Intelligent power monitoring management system based on Internet of Things

Abstract

An intelligent power monitoring management system based on Internet of Things belongs to the technical field of power supervision and comprises a data acquisition module, a communication transmission module, a data processing and analysis module, an edge calculation module, an intelligent control module, an early warning management module and an authority control module. According to the invention, by deploying various types of sensors, real-time acquisition and dynamic monitoring of key parameters of the power system are realized. The system can obtain and analyze data such as voltage, current and power in time, ensures the accuracy and high efficiency of operation of the power system, and effectively avoids energy consumption waste and equipment faults. The edge computing technology is adopted, preprocessing and preliminary analysis are carried out at a data acquisition end, the data processing pressure of a central server is relieved, the bandwidth requirement for data transmission is lowered, and the response speed is increased. Even if the network is interrupted, the edge device can realize short-term data caching, and data continuity is guaranteed. And performing anomaly detection on the power system, identifying potential risks and giving an alarm.

Description

Technical Field

[0001] The present invention belongs to the technical field of power supervision, and more specifically, particularly relates to an intelligent power monitoring and management system based on the Internet of Things. Background Art

[0002] The IoT system relies on network connections for real-time data transmission. However, in complex industrial environments or remote areas, unstable network signals may cause data loss or transmission delays, affecting the accuracy and response speed of the monitoring system. In addition, the connection of a large number of devices may also lead to increased bandwidth pressure, further affecting the transmission quality. The widespread connection of IoT devices and the cloud transmission of data increase the security risks of the system. Since power monitoring data involves critical infrastructure, the system faces a high risk of cyber attacks, including data theft, tampering or destruction. How to protect data privacy and prevent malicious attacks are severe challenges facing the system. In the field of power monitoring, there are many devices and sensors from different suppliers, and the system may need to be compatible with different protocols and standards. Since the standards for IoT devices have not yet been fully unified, there are interoperability issues during system integration, which limits the flexibility and scalability of the system.

[0003] As the number of IoT devices increases, the amount of real-time data that the system needs to process continues to increase. Existing systems may lack data processing and analysis capabilities, especially when faced with large-scale, multi-type data sources. They may not be able to complete complex data analysis in real time, affecting the effectiveness of prediction, diagnosis, and optimization. The IoT power monitoring system relies on a large number of sensors and hardware devices, which are exposed to different environments and may be affected by temperature, humidity, dust or other environmental factors. The maintenance and replacement costs are high, and the system needs to be able to monitor the status of the equipment and issue an alarm when a failure occurs, which adds complexity. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the above or existing problems of the intelligent power monitoring and management system based on the Internet of Things, the present invention is proposed.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] The embodiment of the present invention provides an intelligent power monitoring and management system based on the Internet of Things, including: a data acquisition module, which is used to collect power parameters in real time, including current, voltage, power factor, temperature and power consumption, and perform preliminary processing on the collected data; a communication transmission module, which is used to transmit data from the acquisition end to the server and management platform, and support remote monitoring, configuration and fault diagnosis; a data processing and analysis module, which is used to clean, format and store the collected data, and use machine learning algorithms to perform prediction, anomaly detection and trend analysis; an edge computing module, which is used to reduce the network transmission load and improve the response speed by performing preliminary calculations and preprocessing at the acquisition end, and cache data when the network is interrupted; an intelligent control module, which is used to execute remote control instructions based on the analysis results, including load adjustment, power failure control and time-sharing scheduling; an early warning management module, which is used to monitor the operating status of the power system in real time, detect abnormal situations in time and issue early warning notifications, and generate alarms of different levels according to set thresholds; an authority control module, which is used for user identity authentication, access authority allocation and management.

