Intelligent power distribution monitoring system based on Internet of Things

Abstract

The invention discloses an intelligent power distribution monitoring system based on the Internet of Things, and relates to the technical field of energy monitoring, and the system comprises the following components: a distributed energy data collection module employs a composite sensor fusion networking design, and integrates power, environmental parameters and an electric energy quality sensor to form a compact whole; according to the invention, a hybrid intelligent decision-making algorithm integrating multi-source data driving and model prediction control is adopted, so that massive actual operation data output by the data processing and analysis module is fully utilized to carry out deep mining analysis; accurate in-situ energy consumption scheduling is realized, and in combination with a micro-grid and a distributed energy dynamic prediction model constructed based on a physical model, the energy generation amount, the user power consumption demand and the local power grid state change in a short time in the future are accurately predicted.

Description

Technical Field

[0001] The present invention relates to the field of energy monitoring technology, and in particular to an intelligent power distribution monitoring system based on the Internet of Things. Background Art

[0002] With the transformation of the global energy structure and the rapid development of renewable energy technology, the proportion of distributed energy (such as solar energy, wind energy, etc.) in the energy supply system is increasing. The efficient utilization and management of distributed energy is of great significance to promoting sustainable energy development, improving energy utilization efficiency and ensuring the stable operation of the power grid. Therefore, the development of a monitoring system and method that can achieve efficient and intelligent management of distributed energy in microgrids and stable and optimized operation of distribution systems has become a research hotspot in the current field of energy technology.

[0003] Traditional power distribution monitoring systems have many shortcomings when dealing with distributed energy management. First, in terms of data collection, traditional distributed sensor collection methods are prone to time differences, and the subsequent data fusion processing process is complicated, which affects the accuracy and real-time nature of the data. Secondly, data processing mostly relies on fixed algorithms, which are difficult to adapt to dynamic situations such as the expansion of the energy system scale, equipment upgrades, and changes in user electricity consumption behavior. Furthermore, scheduling decision-making methods often rely too much on historical data or are limited by model assumptions, resulting in insufficient accuracy of scheduling strategies and difficulty in coping with complex and changeable energy system environments. In addition, the management and control methods of power storage and transmission are not intelligent and flexible enough, which affects the balance and stable power supply of electricity. These problems seriously restrict the effective utilization of distributed energy and the stable operation of the distribution system.

[0004] In view of the above problems, it is necessary to optimize the existing intelligent distribution monitoring system based on the Internet of Things. By adopting composite sensor fusion networking technology, adaptive data processing framework based on artificial intelligence deep learning, and hybrid intelligent decision-making algorithm innovation technology integrating multi-source data drive and model predictive control, efficient intelligent management of distributed energy in microgrids and stable and optimized operation of distribution systems can be achieved. Therefore, it is of great significance to develop an intelligent distribution monitoring system based on the Internet of Things that can comprehensively realize the above characteristics. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the prior art and provide an intelligent power distribution monitoring system based on the Internet of Things. It can realize real-time synchronous collection and preliminary fusion processing of data by adopting composite sensor fusion networking technology, thereby improving the quality and efficiency of data collection. It introduces an adaptive data processing framework based on artificial intelligence deep learning, enhances the adaptability and flexibility of data processing, and uses a hybrid intelligent decision-making algorithm that integrates multi-source data drive and model predictive control to optimize the accuracy and flexibility of scheduling decisions. It develops an intelligent two-way energy flow coordination management mechanism, improves the intelligence and stability of power management and control, and at the same time, through user interaction and display modules and remote monitoring and management platforms, convenient management and operation of the system are realized.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: an intelligent power distribution monitoring system based on the Internet of Things, the system includes the following components: a distributed energy data acquisition module, a data processing and analysis module, a comprehensive scheduling and optimization decision module, an electric energy storage and transmission control module and a user interaction and display module;

[0007] The distributed energy data acquisition module adopts a composite sensor fusion networking design, integrating power, environmental parameters and power quality sensors to form a compact whole, and has a built-in intelligent chip to realize real-time synchronous acquisition and preliminary fusion processing of each sensor data. It is installed at the distributed energy generation point and user end according to the deployment strategy, and selects wired or wireless communication methods according to location and needs, and sends the processed data to the data processing and analysis module according to the set cycle or event trigger mechanism;

[0008] The data processing and analysis module operates by building an adaptive data processing framework, setting up a data input interface to verify received data, establishing a hierarchical historical database to store data on different energy sources and user electricity consumption characteristics, and generating cleaning, format unification and feature extraction rules in real time based on data characteristics, and dynamically adjusting rules as the system changes, and processing input data according to the rules before transmitting them to the comprehensive scheduling and optimization decision module;

