grid system, inputting the current data, temperature data and user emotion data into a preset regression coefficient determination model, and obtaining a current regression coefficient, a temperature regression coefficient and a user emotion data regression coefficient.

[0097] S202, obtaining a preset grid state prediction formula, and calculating a grid state prediction observation value according to the current data, temperature data, user emotion data, current regression coefficient, temperature regression coefficient, user emotion data regression coefficient and the preset grid state prediction formula.

[0098] S203, determining a multidimensional vector observation value according to the current data, temperature data and user emotion data, obtaining a preset power grid state update formula, and calculating the actual power grid state observation value according to the multidimensional vector observation value, the power grid state prediction observation value and the preset power grid state update formula; wherein the preset power grid state update formula is: ; in, is the actual observed value of the power grid status; is the pre-calculated Kalman gain; Predict observations for power grid status; is the multidimensional vector observation value; H is the pre-measured observation matrix.

[0099] S204, obtaining a preset grid state observation threshold, and if the actual observation value exceeds the preset grid state observation threshold, sending an alarm message to a control center.

[0100] S205, inputting the current data, temperature data and user emotion data into a preset load prediction model to obtain predicted load data of the power grid.

[0101] The preset load forecasting model can be a mathematical or statistical model used to estimate the power demand of the power grid in a specific time period in the future. It is usually trained based on historical data and related factors (such as current, temperature, and user sentiment data) to capture the laws that affect power demand.

[0102] Predicted load data can be the expected power demand value of the power grid at a certain point in time or time period in the future calculated by the load forecasting model. It reflects the prediction of the power grid load based on the input data (current, temperature, mood, etc.), helping power grid operators to dispatch and allocate resources.

[0103] Current data, temperature data, and emotion data can be organized into input formats to ensure that they can be accepted by the model. The model uses the parameters and features learned internally to calculate the input data and output the predicted load value of the power grid.

[0104] Based on the above technical solution, the optional, preset load forecasting model training process includes: Obtaining historical grid loads corresponding to first historical current data, first historical temperature data, and historical user emotion data; Creating a fourth data set according to the first historical current data, the first historical temperature data, and the historical user emotion data, and labeling a grid load label of the fourth data set according to the historical grid load; A load prediction model is constructed, and the load prediction model is trained according to the fourth data set and the power grid load label until the load prediction model reaches a preset load prediction model training standard.

[0105] In this solution, historical grid load refers to the total amount of electricity consumed by the grid system in a certain period of time in the past, usually expressed in kilowatt-hours (kWh) or megawatts (MW). It reflects the actual demand of the grid.

[0106] The fourth data set may be a set including first historical current data, first historical temperature data, and historical user emotion data. Each data record corresponds to a time point and includes relevant feature information for model training.

[0107] The grid load label can be a target value assigned to each record in the fourth data set, indicating the actual grid load at the corresponding time point. The label is used for supervised learning to help the model learn the relationship between input features and load.

[0108] The preset load forecasting model training standard may refer to performance indicators set during the training process, such as the accuracy of the model, mean square error (MSE), etc. When the model reaches these indicators, the training may end.

[0109] The first historical current data, the first historical temperature data, and the historical user emotion data, as well as the corresponding historical grid load data, can be obtained. These data are integrated to create a data set containing input features (current, temperature, emotion), which is named the fourth data set. Corresponding grid load labels are assigned to each record of the fourth data set, and these labels come from the historical grid load data. According to the data characteristics and prediction requirements, a suitable prediction model is selected, such as linear regression, decision tree, random forest, or deep learning model. The fourth data set and its grid load labels are used for model training. The model should be able to learn from the input current, temperature, and user emotion data to predict the grid load. Set preset load prediction model training standards, such as the accuracy of the model, root mean square error (RMSE), or other performance indicators. When the model meets these standards, the training process ends. Then use the test set (data set not used for training) to verify the effect of the model. If the effect is not good, the model parameters can be adjusted or the model can be replaced. Then the model parameters are optimized through repeated iterations until the model performance reaches the preset training standards.

[0110] In this solution, training the model enables the model to automatically make decisions based on the input data, reducing manual intervention and improving work efficiency. The model can use multiple types of data (such as current, temperature, and user emotion data) at the same time to provide more comprehensive analysis results.

[0111] S206, obtaining real-time output power data, minimum output power, maximum output power and a preset dispatch power calculation formula of each distributed power source of the power grid, and determining the dispatch power of each distributed power source according to the real-time output power data, minimum output power, maximum output power and the preset dispatch power calculation formula; wherein the preset dispatch power calculation formula is: in, is the dispatching power of each distributed power source; is the minimum output power of each distributed power source; is the maximum output power of each distributed power source; To predict load data; is the total output power of other distributed power sources j, which represents the sum of the real-time output powers of all other power sources different from distributed power source i.