[0008] As a preferred solution of the intelligent power monitoring and management system based on the Internet of Things described in the present invention, the real-time collection of power parameters and preliminary processing of the collected data include: selecting the sampling rate according to the dynamic demand of the power system, such as 500ms or 1s interval, increasing the sampling frequency in a high dynamic power environment, and for high-frequency collected data, aggregating the data by time period, such as calculating the average, maximum, minimum and standard deviation of the sampling points every minute; filtering the data points that exceed the preset upper or lower limit to eliminate random noise in the collected signal, so as to improve the accuracy of subsequent analysis.

[0009] As a preferred solution of the intelligent power monitoring and management system based on the Internet of Things described in the present invention, the data is transmitted from the collection end to the server and management platform to support remote monitoring, configuration and fault diagnosis, including: changing the frequency, alarm threshold and transmission protocol of the collection parameters by sending configuration instructions, encrypting the transmission of configuration instructions, and adopting role authority control; the platform receives and analyzes the alarm data, and performs fault diagnosis on abnormal conditions in the power system.

[0010] As a preferred solution of the intelligent power monitoring and management system based on the Internet of Things described in the present invention, the cleaning, formatting and storage of the collected data include:

[0011] For detecting mutation points, the anomaly is judged based on the average value of the data in the sliding window. Assuming that the sampling data x t The average value in the time window N is μ N , with standard deviation σ N， If x t If the deviation from the mean is k times the standard deviation, it is considered abnormal, that is:

[0012] ∣x t -μ N ∣>k·σ N

[0013] For smoothing data, high-frequency noise components are eliminated. Assuming the original signal is S (t) , the signal after low-pass filtering is S filtered (t), using a moving average filter:

[0014]

[0015] Where N is the window size.

[0016] As a preferred solution of the intelligent power monitoring and management system based on the Internet of Things described in the present invention, the prediction, anomaly detection and trend analysis using machine learning algorithms include:

[0017] Long short-term memory network is used to predict power demand. Historical power data is collected and divided into training set and test set. The number of LSTM layers and units is defined to capture the time series characteristics. The LSTM formula is as follows:

[0018] Forget Gate:

[0019] f t =σ(W f ·[h t-1 ,x t ]+b f )

[0020] Input Gate:

[0021] i t =σ(W i ·[h t-1 ,x t ]+b i )

[0022] Memory unit update:

[0023] C t =f t ·C t-1 +i t ·tanh(WC·[h t-1 ,x t ]+b C )

[0024] Output Gate:

[0025] h t =o t tanh(C t )

[0026] Use historical electricity data for training and define the loss function as mean square error:

[0027]

[0028] Among them, n is the total number of samples, y i is the true value of the i-th sample, is the predicted value of the i-th sample;

[0029] Input the latest power data for prediction and output the power demand for several time steps in the future;

[0030] The isolation forest algorithm is used to detect outliers in power data. The parameters of the isolation forest are defined as follows:

[0031]

[0032] Among them, h(x) is the number of splits of data point x from the root node to the leaf node, s(x,n) is the degree of abnormality measured by path length, and c(n) is the path length normalization constant for the number of samples n;

[0033] Input real-time power data, calculate the anomaly score of each data point, set a threshold, and when the score exceeds the threshold, mark the data as abnormal and trigger an alarm;

[0034] The Prophet time series model is used to analyze the trend of power data. The historical data of power load is collected, and the data is decomposed into periodicity and seasonality to capture the changes on different time scales and construct the Prophet model:

[0035] The Prophet model decomposes the time series into trend, seasonality, and holiday components:

[0036] y(t)=g(t)+s(t)+h(t)+∈t

[0037] Among them, g(t) is the trend component, s(t) is the seasonal component, and h(t) is the holiday component;

[0038] Assuming the trend is linear growth or saturated growth, for a linear trend:

[0039] g(t)=(k+a(t))t+b

[0040] Among them, k is the growth rate and b is the offset;

[0041] Through historical data, the Prophet model is fitted and the hyperparameters are adjusted using Bayesian optimization to improve the prediction effect.