[0009] The comprehensive scheduling and optimization decision-making module includes a multi-source data-driven submodule and a model predictive control submodule, wherein the multi-source data-driven module constructs and establishes multiple data access ports, respectively receiving processed multi-dimensional data output from the data processing and analysis module, and using data analysis technology to deeply mine the integrated data set to discover the inherent correlation and potential laws between different data elements. According to these mined relationships and laws, a data-driven energy consumption scheduling model is constructed. The model uses the power distribution plan as the decision variable, and considers various actual operating constraints at the same time. The optimization algorithm is used to solve the optimal power distribution plan that meets the current system operating state. For the model predictive control submodule, a microgrid and distributed energy dynamic prediction model is constructed based on physical principles and system operating characteristics. For the solar power generation part, according to the photoelectric conversion theory of photovoltaic cells, the light intensity prediction model and the temperature impact model of power generation efficiency, a solar power generation power is established at any time. A prediction model for time-varying changes is developed. For wind power generation, a wind power prediction model is constructed by combining aerodynamic principles, wind speed prediction models, and power curve characteristics of wind turbines. At the same time, a comprehensive dynamic prediction model covering energy generation, user demand, and grid status is formed by considering the changing laws of user electricity demand and changes in the state of the local power grid. The dynamic prediction model is used to accurately predict the energy generation, user electricity demand, and changes in the state of the local power grid in a short period of time in the future, and a prediction result data set is obtained. The prediction result data set is integrated with the current real-time data set obtained in the multi-source data-driven submodule, and then input into the optimization decision model again. The preliminary decision on local energy consumption scheduling obtained based on data-driven is corrected and improved. At the same time, a dynamic self-organizing energy optimization strategy for edge nodes is generated, and the generated comprehensive scheduling and optimization strategy containing power distribution instructions and edge node optimization parameters is sent to the power storage and transmission control module and each edge node respectively;

[0010] The electric energy storage and transmission control module relies on an intelligent two-way energy flow coordination management mechanism. The electric energy interaction state perception module deploys multiple types of sensors to collect relevant data, integrates and evaluates the electric energy interaction state, and then feeds back to the intelligent coordination control module. By receiving and analyzing the instructions of the comprehensive scheduling and optimization decision-making module, the charging strategy is formulated according to the state of the energy storage device during charging, and the strategy is formulated based on multiple factors to select equipment and adjust parameters during power transmission. During transmission, the energy flow distribution is optimized according to real-time feedback to ensure safe and efficient electric energy storage and transmission and electric energy balance of the distribution system;

[0011] The user interaction and display module creates an interactive interface for system managers, operation and maintenance personnel and some authorized users, which can display in real time the operating data of each distributed energy generation point, user electricity consumption, the current comprehensive scheduling and optimization strategy and execution effect, and the self-organizing optimization status information of the edge nodes. Users can also use this interface to remotely monitor and set early warning thresholds to intervene in and manage system operations.

[0012] Furthermore, the distributed energy data acquisition module has a built-in intelligent chip to realize real-time synchronous acquisition and preliminary fusion processing of each sensor data, and its fusion formula is: ref =k 1 ×I×(1+k 2 ×(TT std ))×A, where P ref It represents the reference power generation, that is, the power generation that should be generated theoretically according to the current light intensity and temperature. It is the reference value of the reasonable power range estimated after fusion processing. 1 is the photoelectric conversion coefficient, I is the light intensity, k 2 is the temperature influence coefficient, T is the real-time collected photovoltaic panel surface temperature, T std It is the standard temperature reference value, which is used to measure the deviation of the current temperature from the standard state. A is the effective light-receiving area of ​​the photovoltaic panel.

[0013] Furthermore, in the comprehensive scheduling and optimization decision-making module, the multi-source data-driven sub-module constructs and establishes multiple data access ports, which respectively receive the processed multi-dimensional data output from the data processing and analysis module, including distributed energy generation related data, user electricity demand data and local power grid status data. The distributed energy generation related data includes the real-time power generation of each energy generation point, the change trend of power generation over time and the correlation with environmental factors. The user power demand data includes the real-time power consumption of different regions and different types of users, the power consumption fluctuation law and the peak and valley periods of electricity consumption. The local power grid status data includes the power grid topology, the voltage and current conditions of each node and the line load rate.

[0014] Furthermore, in the comprehensive scheduling and optimization decision module, the multi-source data driven submodule uses data analysis technology to conduct in-depth mining of the integrated data set to discover the inherent correlation and potential laws between different data elements. According to these mined relationships and laws, a data-driven energy consumption scheduling model is constructed, and its model formula is: The constraints are: Among them, x ijrepresents the electric power distributed from distributed energy generation point i to user j. Its optimal value is determined by solving the model to achieve a reasonable electric energy distribution plan. I is the set of distributed energy generation points, J is the user set, covering different regions and different types of electricity users, c ij is the unit power transmission loss coefficient from energy generation point i to user j, is the maximum power generation of distributed energy generation point i, which indicates the maximum power limit that the energy generation point can output under current conditions. It is a constraint condition determined by the performance of the energy equipment itself and environmental factors. j is the power demand of user j.

