An energy management system and method based on cloud-edge integrated photovoltaic storage system

Abstract

The present application provides an energy management system based on a cloud-edge integrated photovoltaic storage system, including a cloud module and an edge module; an EMS controller for collecting information from the photovoltaic storage device module and the load module, and sending the collected data to the cloud module; the cloud module receives the data collected by the EMS controller, and predicts the future photovoltaic power generation power / electricity through the photovoltaic power generation prediction model; predicts the future load through the load prediction model; analyzes the peak and valley periods of electricity prices through the electricity price management model, and sends the predicted data to the edge module; the edge module performs energy management according to the battery SOC, rechargeable capacity, photovoltaic predicted power and power generation, superimposed peak and valley arbitrage and safety anti-backflow logic through the control logic of energy management. Through the above-mentioned settings, the present application makes the entire energy management system highly intelligent and automated, realizes the optimization of energy distribution and use, and improves energy utilization efficiency.

Description

Technical Field

[0001] The present application relates to the field of photovoltaic energy storage technology, and in particular to an energy management system and method based on a cloud-edge integrated photovoltaic storage system. Background Art

[0002] With the transformation of the global energy structure and the deepening implementation of sustainable development strategies, the development and utilization of renewable energy has become a key means of alleviating energy pressures and reducing carbon emissions. Photovoltaic and wind power, as primary forms of renewable energy, have experienced rapid growth in recent years due to their clean, renewable, and widespread advantages. However, these renewable energy sources are intermittent and uncertain, making accurate predictions of power generation and timing difficult, posing challenges to the stable operation of power systems.

[0003] In the prior art, a Chinese patent application numbered CN109473999A discloses a method and device for allocating photovoltaic power. This method is applied to a photovoltaic energy storage system. First, the peak-shaving and valley-filling mode of the photovoltaic energy storage system is obtained, and based on the peak-shaving and valley-filling mode, the current energy storage state of the energy storage unit is determined. Then, based on the peak-shaving and valley-filling mode, the photovoltaic power generation power, the load power, the preset minimum value of the power exchanged with the mains, and the preset maximum steady-state value of the power exchanged with the mains, the discharge steady-state power and the charging steady-state power of the energy storage unit are determined. Based on the peak-shaving and valley-filling mode, the current energy storage state, the rated power, the discharge steady-state power or the charging steady-state power of the energy storage unit, the first allocated power of the energy storage unit is determined. This patent revolves around peak-valley arbitrage, but the division of charging during peak periods and discharging during normal periods based on real-time status is too mechanical, and the overall benefits are not good.

[0004] Chinese patent application number CN117762744A discloses a photovoltaic energy storage management system based on big data, specifically relating to the field of photovoltaic energy storage technology, including a photovoltaic storage device data collection module, a data collation module, a neural network module, a database, an evaluation module, and a photovoltaic storage device management module; the photovoltaic storage device management module is respectively connected to the photovoltaic storage device data collection module, the data collation module, the neural network module, the database, and the evaluation module; the photovoltaic storage device data collection module is used to collect the photovoltaic device's main operating information, charging line information, discharge line information, and external environmental information, and transmit the photovoltaic device's main operating information, charging line information, discharge line information, and external environmental information to the data collation module. The patent provides an overview of the system composition, in which predictions are only made based on neural networks from a safety perspective, and the engineering practice significance is limited. In addition, the core control and distribution of photovoltaic energy storage management are basically not involved. Summary of the Invention

[0005] In response to the defects in the existing technology, the purpose of this application is to provide an energy management system and method based on a cloud-edge integrated photovoltaic storage system. For the photovoltaic energy storage system, a complete cloud-edge integrated design solution is provided, and the solution is designed from the perspective of maximizing benefits, taking into account technological innovation and project feasibility, so as to achieve the effect of fully utilizing photovoltaics and minimizing power consumption from the grid.