[0042] As a preferred solution of the intelligent power monitoring and management system based on the Internet of Things described in the present invention, wherein: the preliminary calculation and preprocessing are performed at the acquisition end to reduce the network transmission load, improve the response speed, and cache data when the network is interrupted, including: through preprocessing, reducing the computing burden of the server after receiving the data, accelerating the processing and analysis of the data, and improving the system response speed; the edge device performs part of the calculation locally and directly returns the result to ensure that the information can reach the user end quickly and realize instant feedback. In the event of network interruption or insufficient bandwidth, the acquisition end automatically caches the data and stores it in the local memory. When the network is restored, the cached data is sent to the server in chronological order to ensure the integrity and continuity of the data.

[0043] As a preferred solution of the intelligent power monitoring and management system based on the Internet of Things described in the present invention, wherein: the execution of remote control instructions based on the analysis results, including load adjustment, power-off control and time-sharing scheduling, includes: the system continuously analyzes real-time power parameters, combines prediction and trend analysis to determine the current load status, and when it is detected that the load is close to the set overload threshold, the system will automatically send instructions to adjust the power of non-critical equipment or systems; the system monitors power parameters in real time, and combines the anomaly detection function. When an abnormal situation or an emergency failure occurs, the system automatically generates a power-off control instruction, and the system will selectively interrupt the power supply through the power switch or relay of the remote control device; the system analyzes the peak and trough time periods of power demand in a day through historical data and prediction models, and generates a time-sharing scheduling plan based on the analysis system, restricts the operation of certain equipment in peak hours in advance, or arranges high-energy-consuming tasks to be executed during low hours.

[0044] As a preferred solution of the intelligent power monitoring and management system based on the Internet of Things described in the present invention, the real-time monitoring of the operating status of the power system, timely detection of abnormal situations and issuance of early warning notifications include: the collected data is transmitted to the edge computing module or the cloud server through the communication transmission module, and combined with historical data and set operating standards, the health status of the equipment and the overall system is continuously evaluated; through preset threshold rules and machine learning algorithms, the system automatically analyzes various parameters to detect whether there are abnormal values ​​that deviate from the normal range.

[0045] As a preferred solution of the intelligent power monitoring and management system based on the Internet of Things described in the present invention, the generation of different levels of alarms according to the set thresholds includes:

[0046] When the set voltage reaches 360V or the current reaches 445A, an information alarm is issued:

[0047] If the voltage drops to 360V, the system generates an informational alarm indicating low voltage;

[0048] If the current drops to 445A, the system generates a low-level alarm for the operator’s attention;

[0049] When the set voltage reaches 352V or the current reaches 435A, a warning level alarm is issued:

[0050] If the voltage fluctuates to 352V and the fluctuation lasts for more than 5 minutes, and the current drops to 435A or rises to 555A, a medium alarm is triggered, indicating a local fault;

[0051] When the set voltage drops to 340V, the current is higher than 575A or the temperature exceeds 90℃, a serious alarm is issued:

[0052] If the voltage drops to 340V, there is a problem with the equipment or the power supply system;

[0053] If the current rises to 575A, there is a risk of overload or line abnormality;

[0054] If the temperature reaches 90°C, there is a risk of high temperature failure.

[0055] As a preferred solution of the intelligent power monitoring and management system based on the Internet of Things described in the present invention, the user identity authentication, access permission allocation and management include: the system divides users into different roles, such as ordinary users, operators, administrators and super administrators, and different roles have different permissions: the system sets permission levels for different roles, ordinary users can view data; operators can perform equipment configuration and data debugging; administrators can add and delete users and manage the system; super administrators have full access and control permissions.

[0056] The beneficial effects of the present invention are as follows: the present invention realizes real-time collection and dynamic monitoring of key parameters of the power system by deploying various types of sensors. The system can timely acquire and analyze data such as voltage, current, and power to ensure the accuracy and efficiency of the power system operation, and effectively avoid energy waste and equipment failure. Edge computing technology is used to perform preprocessing and preliminary analysis at the data acquisition end to reduce the data processing pressure of the central server, reduce the bandwidth requirements for data transmission, and improve the response speed. Even in the event of a network interruption, edge devices can achieve short-term data caching to ensure data continuity. By combining machine learning algorithms and data analysis, the operating status of the power system is detected for abnormalities, potential risks are identified, and multi-level alarms are issued. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] 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 labor.