[0015] Furthermore, in the comprehensive scheduling and optimization decision module, for the model predictive control submodule, a dynamic prediction model of microgrid and distributed energy is constructed based on physical principles and system operation characteristics. For the solar power generation part, the prediction model of the solar power generation power changing with time is

[0016] Among them, P solar (t+Δt) represents the predicted value of solar power generation at the future time t+Δt, β 0 is the intercept term, β 1 is the light intensity influence coefficient, β 2 is the temperature influence coefficient, T ref is the reference temperature, β 3 is the influence coefficient of lagged light intensity, γ i is the periodic term coefficient, ω i is the angular frequency of the corresponding period, φ i is the phase angle, ∈ is the random error term, and for wind power generation, the wind power generation power prediction model is: Among them, P wind (t+Δt) represents the predicted wind power generation value at the future time t+Δt, ρ is the air density, A is the swept area of ​​the wind turbine blade, v(t+Δt) is the predicted wind speed at the future time t+Δt, C p (λ, β) is the power coefficient of the wind turbine, η is the comprehensive transmission and power generation efficiency of the entire wind power generation system, and its user electricity demand prediction model is:

[0017] Among them, D user (t+Δt) represents the predicted power demand of a specific user or user group at the future time t+Δt, μ is the long-term average level of power demand, α j(j=1,2,…,m) is the influence coefficient of working day and rest day, d j (t) is a binary variable indicating whether the current time t is a working day or a holiday, β k (k=1,2,…;p) is the seasonal influence coefficient, s k (t) is a categorical variable indicating the current season, ·η is the influence coefficient of the lag term of historical electricity consumption data, and f l (t-τ l ) is the lag τ l The time step, ∈′, is the random error term of the user power demand forecast, and the local power grid state change prediction model is: That

[0018] In, V grid (t+Δt) represents the voltage forecast value of the key node of the local power grid at the future time t+Δt, V grid (t) is the actual voltage value at the current time t, which is obtained through real-time monitoring of the power grid, α i is the active power change influence coefficient, ΔP i (t) represents the change in active power at each distributed energy generation point or load at the current time t, β j is the reactive power variation influence coefficient, ΔQ j (t) is the change in reactive power of each node at the current time t, γ k is the current change influence coefficient, ΔI k (t) is the change in current on the key line at the current moment t, ∈″ is the random error term of the grid voltage prediction. The prediction model results of the above parts are reasonably integrated through data fusion. According to their respective importance and mutual correlation, a unified comprehensive dynamic prediction result covering energy generation, user demand and grid status is formed for subsequent real-time data fusion and scheduling decision-making operations.

[0019] Furthermore, in the comprehensive dispatching and optimization decision module, a unified comprehensive dynamic prediction model covering energy generation, user demand and grid status is formed according to the respective importance and mutual correlation. The algorithm formula of the model is Y(t+Δt)=ω solar ×P solar (t+Δt)+ω wind ×P wind (t+Δt)+ω wer ×D wer (t+Δt)+ω grid ×S grid ((t+Δt), where Y(t+Δt) represents the comprehensive prediction index vector at the future time t+Δt, Δt is the time step of the prediction, ωsolar、 ω wind、 ω user、 ω gridz are the weight coefficients of solar power generation, wind power generation, user power demand, and local grid voltage status in the comprehensive prediction model, respectively. i are other influencing factors, ∈ tatal is the random error term of the comprehensive prediction model.

[0020] Furthermore, in the electric energy storage and transmission control module, a strategy is formulated to control the charging power according to the state of the energy storage device during charging, and the charging power adjustment algorithm is: Among them, P charge Indicates the recommended charging power, V bat is the current terminal voltage of the energy storage device, R int is the current internal resistance of the energy storage device, λ is the weight coefficient of the impact of internal resistance on the remaining life, L remain is the remaining life of the energy storage device.

[0021] Furthermore, in the power storage and transmission control module, a strategy is formulated to select equipment and adjust parameters based on multiple factors during power transmission. In order to select the most suitable equipment combination and adjust its working parameters to achieve stable and efficient transmission, an equipment selection and parameter adjustment algorithm based on comprehensive evaluation of power quality and transmission loss is adopted, and its formula is: Among them, S opt represents the optimal transformer ratio selection result, ω 1 ,ω 2 ,ω 3 They are the weight coefficients of voltage deviation, power loss and total harmonic distortion respectively. ΔU9S) is the voltage deviation corresponding to the transformer ratio of S, ΔP9S) is the power loss corresponding to the transformer ratio of S, and THD9S) is the total harmonic distortion corresponding to the transformer ratio of S.

[0022] Compared with the existing technology, the intelligent power distribution monitoring system based on the Internet of Things has the following beneficial effects:

[0023] 1. The present invention uses a hybrid intelligent decision-making algorithm that integrates multi-source data drive and model predictive control. It not only makes full use of the massive actual operation data output by the data processing and analysis module for in-depth mining and analysis, and realizes accurate on-site energy consumption scheduling, but also combines the microgrid and distributed energy dynamic prediction model built based on the physical model to accurately predict the energy generation, user electricity demand and local power grid status changes in the short term in the future. This hybrid approach overcomes the limitations of pure data drive or model prediction methods, so that system scheduling decisions can not only respond to current system changes in real time, but also plan in advance to deal with upcoming complex situations, thereby significantly enhancing the intelligence and flexibility of system scheduling decisions and ensuring the stable operation and efficient management of the distribution system.