[0112] In this solution, distributed power generation can refer to small power generation equipment installed in distributed locations of the power grid (such as near users). These devices can be solar panels, wind turbines, gas turbines, etc., which usually have a small power generation capacity and can directly provide electricity to local loads, reducing dependence on large centralized power stations.

[0113] Real-time output power data can refer to the actual power value generated by each distributed power source within a specific time, in watts (W) or kilowatts (kW). This data reflects the power generation capacity of the power source during operation, which may change due to factors such as weather and equipment status.

[0114] The minimum output power can refer to the lowest power that a distributed power source can continuously output under normal operating conditions. This is usually the safety range of equipment design and operation, ensuring that the equipment can still work stably under low load conditions.

[0115] The maximum output power refers to the highest power that a distributed power source can output under ideal conditions. Exceeding this power may cause equipment overload or damage, so it needs to be limited in scheduling and operation.

[0116] Dispatching power can refer to the ideal output power of distributed power sources determined based on the grid load demand and the output conditions of other power sources. Dispatching power aims to balance grid supply and demand and ensure stable operation of the grid. Based on real-time output power data, minimum and maximum output power, and the preset dispatching power calculation formula, the output of distributed power sources can be dynamically adjusted to adapt to real-time load changes.

[0117] The generated power of each power source can be collected in real time through sensors and monitoring equipment installed on each distributed power source. Usually, this data is collected and centralized to the control center through a monitoring system. The minimum output power of each distributed power source can be obtained from the technical specifications or operation manual provided by the equipment manufacturer. The maximum output power is also determined by the technical parameters of the equipment. Then, based on the real-time output power data of each distributed power source, the sum of the real-time output power of all other distributed power sources is calculated as the total output power of other distributed power sources j, and then the parameters required by the formula are substituted into the formula to calculate the dispatch power of each distributed power source.

[0118] In this embodiment, by integrating multiple data sources (such as environmental factors and public sentiment), grid load changes can be predicted more accurately, reducing the risk of supply and demand imbalance. Based on real-time data and prediction results, power resources can be allocated more efficiently to ensure that each distributed power source operates within its capacity to avoid overload or idleness.

[0119] Embodiment 3 Figure 3 Schematic diagram of the structure of the power grid system monitoring system based on the Internet of Things provided in the third embodiment of the present application. Figure 3 As shown, specifically including: The acquisition module 301 is used to obtain the current data, temperature data and user emotion data of the power grid system, and input the current data, temperature data and user emotion data into a preset regression coefficient determination model to obtain the current regression coefficient, temperature regression coefficient and user emotion data regression coefficient; The prediction module 302 is used to obtain a preset grid state prediction formula, and calculate the grid state prediction observation value according to the current data, temperature data, user emotion data, current regression coefficient, temperature regression coefficient, user emotion data regression coefficient and the preset grid state prediction formula; The state update module 303 is used to determine the multidimensional vector observation value according to the current data, temperature data and user emotion data, obtain the preset power grid state update formula, and calculate the actual power grid state observation value according to the multidimensional vector observation value, the power grid state prediction observation value and the preset power grid state update formula; wherein the preset power grid state update formula is: ; in, is the actual observed value of the power grid status; is the Kalman gain; Predict observations for power grid status; is the multidimensional vector observation value; H is the pre-measured observation matrix.

[0120] The alarm module 304 is used to obtain a preset grid state observation threshold, and send an alarm message to the control center if the actual observation value exceeds the preset grid state observation threshold.

[0121] In an embodiment of the present application, an acquisition module is used to obtain current data, temperature data and user emotion data of a power grid system, and input the current data, temperature data and user emotion data into a preset regression coefficient determination model to obtain a current regression coefficient, a temperature regression coefficient and a user emotion data regression coefficient; a prediction module is used to obtain a preset power grid state prediction formula, and calculate a power grid state prediction observation value according to the current data, temperature data, user emotion data, current regression coefficient, temperature regression coefficient, user emotion data regression coefficient and a preset power grid state prediction formula; a state update module is used to determine a multidimensional vector observation value according to the current data, temperature data and user emotion data, obtain a preset power grid state update formula, and calculate a power grid state actual observation value according to the multidimensional vector observation value, the power grid state prediction observation value and the preset power grid state update formula; wherein the preset power grid state update formula is: ; in, is the actual observed value of the power grid status; is the Kalman gain; Predict observations for power grid status; is a multidimensional vector observation value; H is a pre-measured observation matrix; the alarm module is used to obtain a preset grid state observation threshold, and if the actual observation value exceeds the preset grid state observation threshold, an alarm message is sent to the control center. Through the above-mentioned Internet of Things-based grid system monitoring system, by obtaining real-time current, temperature and user emotion data, the grid state can be dynamically monitored and the response capability to potential problems can be improved. The use of regression coefficients and prediction formulas can more accurately predict the grid state and help managers make timely decisions. Timely discovery and handling of potential risks can enhance the reliability and security of the grid system and ensure the continuity of power supply.