[0058] Figure 1 A schematic diagram of the structure of an intelligent power monitoring and management system based on the Internet of Things provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0060] 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.

[0061] 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 does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0062] Example 1

[0063] Reference below Figure 1 , Figure 1 This is a schematic diagram of the structure of an intelligent power monitoring and management system based on the Internet of Things provided by an embodiment of the present invention. It should be noted that the embodiments of the present invention can be applied to any applicable scenario.

[0064] S1: Real-time collection of power parameters, including current, voltage, power factor, temperature and power consumption, and preliminary processing of the collected data.

[0065] Preferably, the sampling rate is selected according to the dynamic demand of the power system, such as 500ms or 1s interval, and the sampling frequency is increased in a highly dynamic power environment. For data collected at a high frequency, the data is aggregated by time period, such as calculating the average, maximum, minimum and standard deviation of the sampling points every minute;

[0066] Filter data points that exceed the preset upper or lower limit to eliminate random noise in the acquired signal in order to improve the accuracy of subsequent analysis.

[0067] S2: Transmits data from the collection end to the server and management platform to support remote monitoring, configuration and fault diagnosis.

[0068] Preferably, the frequency, alarm threshold, transmission protocol and other settings of the acquisition parameters are changed by sending configuration instructions, the configuration instructions are transmitted in encrypted form, and role-based authority control is adopted; the platform receives and analyzes the alarm data, and performs fault diagnosis on abnormal conditions occurring in the power system.

[0069] Furthermore, the acquisition frequency is set to 500ms, the upper limit of the voltage alarm is 240V, the lower limit is 200V, and the transmission protocol is HTTPS. The platform receives the encrypted alarm data sent by the acquisition end device in real time, and decrypts and analyzes the alarm content; based on the alarm data and historical data, the platform uses algorithms to analyze the possible causes of the alarm. Combining historical abnormal data and existing alarm information, the system may use a support vector machine model for preliminary fault classification and location; the system infers the cause of the fault based on the characteristics of the alarm data (such as voltage and current fluctuations). For example, when the current and voltage fluctuate greatly, the platform determines that it is a problem of instantaneous overload of the power grid, aging of equipment, etc.; the platform generates a diagnostic report, sends the cause of the fault, location information and recommended measures to relevant personnel, and records it in the alarm log for future reference and analysis.

[0070] S3: Cleans, formats, and stores the collected data, and uses machine learning algorithms for prediction, anomaly detection, and trend analysis.

[0071] Preferably, for detecting mutation points, the abnormality is determined based on the average value of the data in the sliding window. Assume that the sampling data x t The average value in the time window N is μ N , with standard deviation σ N， If x t If the deviation from the mean is k times the standard deviation, it is considered abnormal, that is:

[0072] ∣x t -μ N ∣>k·σ N

[0073] For smoothing data, high-frequency noise components are eliminated. Assuming the original signal is S (t) , the signal after low-pass filtering is S filtered (t), using a moving average filter:

[0074]

[0075] Where N is the window size.