[0024] 2. Through the composite sensor fusion networking technology adopted in the distributed energy data acquisition module of the present invention, the system can synchronously collect and preliminarily fuse and process various distributed energy and user electricity consumption data in real time, avoiding the time difference and subsequent complex fusion processing problems existing in traditional distributed sensor data collection. At the same time, the data processing and analysis module dynamically generates targeted data cleaning, format unification and feature extraction rules based on the historical data of different distributed energy types and user electricity consumption characteristics, thereby significantly improving the accuracy and efficiency of data collection and processing, and providing a solid data foundation for subsequent comprehensive scheduling and optimization decisions.

[0025] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0027] Figure 1 This is the process operation diagram of the intelligent power distribution monitoring system based on the Internet of Things.

[0028] Figure 2 Flowchart of the intelligent power distribution monitoring system based on the Internet of Things. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] Embodiment 1

[0031] This embodiment describes in detail the application of the intelligent power distribution monitoring system based on the Internet of Things in the power distribution monitoring of microgrids in small industrial parks. Through the present invention, data accuracy, energy utilization, and power supply reliability are significantly improved, and operating costs are greatly reduced, providing a solid guarantee for the stable and efficient operation of microgrids in small industrial parks.

[0032] First, in the small industrial park, the distributed energy data acquisition module carefully installs a composite sensor at the solar photovoltaic panel array on the roof of each factory building. The sensor cleverly integrates the power, light intensity, temperature and power quality monitoring functions, and the built-in smart chip realizes real-time synchronous collection and preliminary fusion processing of various data. The fusion formula is p ref =k 1 ×I×(1+k 2 ×(TT std ))×A, where p ref represents the reference power generation, that is, the power generation that should be generated theoretically according to the current light intensity and temperature, k 1 is the photoelectric conversion coefficient, I is the light intensity, k 2 is the temperature influence coefficient, T is the real-time collected photovoltaic panel surface temperature, T std is the standard temperature reference value (usually set to the temperature of the photovoltaic panel under standard test conditions, such as 25°C), A is the effective light-receiving area of ​​the photovoltaic panel. For example, at a certain moment, the light intensity I is 800 lux, the temperature T is 30°C, and the photoelectric conversion coefficient k is 1 is 0.15, the temperature influence coefficient k 2 is -0.005, the effective light receiving area A of the photovoltaic panel is 10 square meters, and the reference power generation power p can be calculated by the formula ref This value can be used as an important basis for judging whether the actual power generation is abnormal. A composite sensor is also installed at the distribution box in the park to collect power consumption, voltage, current and other data of each enterprise user. The collected data is stably transmitted to the data processing and analysis module via wired Ethernet.

[0033] After the data processing and analysis module receives the data, the adaptive data processing framework based on artificial intelligence deep learning generates corresponding data processing rules according to the rich historical data of solar power generation in the park and the electricity consumption characteristics of enterprises. The framework uses a large amount of historical data for training, so that it can generate cleaning, format unification and feature extraction rules in real time according to the characteristics of the data, and dynamically adjust the rules as the system changes. For example, when an enterprise in the park upgrades its equipment and the electricity consumption pattern changes, the framework can automatically identify and adjust the data processing rules. During the data cleaning process, it can identify and eliminate abnormal power generation data caused by partial shading of photovoltaic panels or equipment failure according to the set threshold (such as a certain fluctuation range of reference power generation), and then transmit the processed data to the comprehensive scheduling and optimization decision module.

[0034] The comprehensive scheduling and optimization decision-making module uses the multi-source data-driven sub-module to receive and integrate data and build an energy consumption scheduling model. The model formula is: The constraints are where x ij represents the electric power distributed from distributed energy generation point i to user j, where I is the set of distributed energy generation points, J is the set of users, and c ij is the unit power transmission loss coefficient from energy generation point i to user j, is the maximum power generation of distributed energy generation point i, D j is the power demand of user j. Considering the diversity of production of enterprises in the park, for example, if an enterprise consumes a lot of electricity during the day and it coincides with the peak period of solar power generation, the model calculation is used to prioritize the allocation of solar power to the enterprise to achieve efficient use of electricity. At the same time, the model prediction control submodule is based on the constructed dynamic prediction model and combines the weather forecast to predict the change of solar power generation in the next few hours (such as according to the light intensity prediction model where p solar (t+Δt) represents the predicted value of solar power generation at the future time t+Δt, β 0 is the intercept term, β 1 is the light intensity influence coefficient, β 2 is the temperature influence coefficient, T ref is the reference temperature, β 3 is the influence coefficient of lagged light intensity, y i is the periodic coefficient w i is the angular frequency of the corresponding period, is the phase angle, E is the random error term, and the cloud movement in the weather forecast is combined to estimate the change in light intensity) and the company's production plan (for example, if it is known that a company will have large-scale equipment maintenance in the afternoon and the electricity demand will decrease), the scheduling strategy is optimized and adjusted. For example, if it is predicted that the light intensity will weaken in the next two hours, and the production task of a company is about to end and the electricity demand will decrease, the model will adjust the electricity distribution in advance and store or distribute the excess electricity to other companies in need.