[0006] In one aspect of the present application, an energy management system based on a cloud-edge integrated photovoltaic storage system is provided, comprising: a cloud module and an edge module;

[0007] The cloud module includes: photovoltaic power generation prediction model, electricity price management model and load prediction model;

[0008] The edge module includes: an EMS controller with energy management control logic and information collection functions;

[0009] The EMS controller is used to collect information of the optical storage device module and the load module, and send the collected data to the cloud module;

[0010] The cloud module receives data collected by the EMS controller, predicts future photovoltaic power generation and power generation through the photovoltaic power generation prediction model; predicts future load through the load prediction model; analyzes peak and valley periods of electricity prices through the electricity price management model, and sends the predicted data to the edge module;

[0011] The edge module performs energy management through the EMS controller based on battery SOC, rechargeable capacity, future photovoltaic power generation power and power generation, superimposed peak-valley arbitrage and safety anti-backflow logic.

[0012] Furthermore, the photovoltaic power generation prediction model includes:

[0013] The first data collection unit is used to collect meteorological data of the equipment location, historical power generation data of the equipment, equipment status data, and equipment geographic information;

[0014] A first data preprocessing unit, used for cleaning the collected data and preprocessing feature engineering data;

[0015] The first prediction unit uses a linear regression equation to input the pre-processed data into the photovoltaic power generation prediction model to predict the future photovoltaic power generation power and power generation.

[0016] Furthermore, the linear regression equation is:

[0017] p=β+β1T+β2H+β3S+β4C+β5S pv +β6S inv +β7L lat +β8Llon +β9A+∈;

[0018] Among them, β, β1, β2, β3, β4, β5, β6, β7, β8, and β9 are coefficients to be fitted, which need to be estimated through training data; ∈ is the correction deviation; p is the photovoltaic power generation prediction data; S pv is the photovoltaic panel status; S inv is the inverter state; L lat is the latitude of the geographical location; L lon is the longitude of the geographical location; C is the cloud cover; S is the radiation intensity; T is the temperature; H is the humidity; and A is the altitude.

[0019] Furthermore, the load prediction model includes:

[0020] The second data collection unit is used to collect historical electricity consumption data and electricity consumption plans;

[0021] a second data preprocessing unit, configured to perform normalization processing on the collected data;

[0022] The second prediction unit is used to construct a state equation and an observation equation to predict future load based on the data processed by the second data preprocessing unit and the state space model.

[0023] Furthermore, the state equation is:

[0024] The observation equation is:

[0025] Among them, X t is the state vector at time t, Y t is the prediction vector at time t, u t is the control input at time t, A, B, C, D are the system matrices, and w t and v t are process noise and observation noise.

[0026] Furthermore, the predicting of future load specifically includes: estimating parameters of a state space model using a Bayesian method; and performing state estimation and prediction using a Kalman filter.

[0027] Furthermore, the electricity price management model includes:

[0028] a third data collection unit, configured to obtain electricity price data from a website of an electricity company or an electricity market operator;

[0029] The third data preprocessing unit fills or deletes missing values, ensures that the electricity price data has a consistent timestamp format, and aligns the data;

[0030] Data visualization unit to show how electricity prices change over time;

[0031] The peak and valley period recognition unit uses the K-means clustering algorithm to automatically identify the peak period, normal period, and valley period in electricity prices.

[0032] Furthermore, the EMS controller includes:

[0033] The battery management module is used to monitor the battery's SOC and the remaining capacity of the battery that can be charged;

[0034] Photovoltaic power generation management module, used to monitor photovoltaic power generation in real time and report it to the cloud module, helping the cloud module to correct the predicted values ​​of photovoltaic power generation and power generation in real time;

[0035] The charging management module calculates the SOC value of the battery pre-charge at night and controls the battery charging based on the load power consumption, photovoltaic power generation and peak and valley time periods of electricity prices in the next day;

[0036] The discharge management module prioritizes battery discharge or reduces grid power purchases based on electricity price periods and battery SOC;

[0037] The anti-backflow strategy module monitors the anti-backflow meter in real time and adjusts the photovoltaic power generation or battery charging and discharging strategy to ensure that no excess power flows back to the grid.