[0076] Long short-term memory network is used to predict power demand. Historical power data is collected and divided into training set and test set. The number of LSTM layers and units is defined to capture the time series characteristics. The LSTM formula is as follows:

[0077] Forget Gate:

[0078] f t =σ(W f ·[h t-1 ,x t ]+b f )

[0079] Input Gate:

[0080] i t =σ(W i ·[h t-1 ,x t ]+b i )

[0081] Memory unit update:

[0082] C t =f t ·C t-1 +i t ·tanh(WC·[h t-1 ,x t ]+b C )

[0083] Output Gate:

[0084] h t =o t tanh(C t )

[0085] Use historical electricity data for training and define the loss function as mean square error:

[0086]

[0087] Among them, n is the total number of samples, y i is the true value of the i-th sample, is the predicted value of the i-th sample;

[0088] Input the latest power data for prediction and output the power demand for several time steps in the future;

[0089] The isolation forest algorithm is used to detect outliers in power data. The parameters of the isolation forest are defined as follows:

[0090]

[0091] Among them, h(x) is the number of splits of data point x from the root node to the leaf node, s(x,n) is the degree of abnormality measured by path length, and c(n) is the path length normalization constant for the number of samples n;

[0092] Input real-time power data, calculate the anomaly score of each data point, set a threshold, and when the score exceeds the threshold, mark the data as abnormal and trigger an alarm;

[0093] The Prophet time series model is used to analyze the trend of power data. The historical data of power load is collected, and the data is decomposed into periodicity and seasonality to capture the changes on different time scales and construct the Prophet model:

[0094] The Prophet model decomposes the time series into trend, seasonality, and holiday components:

[0095] y(t)=g(t)+s(t)+h(t)+∈t

[0096] Among them, g(t) is the trend component, s(t) is the seasonal component, and h(t) is the holiday component;

[0097] Assuming the trend is linear growth or saturated growth, for a linear trend:

[0098] g(t)=(k+a(t))t+b

[0099] Among them, k is the growth rate and b is the offset;

[0100] Through historical data, the Prophet model is fitted and the hyperparameters are adjusted using Bayesian optimization to improve the prediction effect.

[0101] Furthermore, the Prophet model is used for data fitting. Prophet decomposes the time series data into the following components:

[0102] Trend component: reflects the long-term trend of data changes, such as the gradual increase or decrease of power load;

[0103] Seasonal component: reflects seasonal or cyclical changes in data, such as increased electricity load in summer;

[0104] Holiday component: Electricity load changes during certain holidays or festivals;

[0105] When training the model, Prophet automatically does the following: Identifies seasonal changes in the data, such as annual fluctuations in electricity load, identifies long-term trends, and handles holiday effects to adjust for fluctuations in electricity demand;

[0106] Prophet uses the Bayesian optimization algorithm to automatically adjust the model's hyperparameters. The optimized model can more accurately predict future power loads. After the model is trained, it can use historical data to predict future power loads. For example, predicting power demand for the next 7 days will generate prediction results for the next 7 days based on historical data, and the output is the power load forecast value for each day in the future.

[0107] The forecast results will be displayed in the form of a chart, including historical data (blue line), which represents past power load data, forecast data (orange line), which represents future power load forecast data, trend line (green line), which represents changes in long-term trends, such as the growth of power load, and seasonal fluctuations (red line), which reflects seasonal changes, such as peak power consumption period.

[0108] S4: By performing preliminary calculations and preprocessing at the acquisition end, the network transmission load is reduced, the response speed is improved, and data is cached when the network is interrupted.

[0109] Preferably, preprocessing is used to reduce the computing burden of the server after receiving the data, speed up the processing and analysis of the data, and improve the system response speed;

[0110] Edge devices perform some calculations locally and return the results directly, ensuring that information can reach the user end quickly and achieve instant feedback.

[0111] In the event of a network outage or insufficient bandwidth, the acquisition end automatically caches the data and stores it in local storage. When the network is restored, the cached data is sent to the server in chronological order to ensure data integrity and continuity.

[0112] S5: Execute remote control instructions based on the analysis results, including load adjustment, power failure control and time-sharing scheduling.

[0113] Preferably, the system continuously analyzes real-time power parameters and combines forecasting and trend analysis to determine the current load status.

[0114] When it is detected that the load is close to the set overload threshold, the system will automatically send instructions to adjust the power of non-critical equipment or systems;

[0115] The system monitors power parameters in real time and combines with anomaly detection function. When an abnormal situation or emergency failure occurs, the system automatically generates a power-off control command and selectively interrupts the power supply through the power switch or relay of the remote control device.