[0035] The power storage and transmission control module uses the charging power adjustment algorithm based on the comprehensive dispatching instructions and the current terminal voltage, internal resistance and remaining life of the energy storage device (such as the remaining life estimated based on the battery internal resistance data and remaining power provided by the battery management system). Reasonably control the charging power, where p c harge indicates the recommended charging power, v bat is the current terminal voltage of the energy storage device, R int is the current internal resistance of the energy storage device, λ is the weight coefficient of the impact of internal resistance on the remaining life, L remain is the remaining life of the energy storage device. Assuming that the current terminal voltage of the energy storage device is v bat is 480 volts, internal resistance R iπt The weight coefficient λ of the influence of the internal resistance on the remaining life is 0.5 ohms, and the remaining life L remain is 0.7, then the appropriate charging power p can be calculated charge , ensuring the safe and efficient charging process, and when delivering electricity to enterprises, through equipment selection and parameter adjustment algorithms based on comprehensive evaluation of power quality and transmission loss Where S opt represents the optimal transformer ratio selection result, w 1 、w 2 、w 3 They are the weight coefficients of voltage deviation, power loss, and total harmonic distortion. ΔU(S) is the voltage deviation corresponding to the transformer ratio S, Δp(S) is the power loss corresponding to the transformer ratio S, and THD(S) is the total harmonic distortion corresponding to the transformer ratio S. Selecting an appropriate transformer ratio (such as based on the requirements of the enterprise's electrical equipment for voltage stability and the current grid line load) ensures stable and efficient transmission of electric energy. For example, an enterprise has high requirements for voltage stability. 1 The weight is larger, and the system will give priority to selecting the transformer ratio that can minimize the voltage deviation, while taking into account factors such as power loss and total harmonic distortion.

[0036] The user interaction and display module is set up in the power distribution room of the park. Operation and maintenance personnel can use this interface to view the power consumption of each enterprise, solar power generation power, the execution effect of the current scheduling strategy and other information in real time. They can also set early warning thresholds. For example, when the power consumption of an enterprise suddenly exceeds the set threshold, an alarm will be issued to timely troubleshoot or adjust the production plan. The remote monitoring and management platform allows the park management department to remotely monitor the operating status of the entire microgrid system in the office, and perform remote parameter adjustments and system upgrades.

[0037] In summary, by adopting the present invention, the operating cost of the park is greatly reduced, providing a solid guarantee for the stable and efficient operation of the microgrid in the small industrial park, and effectively promoting the effective utilization and sustainable development of distributed energy in the park.

[0038] Embodiment 2

[0039] This embodiment describes in detail the application of the intelligent distribution monitoring system based on the Internet of Things in the distribution monitoring of rural distributed energy microgrids. Through the present invention, the effective utilization and sustainable development of rural distributed energy are effectively promoted, and a stable, efficient and intelligent distribution monitoring solution is provided for rural areas.

[0040] In rural areas, the distributed energy data acquisition module installs a composite sensor on the wind turbine tower at the village entrance. The sensor integrates wind speed, wind direction, power generation and power quality monitoring functions. The built-in smart chip realizes real-time synchronous collection and preliminary fusion processing of various data. Smart meter sensors are installed at farmers' meter boxes to collect farmers' electricity consumption data. Taking into account the characteristics of the rural environment, some sensors use Zigbee wireless communication to transmit data to the data processing and analysis module. For example, the composite sensor on the wind turbine tower can accurately collect environmental data such as wind speed and wind direction, as well as the real-time power generation and power quality information of the wind turbine. These data are stably transmitted through the Zigbee network to ensure the timeliness and accuracy of the data.

[0041] The data processing and analysis module uses an adaptive data processing framework to process the collected data based on the rich historical data characteristics of rural wind power generation and residential electricity consumption (such as the seasonal and time-based patterns of residential electricity consumption, peak electricity consumption at night in summer, etc.). The framework uses a deep learning model to analyze a large amount of historical data and can automatically generate data processing rules that are adapted to the characteristics of rural electricity consumption. For example, when cleaning data, based on the relationship model between wind speed and power generation, it identifies and eliminates the impact of abnormal wind speed data caused by extreme weather on power generation calculations, and then sends the processed data to the comprehensive scheduling and optimization decision-making module.