[0038] A second aspect of the present application provides an energy management method based on a cloud-edge integrated photovoltaic storage system, the method comprising:

[0039] The EMS controller collects information from the optical storage device module and the load module and sends the data to the cloud module;

[0040] The cloud module predicts future photovoltaic power generation and power generation through a photovoltaic power generation prediction model, predicts future load through a load prediction model, and analyzes peak and valley periods of electricity prices through an electricity price management model;

[0041] Execute the control logic of energy management in the EMS controller based on the battery SOC, rechargeable capacity, future photovoltaic power and power generation, superimposed peak-valley arbitrage and safety anti-reverse flow logic;

[0042] The EMS controller monitors photovoltaic power generation, battery SOC, and backflow prevention in real time, and adjusts photovoltaic power generation and battery charging and discharging strategies according to the instructions of the cloud module;

[0043] The cloud module calculates revenue based on real-time data and optimizes energy management strategies according to revenue conditions.

[0044] Furthermore, the photovoltaic power generation prediction model uses a linear regression method, the load prediction model uses a state space model, and the electricity price management model uses a K-means clustering algorithm;

[0045] The control logic of the energy management includes battery management, photovoltaic power generation management, charging management, discharge management and anti-backflow strategy;

[0046] The EMS controller dynamically adjusts photovoltaic power generation and battery charging and discharging strategies based on the predicted data and real-time data from the cloud module to achieve energy optimization management.

[0047] Compared with the prior art, the present invention has at least one of the following beneficial effects:

[0048] 1. This application uses the photovoltaic power generation prediction model, electricity price management model and load prediction model of the cloud module to accurately predict future photovoltaic power generation, electricity price changes and load demand. The electricity price management model can analyze the peak and valley periods of electricity prices, so that the energy management system can charge when electricity prices are low and discharge when electricity prices are peak, thereby optimizing electricity costs; the edge module takes into account the safety and anti-backflow logic when managing energy, ensuring that the system can operate safely in the event of power grid failure or abnormal conditions, and preventing backflow from impacting the power grid. The collaborative work of the cloud module and the edge module makes the entire energy management system highly intelligent and automated, and can make fast and accurate decisions based on real-time data, optimize energy distribution and use, and improve energy utilization efficiency.

[0049] 2. This application effectively integrates and optimizes electricity price data obtained from power companies or market operators by integrating data collection, preprocessing, visualization and automatic identification of peak and valley periods, ensuring data consistency and integrity, reducing the possibility of manual intervention and errors, and visually displaying the trend of electricity price changes. It also uses the K-means clustering algorithm to accurately identify peak periods, flat periods and valley periods in electricity prices. Based on the changing patterns of electricity prices, it optimizes reasonable power dispatching plans and achieves optimal power resource allocation and economic benefit planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Other features, objects and advantages of the present application will become more apparent upon reading the detailed description of non-limiting embodiments with reference to the following drawings:

[0051] Figure 1 This is a block diagram of an energy management system based on a cloud-edge integrated photovoltaic storage system according to an embodiment of the present application.

[0052] Figure 2 This is a structural diagram of an energy management system based on a cloud-edge integrated photovoltaic storage system according to an embodiment of the present application.

[0053] Figure 3 This is a schematic diagram of the structure of an optical storage system according to an embodiment of the present application.

[0054] Figure 4 This is a flow chart of an energy management method based on a cloud-edge integrated photovoltaic storage system according to an embodiment of the present application. DETAILED DESCRIPTION

[0055] The present application is described in detail below with reference to specific embodiments. The following embodiments will help those skilled in the art to further understand the present application, but are not intended to limit the present application in any form. It should be noted that those skilled in the art may make several variations and improvements without departing from the scope of the present application. These all fall within the scope of protection of the present application.

[0056] Reference Figure 1 As shown, an energy management system based on a cloud-edge integrated photovoltaic storage system according to an embodiment of the present application includes a cloud module and an edge module.

[0057] The cloud module includes a photovoltaic power generation prediction model, an electricity price management model, and a load prediction model; the edge module includes an EMS controller with energy management control logic and information collection functions.

[0058] The EMS controller is used to collect information from the photovoltaic storage device module and the load module, and send the collected data to the cloud module in a unified manner; the cloud module receives the data collected by the EMS controller, and predicts the future photovoltaic power generation power and power generation through the photovoltaic power generation prediction model; predicts the future load through the load prediction model; analyzes the peak and valley periods of electricity prices through the electricity price management model, and sends the predicted data to the edge module; the edge module uses the energy management control logic to perform energy management based on the battery SOC, rechargeable capacity, future photovoltaic power generation power and power generation, superimposed peak and valley arbitrage and safety anti-backflow logic.