[0116] The system uses historical data and prediction models to analyze the peak and trough periods of electricity demand in a day, and generates a time-sharing scheduling plan based on the analysis system, restricting the operation of certain equipment during peak periods in advance, or scheduling high-energy-consuming tasks to be executed during off-peak periods.

[0117] Furthermore, when the system detects an abnormal situation, it automatically generates a power-off control instruction based on the nature of the fault and the device priority:

[0118] Overload: If the current threshold is exceeded, a power-off command is automatically generated to prevent equipment failure due to overload;

[0119] Voltage abnormality: If the voltage is too high or too low, a power-off command is generated to prevent equipment from being damaged due to unstable voltage;

[0120] Emergency failure: If an emergency such as a short circuit is detected, a power-off command is immediately generated to prioritize system safety.

[0121] S6: Real-time monitoring of the operating status of the power system, timely detection of abnormal conditions and issuance of early warning notifications, and generation of different levels of alarms according to the set thresholds.

[0122] Preferably, the collected data is transmitted to the edge computing module or cloud server through the communication transmission module, and combined with historical data and set operating standards, the health status of the equipment and the overall system is continuously evaluated;

[0123] Through preset threshold rules and machine learning algorithms, the system automatically analyzes various parameters to detect whether there are abnormal values ​​that deviate from the normal range.

[0124] When the set voltage reaches 360V or the current reaches 445A, an information alarm is issued:

[0125] If the voltage drops to 360V, the system generates an informational alarm indicating low voltage;

[0126] If the current drops to 445A, the system generates a low-level alarm for the operator’s attention;

[0127] When the set voltage reaches 352V or the current reaches 435A, a warning level alarm is issued:

[0128] If the voltage fluctuates to 352V and the fluctuation lasts for more than 5 minutes, and the current drops to 435A or rises to 555A, a medium alarm is triggered, indicating a local fault;

[0129] When the set voltage drops to 340V, the current is higher than 575A or the temperature exceeds 90℃, a serious alarm is issued:

[0130] If the voltage drops to 340V, there is a problem with the equipment or the power supply system;

[0131] If the current rises to 575A, there is a risk of overload or line abnormality;

[0132] If the temperature reaches 90°C, there is a risk of high temperature failure.

[0133] S7: User authentication, access rights allocation and management.

[0134] Preferably, the system divides users into different roles, such as ordinary users, operators, administrators and super administrators, and different roles have different permissions:

[0135] The system sets permission levels for different roles. Ordinary users can view data; operators can configure equipment and debug data; administrators can add and delete users and manage the system; and super administrators have full access and control permissions.

[0136] Furthermore, ordinary users log in to the system and view the real-time and historical data of power parameters. After entering the user name and password to log in to the system, the user enters the data viewing interface and chooses to view the power data for the day. The system displays the curve graphs of the power current and voltage for the day and the real-time values ​​of the temperature of each device;

[0137] The operator enters the system and adjusts the sampling frequency of the device from 1 second to 0.5 seconds. The operator enters the user name and password to enter the system. The operator selects a device and clicks "Parameter Settings" to adjust the sampling frequency. The system records that the sampling frequency has changed from 1 second to 0.5 seconds and displays that the parameters have been updated successfully.

[0138] The administrator adds a new user, enters the user name and password, enters the system, and enters the user management interface: selects the "Add User" option, the system confirms that the administrator has user management permissions, and after the administrator confirms the information, clicks "Save". The system records the new operation and time, and sends a notification to the new user.