[0042] The multi-source data-driven submodule of the comprehensive scheduling and optimization decision-making module analyzes and integrates data to build an energy consumption scheduling model suitable for rural areas, giving priority to residents' daily electricity consumption (such as reasonably allocating wind power generation and energy storage electricity according to the peak hours of residents' electricity consumption). The model prediction control submodule is based on the wind power generation power prediction model (taking into account the seasonal variation of wind speed and short-term weather forecasts) and the residents' electricity demand prediction model (combined with local agricultural activities and living habits) to predict future energy supply and demand. For example, the wind power generation power prediction model where p uind (t+Δt) represents the predicted wind power generation value at the future time t+Δt, P is the air density, A is the swept area of ​​the wind turbine blade, v(t+Δt) is the predicted wind speed at the future time t+Δt, C 卩 (λ,β) is the power coefficient of the wind turbine, λ is the tip speed ratio, β is the pitch angle, η is the comprehensive transmission and power generation efficiency of the entire wind power generation system, and the residential electricity demand prediction model Where D user (t+Δt) represents the predicted power demand value of a certain user or user group at the future time t+Δt, μ is the long-term average level of power demand, αj (j=1,2,…,m) is the influence coefficient of working days and rest days, d j (t) is a binary variable indicating whether the current time t is a working day or a holiday, β k (k=1,2,…;p) is the seasonal influence coefficient, s k (t) is the classification variable representing the current season, yl (l = 1, 2, ..., q) is the influence coefficient of the lag term of historical electricity consumption data, and f l (t-τ l ) is the lag πη time step, E′ is the random error term of the user electricity demand forecast. For example, if it is predicted that the wind is weak at night in winter and the residents’ electricity demand increases (because it is cold in winter and residents use heating equipment frequently), the energy storage equipment is arranged to discharge and supplement electricity in advance to ensure the normal electricity consumption of residents.

[0043] When charging, the energy storage and transmission control module uses the charging power adjustment algorithm according to the state of the energy storage battery (such as estimating the remaining life through internal resistance monitoring and remaining power). Reasonably control the charging power, where p charge Indicates the recommended charging power, v bat is the current terminal voltage of the energy storage device, R int is the current internal resistance of the energy storage device, λ is the weight coefficient of the impact of internal resistance on the remaining life, L remain is the remaining life of the energy storage device. Assume that at a certain moment, the voltage at the end of the energy storage device v batis 400 volts, the internal resistance R int is 0.6 ohms, the weight coefficient of the influence of internal resistance on the remaining life is λ is 0.15, and the remaining life L remain is 0.6, and the appropriate charging power p is calculated by the formula charge , ensuring the safe and efficient charging process. When transmitting electricity, according to the characteristics of rural power grids, through equipment selection and parameter adjustment algorithms where s opt Indicates the optimal transformer ratio selection result, W 1 , W 2 , W 3 They are the weight coefficients of voltage deviation, power loss, and total harmonic distortion. ΔU(S) is the voltage deviation corresponding to the transformer ratio of S, ΔP(S) is the power loss corresponding to the transformer ratio of S, and THD(S) is the total harmonic distortion corresponding to the transformer ratio of S. Selecting a suitable combination of transformer ratio and switchgear can ensure stable transmission of electric energy and reduce damage to rural electrical equipment caused by voltage fluctuations. For example, considering the aging of lines in some rural areas and the large voltage drop, the voltage deviation weight coefficient w will be increased when selecting the transformer ratio. 1 To ensure that the voltage at the user end is stable within a reasonable range.

[0044] The user interaction and display module is set up in the village power service station. The staff can use this interface to understand the energy usage of the entire village. Villagers can also view their own electricity usage information through authorization. The remote monitoring and management platform facilitates the power department to remotely monitor the operation of rural microgrids and promptly discover and deal with problems, such as remotely diagnosing sensor failures or adjusting scheduling strategies to adapt to special rural electricity demand (such as a significant increase in electricity consumption during the Spring Festival). For example, during the Spring Festival, rural residents' electricity consumption will increase significantly due to reunions and celebrations. The power department adjusts the scheduling strategy in advance through the remote monitoring platform to increase energy supply.

[0045] To sum up, through the present invention, the effective utilization and sustainable development of rural distributed energy can be effectively promoted, stable, efficient and intelligent distribution monitoring solutions can be provided for rural areas, the rural electricity consumption situation can be significantly improved, and rural energy construction and development can be promoted.