[0059] This application uses the photovoltaic power generation prediction model, electricity price management model and load prediction model of the cloud module to accurately predict future photovoltaic power generation, electricity price changes and load demand. The electricity price management model can analyze the peak and valley periods of electricity prices, so that the energy management system can charge when electricity prices are low and discharge when electricity prices are peak, thereby optimizing electricity costs; the edge module takes into account the safety and anti-backflow logic when managing energy, ensuring that the system can operate safely in the event of power grid failure or abnormal conditions, and preventing backflow from impacting the power grid. The collaborative work of the cloud module and the edge module makes the entire energy management system highly intelligent and automated, and can make fast and accurate decisions based on real-time data, optimize energy distribution and use, and improve energy utilization efficiency.

[0060] Specifically, cloud-edge integration refers to deploying light and wind power generation prediction models, electricity price management models and load prediction models to the cloud module; the control logic and information collection functions of specific energy management are deployed to the edge module, through the EMS controller, that is, EMS serves as a data collection module and management module; EMS is used to collect information from the photovoltaic storage device module and the load module; the collected data is sent to the cloud in a unified manner; the edge module performs energy management based on the received prediction data, combined with information such as battery SOC, rechargeable capacity, future photovoltaic power and power generation, as well as safety anti-backflow logic, through the control logic of energy management, including the formulation and execution of charging, discharging, energy storage and other strategies. By predicting photovoltaic power and power generation, as well as predicting load, the valley electricity price is used to supplement the power difference, and the control difference is as small as possible, so that photovoltaic power can be fully utilized and the power consumption of the grid is as small as possible.

[0061] In some possible embodiments, the photovoltaic power generation prediction model includes: a first data collection unit, used to collect meteorological data of the equipment location, equipment historical power generation data, equipment status data, and equipment geographic information; a first data preprocessing unit, used to clean the collected data and preprocess the feature engineering data; a first prediction unit, using a linear regression equation, inputs the preprocessed data into the photovoltaic power generation prediction model to predict future photovoltaic power generation power and power generation.

[0062] Specifically, the photovoltaic power generation prediction module first collects meteorological data of the equipment location, such as light intensity, temperature, humidity, wind speed, etc., through the first data collection unit; collects historical power generation data of the equipment, including power generation power and power generation in the past period of time; collects equipment status data, such as equipment working status, fault records, etc.; collects geographical information of the equipment, such as longitude, latitude, altitude, etc., and uses the above data as input of the power generation prediction module. Then, the collected data is cleaned and feature engineering data preprocessed through the first data preprocessing unit to obtain features useful for photovoltaic power generation prediction, improve the prediction accuracy and stability of the model, and input the data into the first prediction unit of the photovoltaic power generation prediction model. Finally, the first prediction unit uses the linear regression equation to predict future photovoltaic power generation power and power generation.

[0063] Among them, data cleaning includes: removing outliers, missing values ​​or duplicate values ​​to ensure data quality; feature engineering data preprocessing to extract features useful for photovoltaic power generation prediction, including time series characteristics of light intensity, relationship characteristics between temperature and power generation, etc.

[0064] By collecting rich data and preprocessing it, we extract features that are useful for photovoltaic power generation prediction. We use the linear regression equation as the basic prediction model to improve the accuracy of the prediction model. By predicting future photovoltaic power and power generation, we provide support for the system's energy management decisions, help the system better arrange the charging and discharging plans of energy storage equipment, optimize energy utilization, and adjust the power generation and energy storage strategies according to the prediction results to reduce costs and improve economic benefits.

[0065] In the above embodiment, the linear regression equation is:

[0066] p=β+β1T+β2H+β3S+β4C+β5S pv +β6S inv +β7L lat +β8L lon +β9A+∈;

[0067] Among them, β, β1, β2, β3, β4, β5, β6, β7, β8, and β9 are coefficients to be fitted, which need to be estimated through training data; ε is the correction deviation; p is the predicted photovoltaic power generation power; S pv is the photovoltaic panel status; S inv is the inverter state; L lat is the latitude of the geographical location; L lon is the longitude of the geographical location; C is the cloud cover; S is the radiation intensity; T is the temperature; H is the humidity; and A is the altitude.