[0139] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. 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. An intelligent power monitoring and management system based on the Internet of Things, characterized in that: include: Data acquisition module, used to collect power parameters in real time, including current, voltage, power factor, temperature and power consumption, and perform preliminary processing on the collected data; Communication transmission module, used to transmit data from the acquisition end to the server and management platform, supporting remote monitoring, configuration and fault diagnosis; Data processing and analysis module, which is used to clean, format and store the collected data, and use machine learning algorithms for prediction, anomaly detection and trend analysis; Edge computing module, which is used to reduce network transmission load, improve response speed, and cache data when the network is interrupted by performing preliminary calculations and preprocessing at the acquisition end; Intelligent control module, used to execute remote control instructions based on analysis results, including load adjustment, power-off control and time-sharing scheduling; The early warning management module is used to monitor the operating status of the power system in real time, detect abnormal conditions in a timely manner and issue early warning notifications, and generate different levels of alarms according to the set thresholds; The permission control module is used for user authentication, access permission allocation and management.

2. The intelligent power monitoring and management system based on the Internet of Things as claimed in claim 1, characterized in that: The real-time collection of power parameters and preliminary processing of the collected data include: Select the sampling rate according to the dynamic needs of the power system, such as 500ms or 1s intervals. Increase the sampling frequency in a highly dynamic power environment. For data collected at a high frequency, aggregate the data by time period, such as calculating the average, maximum, minimum, and standard deviation of the sampling points every minute. Filter data points that exceed the preset upper or lower limit to eliminate random noise in the acquired signal in order to improve the accuracy of subsequent analysis.

3. The intelligent power monitoring and management system based on the Internet of Things as claimed in claim 1, characterized in that: The data is transmitted from the acquisition end to the server and management platform to support remote monitoring, configuration and fault diagnosis, including: By sending configuration instructions, the frequency, alarm threshold, transmission protocol and other settings of the collection parameters can be changed. The configuration instructions are transmitted in encrypted form and role-based permission control is adopted. The platform receives and analyzes alarm data and performs fault diagnosis on abnormal conditions in the power system.

4. The intelligent power monitoring and management system based on the Internet of Things as claimed in claim 1, characterized in that: The cleaning, formatting and storing of the collected data include: For detecting mutation points, the anomaly is judged based on the average value of the data in the sliding window. Assuming that the sampling data x t The average value in the time window N is μ N , with standard deviation σ N， If x t If the deviation from the mean is k times the standard deviation, it is considered abnormal, that is: ∣x t -m N ∣>k·s N For smoothing data, high-frequency noise components are eliminated. Assuming the original signal is S (t) , the signal after low-pass filtering is S filtered (t), using a moving average filter: Where N is the window size.

5. The intelligent power monitoring and management system based on the Internet of Things as claimed in claim 1, characterized in that: The use of machine learning algorithms for prediction, anomaly detection, and trend analysis includes: Long short-term memory network is used to predict power demand. Historical power data is collected and divided into training set and test set. The number of LSTM layers and units is defined to capture the time series characteristics. The LSTM formula is as follows: Forget Gate: f t =σ(W f ·[h t-1 ,x t ]+b f ) Input Gate: i t =σ(W i ·[h t-1 ,x t ]+b i ) Memory unit update: C t =f t ·C t-1 +i t ·tanh(WC·[h t-1 ,x t ]+b C ) Output Gate: h t =o t ·tanh(C t ) Use historical electricity data for training and define the loss function as mean square error: Among them, n is the total number of samples, y i is the true value of the i-th sample, is the predicted value of the i-th sample; Input the latest power data for prediction and output the power demand for several time steps in the future; The isolation forest algorithm is used to detect outliers in power data. The parameters of the isolation forest are defined as follows: Among them, h(x) is the number of splits of data point x from the root node to the leaf node, s(x,n) is the degree of abnormality measured by path length, and c(n) is the path length normalization constant for the number of samples n; Input real-time power data, calculate the anomaly score of each data point, set a threshold, and when the score exceeds the threshold, mark the data as abnormal and trigger an alarm; The Prophet time series model is used to analyze the trend of power data. The historical data of power load is collected, and the data is decomposed into periodicity and seasonality to capture the changes on different time scales and construct the Prophet model: The Prophet model decomposes the time series into trend, seasonality, and holiday components: y(t)=g(t)+s(t)+h(t)+∈t Among them, g(t) is the trend component, s(t) is the seasonal component, and h(t) is the holiday component; Assuming the trend is linear growth or saturated growth, for a linear trend: g(t)=(k+a(t))t+b Among them, k is the growth rate and b is the offset; Through historical data, the Prophet model is fitted and the hyperparameters are adjusted using Bayesian optimization to improve the prediction effect.