[0046] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. The intelligent power distribution monitoring system based on the Internet of Things is characterized by: The system includes the following components: distributed energy data acquisition module, data processing and analysis module, comprehensive scheduling and optimization decision module, electric energy storage and transmission control module and user interaction and display module; The distributed energy data acquisition module adopts a composite sensor fusion networking design, integrating power, environmental parameters and power quality sensors to form a compact whole, and has a built-in intelligent chip to realize real-time synchronous acquisition and preliminary fusion processing of each sensor data. It is installed at the distributed energy generation point and user end according to the deployment strategy, and selects wired or wireless communication methods according to location and needs, and sends the processed data to the data processing and analysis module according to the set cycle or event trigger mechanism; The data processing and analysis module operates by building an adaptive data processing framework, setting up a data input interface to verify received data, establishing a hierarchical historical database to store data on different energy sources and user electricity consumption characteristics, and generating cleaning, format unification and feature extraction rules in real time based on data characteristics, and dynamically adjusting rules as the system changes, and processing input data according to the rules before transmitting them to the comprehensive scheduling and optimization decision module; The comprehensive scheduling and optimization decision-making module includes a multi-source data-driven submodule and a model predictive control submodule, wherein the multi-source data-driven module constructs and establishes multiple data access ports, respectively receiving processed multi-dimensional data output from the data processing and analysis module, and using data analysis technology to deeply mine the integrated data set to discover the inherent correlation and potential laws between different data elements. According to these mined relationships and laws, a data-driven energy consumption scheduling model is constructed. The model uses the power distribution plan as the decision variable, and considers various actual operating constraints at the same time. The optimization algorithm is used to solve the optimal power distribution plan that meets the current system operating state. For the model predictive control submodule, a microgrid and distributed energy dynamic prediction model is constructed based on physical principles and system operating characteristics. For the solar power generation part, according to the photoelectric conversion theory of photovoltaic cells, the light intensity prediction model and the temperature impact model of power generation efficiency, a solar power generation power is established at any time. A prediction model for time-varying changes is developed. For wind power generation, a wind power prediction model is constructed by combining aerodynamic principles, wind speed prediction models, and power curve characteristics of wind turbines. At the same time, a comprehensive dynamic prediction model covering energy generation, user demand, and grid status is formed by considering the changing laws of user electricity demand and changes in the state of the local power grid. The dynamic prediction model is used to accurately predict the energy generation, user electricity demand, and changes in the state of the local power grid in a short period of time in the future, and a prediction result data set is obtained. The prediction result data set is integrated with the current real-time data set obtained in the multi-source data-driven submodule, and then input into the optimization decision model again. The preliminary decision on local energy consumption scheduling obtained based on data-driven is corrected and improved. At the same time, a dynamic self-organizing energy optimization strategy for edge nodes is generated, and the generated comprehensive scheduling and optimization strategy containing power distribution instructions and edge node optimization parameters is sent to the power storage and transmission control module and each edge node respectively; The electric energy storage and transmission control module relies on an intelligent two-way energy flow coordination management mechanism. The electric energy interaction state perception module deploys multiple types of sensors to collect relevant data, integrates and evaluates the electric energy interaction state, and then feeds back to the intelligent coordination control module. By receiving and analyzing the instructions of the comprehensive scheduling and optimization decision-making module, the charging strategy is formulated according to the state of the energy storage device during charging, and the strategy is formulated based on multiple factors to select equipment and adjust parameters during power transmission. During transmission, the energy flow distribution is optimized according to real-time feedback to ensure safe and efficient electric energy storage and transmission and electric energy balance of the distribution system; The user interaction and display module creates an interactive interface for system managers, operation and maintenance personnel and some authorized users, which can display in real time the operating data of each distributed energy generation point, user electricity consumption, the current comprehensive scheduling and optimization strategy and execution effect, and the self-organizing optimization status information of the edge nodes. Users can also use this interface to remotely monitor and set early warning thresholds to intervene in and manage system operations.

2. The intelligent power distribution monitoring system based on the Internet of Things according to claim 1 is characterized in that: The distributed energy data acquisition module has a built-in intelligent chip to realize real-time synchronous acquisition and preliminary fusion processing of each sensor data, and its fusion formula is: ref =k1×I×(1+k2×(TT std ))×A, where p ref It represents the reference power generation, that is, the power generation that should be generated theoretically according to the current light intensity and temperature. It is a reference value of the reasonable power range estimated after fusion processing. k1 is the photoelectric conversion coefficient, I is the light intensity, k2 is the temperature influence coefficient, T is the real-time collected surface temperature of the photovoltaic panel, and T std It is the standard temperature reference value, which is used to measure the deviation of the current temperature from the standard state. A is the effective light-receiving area of ​​the photovoltaic panel.

3. The intelligent power distribution monitoring system based on the Internet of Things according to claim 1 is characterized in that: In the comprehensive scheduling and optimization decision-making module, a multi-source data-driven sub-module is constructed to establish multiple data access ports, which respectively receive the processed multi-dimensional data output from the data processing and analysis module, including distributed energy generation related data, user electricity demand data and local power grid status data. The distributed energy generation related data includes the real-time power generation of each energy generation point, the change trend of power generation over time and the correlation with environmental factors. The user power demand data includes the real-time power consumption of different regions and different types of users, the power consumption fluctuation law and the peak and valley periods of electricity consumption. The local power grid status data includes the power grid topology, the voltage and current conditions of each node and the line load rate.

4. The intelligent power distribution monitoring system based on the Internet of Things according to claim 1 is characterized in that: In the comprehensive scheduling and optimization decision module, the multi-source data driven submodule uses data analysis technology to conduct in-depth mining of the integrated data set to discover the inherent correlation and potential laws between different data elements. Based on these mined relationships and laws, a data-driven energy consumption scheduling model is constructed. The model formula is: The constraints are: Among them, x ij represents the electric power distributed from distributed energy generation point i to user j. Its optimal value is determined by solving the model to achieve a reasonable electric energy distribution plan. I is the set of distributed energy generation points, J is the user set, covering different regions and different types of electricity users, c ij is the unit power transmission loss coefficient from energy generation point i to user j, is the maximum power generation of distributed energy generation point i, which indicates the maximum power limit that the energy generation point can output under current conditions. It is a constraint condition determined by the performance of the energy equipment itself and environmental factors. j is the power demand of user j.