[0068] Specifically, the power generation power P is predicted to be photovoltaic p = β + β1T + β2H + β 3s +β4C+β5S pv +β6S inv +β7L lat +β8L lon +β9A+∈, where β, β1, β2, β3, β4, β5, β6, β7, β8, and β9 are coefficients to be fitted and need to be estimated through training data, ∈ is the correction deviation, P is the historical power generation data, and the photovoltaic panel status S pv , inverter state S inv , the geographical location L can be expressed as latitude L lat , longitude L lon , cloud cover C, radiation intensity S, temperature T, humidity H, altitude A; the radiation coefficient is the normalized result of the radiation amount; through regression analysis of historical data, each coefficient can be estimated, and the trained model can be used to estimate the power generation in the next 24 hours on an hourly basis.

[0069] In some possible embodiments, the load prediction model includes: a second data collection unit, used to collect historical electricity consumption data and electricity consumption plans; a second data preprocessing unit, used to normalize the collected data; and a second prediction unit, used to construct a state equation and an observation equation to predict future load based on the data processed by the second data preprocessing unit and the state space model.

[0070] like Figure 3 As shown, the core parts are the cloud platform and the site-side EMS, which together form the cloud-edge integrated photovoltaic storage system energy management system. In order to facilitate the understanding of the object-oriented system and the scope of application of this patent. Figure 3 It is mainly divided into time-varying energy systems, such as photovoltaics and inverters, wind turbines and converters; storage energy systems, such as lithium batteries / sodium batteries and battery management systems BMS, as well as bidirectional inverters PCS, supercapacitors; supplementary energy systems: such as gasoline / diesel generators and controllers ECU, fuel cells; energy storage support systems: such as thermal management systems, fire protection systems; and detection and execution systems: such as electricity meters, switch components, etc.; load systems: such as adjustable loads and necessary loads.

[0071] This application first collects historical electricity consumption data and electricity consumption plans through the second data collection unit, and then the second data preprocessing unit normalizes the collected data to achieve data consistency and comparability; finally, the second prediction unit uses the processed data and the preset state space model to construct a state equation and an observation equation to predict future loads. The load prediction model integrates data collection, preprocessing, and a prediction unit based on a state space model, and predicts future electricity consumption by importing electricity consumption plans (long-term, temporary) and historical electricity consumption information. The data is normalized and input into the load prediction model to predict future loads. It can efficiently utilize historical electricity consumption data and electricity consumption plans, and after normalization, construct accurate state equations and observation equations to achieve accurate prediction of future loads.

[0072] In the above embodiment, the state equation is: The observation equation is:

[0073] Among them, X t is the state vector at time t, Y t is the prediction vector at time t, u t is the control input at time t, A, B, C, D are the system matrices, and w t and v t are process noise and observation noise.

[0074] The state space model is used to predict the load. The state space model is used to describe the changes in the system state by using the state equation and the observation equation to observe the changes between the state equation.

[0075] Specifically, the state space model is used to predict the load. The state space model uses the state equation to describe the change of the system state and the observation equation to observe the change between the value and the state equation. The state equation is The observation equation is X t is the state vector at time t, Y t is the forecast vector at time t (e.g., predicted load), u t is the control input at time t (historical electricity consumption data), A, B, C, D are system matrices, w t and v t The dynamic behavior of the system is described by choosing appropriate state variables.

[0076] One implementation method is to select the load trend, seasonal component, and random fluctuation as state variables and construct a state equation describing the evolution of the state variables based on historical power consumption data and power consumption plans.

[0077] The trend and seasonal components of the load are T t 、S t ; Then the state equation can be written as T t =T t-1 +b t +w 1t and S t =S t-1 +w 2t ;

[0078] where b t is the trend change, w 1t and w 2t is the added noise; describes the relationship between the observed value (i.e., load data) and the state variable: y t =T t +S t +D t +v t ;D t is the impact of electricity consumption plan, v t is the observation noise; finally, the future load is predicted.