[0122] The Internet of Things-based power grid system monitoring system provided in the embodiment of the present application can achieve Figure 1 To avoid repetition, the various processes implemented by the method embodiment are not described here.

[0123] Embodiment 4 like Figure 4 As shown, an embodiment of the present application also provides an electronic device 400, including a processor 401, a memory 402, and a program or instruction stored in the memory 402 and executable on the processor 401. When the program or instruction is executed by the processor 401, each process of the above-mentioned embodiment of the method for monitoring the power grid system based on the Internet of Things is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0124] It should be noted that the electronic devices in the embodiments of the present application include the mobile electronic devices and non-mobile electronic devices mentioned above.

[0125] Embodiment 5 An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, each process of the above-mentioned cable installation process based on the tension adaptive control system embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0126] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0127] It should be noted that, in this article, the term "comprises", "includes" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or system including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, the elements defined by the sentence "including one..." do not exclude the existence of other identical elements in the process, method, article or system including the element. In addition, it should be pointed out that the scope of the method and system in the embodiment of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0128] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0129] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

[0130] The above are only preferred embodiments of the present application and the technical principles used. The present application is not limited to the specific embodiments described herein, and various obvious changes, readjustments and substitutions that can be made by those skilled in the art will not deviate from the scope of protection of the present application. Therefore, although the present application is described in more detail through the above embodiments, the present application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the claims.

Claims

1. A power grid system monitoring method based on the Internet of Things, characterized in that: The method comprises: Acquire current data, temperature data and user emotion data of the power grid system, input the current data, temperature data and user emotion data into a preset regression coefficient determination model, and obtain a current regression coefficient, a temperature regression coefficient and a user emotion data regression coefficient; Obtain a preset grid state prediction formula, and calculate a grid state prediction observation value according to the current data, temperature data, user emotion data, current regression coefficient, temperature regression coefficient, user emotion data regression coefficient, and the preset grid state prediction formula; Determine a multidimensional vector observation value according to the current data, temperature data and user emotion data, obtain a preset power grid state update formula, and calculate the actual power grid state observation value according to the multidimensional vector observation value, the power grid state prediction observation value and the preset power grid state update formula; wherein the preset power grid state update formula is: ； in, is the actual observed value of the power grid status; is the pre-calculated Kalman gain; Predict observations for power grid status; is the multidimensional vector observation value; H is the pre-measured observation matrix; A preset power grid state observation threshold is obtained, and if the actual observation value exceeds the preset power grid state observation threshold, an alarm message is sent to a control center.

2. The method for monitoring a power grid system based on the Internet of Things according to claim 1, characterized in that: The preset grid status prediction formula is: ； in, Predict observations for power grid status; is a preset bias item; is the current regression coefficient; is the current data; is the temperature regression coefficient; is the temperature data; is the regression coefficient of user sentiment data; User sentiment data.

3. The method for monitoring a power grid system based on the Internet of Things according to claim 1, characterized in that: After obtaining the preset power grid state observation threshold, if the actual observation value exceeds the preset power grid state observation threshold, sending an alarm message to the control center, the method further includes: Inputting the current data, temperature data and user emotion data into a preset load prediction model to obtain predicted load data of the power grid; The real-time output power data, minimum output power, maximum output power and preset dispatching power calculation formula of each distributed power source of the power grid are obtained, and the dispatching power of each distributed power source is determined according to the real-time output power data, minimum output power, maximum output power and preset dispatching power calculation formula; wherein the preset dispatching power calculation formula is: in, is the dispatching power of each distributed power source; is the minimum output power of each distributed power source; is the maximum output power of each distributed power source; To predict load data; is the total output power of other distributed power sources j, which represents the sum of the real-time output powers of all other power sources different from distributed power source i.