6. The intelligent power monitoring and management system based on the Internet of Things as claimed in claim 1, characterized in that: The method of performing preliminary calculation and preprocessing at the acquisition end to reduce the network transmission load, improve the response speed, and cache data when the network is interrupted includes: Through preprocessing, the server's computing burden after receiving data is reduced, data processing and analysis are accelerated, and system response speed is improved; Edge devices perform some calculations locally and return the results directly, ensuring that information can reach the user end quickly and achieve instant feedback. In the event of a network outage or insufficient bandwidth, the acquisition end automatically caches the data and stores it in local storage. When the network is restored, the cached data is sent to the server in chronological order to ensure data integrity and continuity.

7. The intelligent power monitoring and management system based on the Internet of Things as claimed in claim 1, characterized in that: The remote control instructions are executed based on the analysis results, including load adjustment, power-off control and time-sharing scheduling, including: The system continuously analyzes real-time power parameters and combines forecasting and trend analysis to determine the current load status. When it detects that the load is approaching the set overload threshold, the system automatically sends instructions to adjust the power of non-critical equipment or systems; The system monitors power parameters in real time and combines with anomaly detection function. When an abnormal situation or emergency failure occurs, the system automatically generates a power-off control command and selectively interrupts the power supply through the power switch or relay of the remote control device. The system uses historical data and prediction models to analyze the peak and trough periods of electricity demand in a day, and generates a time-sharing scheduling plan based on the analysis system, restricting the operation of certain equipment during peak periods in advance, or scheduling high-energy-consuming tasks to be executed during off-peak periods.

8. The intelligent power monitoring and management system based on the Internet of Things as claimed in claim 1, characterized in that: The real-time monitoring of the operating status of the power system, timely detection of abnormal conditions and issuance of early warning notifications include: The collected data is transmitted to the edge computing module or cloud server through the communication transmission module, and combined with historical data and set operating standards, the health status of the equipment and the overall system is continuously evaluated; Through preset threshold rules and machine learning algorithms, the system automatically analyzes various parameters to detect whether there are abnormal values ​​that deviate from the normal range.

9. The intelligent power monitoring and management system based on the Internet of Things as claimed in claim 1, characterized in that: Generating different levels of alarms according to the set thresholds includes: When the set voltage reaches 360V or the current reaches 445A, an information alarm is issued: If the voltage drops to 360V, the system generates an informational alarm indicating low voltage; If the current drops to 445A, the system generates a low-level alarm for the operator’s attention; When the set voltage reaches 352V or the current reaches 435A, a warning level alarm is issued: If the voltage fluctuates to 352V and the fluctuation lasts for more than 5 minutes, and the current drops to 435A or rises to 555A, a medium alarm is triggered, indicating a local fault; When the set voltage drops to 340V, the current is higher than 575A or the temperature exceeds 90℃, a serious alarm is issued: If the voltage drops to 340V, there is a problem with the equipment or the power supply system; If the current rises to 575A, there is a risk of overload or line abnormality; If the temperature reaches 90°C, there is a risk of high temperature failure.

10. The intelligent power monitoring and management system based on the Internet of Things according to claim 1, characterized in that: The user identity authentication, access rights allocation and management include: The system divides users into different roles, such as common users, operators, administrators, and super administrators. Different roles have different permissions: The system sets permission levels for different roles. Ordinary users can view data; operators can configure equipment and debug data; administrators can add and delete users and manage the system; and super administrators have full access and control permissions.

Images