5. The intelligent power distribution monitoring system based on the Internet of Things according to claim 1 is characterized in that: In the comprehensive scheduling and optimization decision module, for the model predictive control submodule, a dynamic prediction model of microgrid and distributed energy is constructed based on physical principles and system operation characteristics. For the solar power generation part, the prediction model of solar power generation power changing with time is: Among them, psolar(t+Δt) represents the predicted value of solar power generation at the future time t+Δt, β0 is the intercept term, β1 is the light intensity influence coefficient, β2 is the temperature influence coefficient, T ref is the reference temperature, β3 is the influence coefficient of the lagged light intensity, y i is the periodic term coefficient, w i is the angular frequency of the corresponding period, is the phase angle, E is the random error term, and for wind power generation, the wind power generation power prediction model is: Among them, p wind (t+Δt) represents the predicted wind power generation value at the future time t+Δt, P is the air density, A is the swept area of ​​the wind turbine blade, v(t+Δt) is the predicted wind speed at the future time t+Δt, C p (λ, β) is the power coefficient of the wind turbine, η is the comprehensive transmission and power generation efficiency of the entire wind power generation system, and its user electricity demand prediction model is: Among them, D user (t+Δt) represents the predicted power demand of a specific user or user group at the future time t+Δt, μ is the long-term average level of power demand, α j (j=1, 2, ..., m) is the influence coefficient of working day and rest day, d j (t) is a binary variable indicating whether the current time t is a working day or a holiday, β k (k=1,2,…;p) is the seasonal influence coefficient, s k (t) is a categorical variable indicating the current season, η is the influence coefficient of the lag term of historical electricity consumption data, and f l (t-τ l ) is the lag τ1 time step, E′ is the random error term of the user power demand forecast, and the local power grid state change prediction model is: Where Vgrid(t+Δt) represents the voltage forecast value of the key nodes of the local power grid at the future time t+Δt, v grid (t) is the actual voltage value at the current time t, which is obtained through real-time monitoring of the power grid, α i is the active power change influence coefficient, Δp i (t) represents the change in active power at each distributed energy generation point or load at the current time t, βj is the reactive power change influence coefficient, ΔQ j (t) is the change in reactive power of each node at the current time t, Yk is the current change influence coefficient, ΔI k (t) is the change in current on the key line at the current moment t, and E″ is the random error term of the grid voltage prediction. The prediction model results of the above parts are reasonably integrated through data fusion. According to their respective importance and mutual correlation, a unified comprehensive dynamic prediction result covering energy generation, user demand and grid status is formed, which is used for subsequent real-time data fusion and scheduling decision-making operations.

6. The intelligent power distribution monitoring system based on the Internet of Things according to claim 5 is characterized in that: In the comprehensive dispatch and optimization decision module, a unified comprehensive dynamic prediction model covering energy generation, user demand and power grid status is formed according to the respective importance and mutual correlation. The algorithm formula of the model is Y(t+Δt)=w solar ×p solar (t+Δt)+w wind ×p wind (t+Δt)+w wer ×D wer (t+Δt)+w grid ×s grid ((t+Δt), where Y(t+Δt) represents the comprehensive prediction index vector at the future time t+Δt, Δt is the time step of the prediction, and w solar 、w wind 、w user 、w gridz are the weight coefficients of solar power generation, wind power generation, user power demand, and local grid voltage status in the comprehensive prediction model. are other influencing factors, E tatal is the random error term of the comprehensive prediction model.

7. The intelligent power distribution monitoring system based on the Internet of Things according to claim 1 is characterized in that: In the electric energy storage and transmission control module, a strategy is formulated to control the charging power according to the state of the energy storage device during charging, and the charging power adjustment algorithm is: Where Pcharge represents the recommended charging power, v bat is the current terminal voltage of the energy storage device, R int is the current internal resistance of the energy storage device, λ is the weight coefficient of the impact of internal resistance on the remaining life, L remain is the remaining life of the energy storage device.

8. The intelligent power distribution monitoring system based on the Internet of Things according to claim 1 is characterized in that: In the electric energy storage and transmission control module, a strategy is formulated to select equipment and adjust parameters based on multiple factors during power transmission. In order to select the most suitable equipment combination and adjust its working parameters to achieve stable and efficient transmission, an equipment selection and parameter adjustment algorithm based on comprehensive evaluation of power quality and transmission loss is adopted, and its formula is: Among them, S opt It represents the optimal transformer ratio selection result. w1, w2, and w3 are the weight coefficients of voltage deviation, power loss, and total harmonic distortion, respectively. ΔU(S) is the voltage deviation corresponding to the transformer ratio of S. ΔP(S) is the power loss corresponding to the transformer ratio of S. THD(S) is the total harmonic distortion corresponding to the transformer ratio of S.

Images