4. The method for monitoring a power grid system based on the Internet of Things according to claim 1, characterized in that: After obtaining the preset power grid state observation threshold, if the actual observation value exceeds the preset power grid state observation threshold, sending an alarm message to the control center, the method further includes: Acquire voltage data, frequency data, and device status data of the power grid system, input the current data, temperature data, voltage data, frequency data, and device status data into a preset fault detection model to obtain a fault score, and determine whether a power grid fault exists according to the fault score and a preset fault threshold; If the fault score exceeds the preset fault threshold, it is determined that a power grid fault exists, and the fault cause is determined by a preset fault detection model, and the fault cause is sent to a control center.

5. The method for monitoring a power grid system based on the Internet of Things according to claim 1, characterized in that: The preset regression coefficient determines the model training process, including: Obtaining first historical current data, first historical temperature data, historical user emotion data, historical current regression coefficient, historical temperature regression coefficient, and historical user emotion data regression coefficient; labeling a current regression coefficient label of the first historical current data according to the historical current regression coefficient, labeling a temperature regression coefficient label of the first historical temperature data according to the historical temperature regression coefficient, and labeling a user emotion data regression coefficient label of the historical user emotion data according to the historical user emotion data regression coefficient; Creating a first data set based on the first historical current data and the current regression coefficient label, creating a second data set based on the first historical temperature data and the temperature regression coefficient label, and creating a third data set based on the historical user emotion data and the user emotion data regression coefficient label; A regression coefficient determination model is constructed, and the regression coefficient determination model is trained according to the first data set, the second data set, and the third data set until the regression coefficient determination model reaches a preset regression coefficient determination model training standard.

6. The method for monitoring a power grid system based on the Internet of Things according to claim 3, characterized in that: The training process of the preset load forecasting model includes: Acquire the historical grid load corresponding to the first historical current data, the first historical temperature data, and the historical user emotion data; Creating a fourth data set according to the first historical current data, the first historical temperature data, and the historical user emotion data, and labeling a grid load label of the fourth data set according to the historical grid load; Constructing a load prediction model, and training the load prediction model according to the fourth data set and the power grid load label until the load prediction model reaches a preset load prediction model training standard.

7. The method for monitoring a power grid system based on the Internet of Things according to claim 4, characterized in that: The training process of the preset fault detection model includes: Acquire first historical voltage data, first historical frequency data, and first historical device status data, and acquire historical fault scores corresponding to the first historical current data, first historical temperature data, first historical voltage data, first historical frequency data, and first historical device status data; Creating a fifth data set according to the first historical current data, the first historical temperature data, the first historical voltage data, the first historical frequency data, and the first historical device status data, and labeling a fault score label of the fifth data set according to the historical fault score; Obtain historical fault records, and determine historical fault causes and operating parameter sets corresponding to each historical fault cause according to the historical fault records; wherein the operating parameter set includes second historical current data, second historical temperature data, second historical voltage data, second historical frequency data, and second historical device status data; Marking a fault cause label of the operating parameter set according to the historical fault cause; A fault detection model is constructed, and the fault detection model is trained according to the fifth data set, the fault score label, the operating parameter set, and the fault cause label until the fault detection model reaches a preset fault detection model training standard.

8. A power grid system monitoring system based on the Internet of Things, characterized in that: The system comprises: The acquisition module is used to obtain the current data, temperature data and user emotion data of the power grid system, and input the current data, temperature data and user emotion data into a preset regression coefficient determination model to obtain the current regression coefficient, temperature regression coefficient and user emotion data regression coefficient; A prediction module, used to obtain a preset grid state prediction formula, and calculate a grid state prediction observation value according to the current data, temperature data, user emotion data, current regression coefficient, temperature regression coefficient, user emotion data regression coefficient and the preset grid state prediction formula; A state update module is used to determine a multidimensional vector observation value according to the current data, temperature data and user emotion data, obtain a preset power grid state update formula, and calculate the actual power grid state observation value according to the multidimensional vector observation value, the power grid state prediction observation value and the preset power grid state update formula; wherein the preset power grid state update formula is: ； in, is the actual observed value of the power grid status; is the Kalman gain; Predict observations for power grid status; is the multidimensional vector observation value; H is the pre-measured observation matrix; The alarm module is used to obtain a preset grid state observation threshold, and if the actual observation value exceeds the preset grid state observation threshold, an alarm message is sent to the control center.

9. An electronic device, characterized in that: It includes a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the power grid system monitoring method based on the Internet of Things as described in any one of claims 1 to 7.

10. A readable storage medium, characterized in that: The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, the steps of the power grid system monitoring method based on the Internet of Things are implemented as described in any one of claims 1-7.

Images