[0079] In the above embodiment, the load trend is strongly correlated with the seasonal component, for example, the power consumption of air conditioners in summer is significantly higher than that in autumn; the load trend is the change in the load in a certain seasonality over a period of time, such as a day or even a week;

[0080] Specifically, the Bayesian method is used to estimate the parameters of the state-space model, and finally the Kalman filter is used for state estimation and prediction. The Kalman filter is deleted for state estimation and prediction, and only the Bayesian method is used to estimate the parameters of the state-space model. The Bayesian method is an existing and common method.

[0081] In some possible embodiments, the electricity price management model includes: a third data collection unit, used to obtain electricity price data from the website of the power company or the power market operator; a third data preprocessing unit, used to fill or delete missing values, ensure that the electricity price data has a consistent timestamp format, and align the data; a data visualization unit, used to display the changes in electricity prices over time; a peak and valley period identification unit, which uses a K-means clustering algorithm to automatically identify peak periods, flat periods, and valley periods in electricity prices.

[0082] Specifically, it goes through four processes: data collection, data preprocessing, data visualization, and identification of peak and valley periods of electricity prices. The first step is data collection: obtaining data from the power company or power market operator's website. The data format may be CSV file, JSON file or obtained through API interface. The second step is data preprocessing: filling or deleting missing values, and ensuring that all electricity price data has a consistent timestamp format. Finally, if the data comes from multiple sources, ensure data alignment. The third step is data visualization: using time series graphs to show the changes in electricity prices over time. The fourth step is peak and valley period identification: using K-means clustering unsupervised learning algorithm to automatically identify the three periods in electricity prices, namely peak period, flat period, and valley period.

[0083] By integrating data collection, preprocessing, visualization and automatic identification of peak and valley periods, the system effectively integrates and optimizes electricity price data obtained from power companies or market operators, ensuring data consistency and integrity, reducing the possibility of manual intervention and errors, and visually displaying the trend of electricity price changes. It also uses the K-means clustering algorithm to accurately identify peak, flat and valley periods in electricity prices. Based on the changing patterns of electricity prices, it optimizes reasonable power dispatching plans and achieves optimal power resource allocation and economic benefit planning.

[0084] like Figure 2 As shown, in some possible embodiments, the EMS controller includes: a battery management module, configured to monitor the SOC of the battery and the remaining capacity of the battery that can currently be charged.

[0085] Specifically, instructions are sent to the EMS through the cloud to set the maximum and minimum SOC thresholds of the battery to 95% and 15% respectively; the cloud-based electricity price prediction model predicts the peak and valley time periods of electricity prices; and the cloud-based load prediction model predicts the electricity consumption in the next day.

[0086] The photovoltaic power generation management module is used to monitor photovoltaic power generation in real time and report it to the cloud, helping the cloud to correct the predicted value of photovoltaic power generation in real time.

[0087] The cloud-based photovoltaic power generation prediction model predicts photovoltaic power generation and, based on historical data and weather forecasts, predicts photovoltaic power generation for the next day. The EMS side monitors photovoltaic power generation in real time and reports it to the cloud to help the cloud correct the predicted photovoltaic power generation value in real time. Battery management is divided into battery charging management and discharge management, and SOC monitoring monitors the battery SOC in real time.

[0088] The charging management module calculates the SOC value of the battery pre-charge at night and controls the battery charging based on the load power consumption, photovoltaic power generation and peak and valley time periods of electricity prices in the next day.

[0089] By predicting the load's power consumption for the next day, as well as the peak and valley periods of photovoltaic power generation and electricity prices, the cloud-based energy management system calculates the battery's SOC value for pre-charging at night and sends this data to the EMS. The EMS then controls battery charging to the cloud-specified SOC value. When daytime photovoltaic power generation exceeds the current load demand and the battery SOC is below the maximum threshold, battery charging is prioritized. When electricity prices are low and the battery SOC is below the maximum threshold, charging is performed from the grid.

[0090] The discharge management module prioritizes battery discharge or reduces grid power purchases based on electricity price periods and battery SOC.

[0091] When electricity prices are at peak times and the battery SOC is above the minimum threshold, battery discharge is prioritized to reduce grid power purchases. When photovoltaic power generation is insufficient to meet load demand and the battery SOC is above the minimum threshold, battery discharge is used.

[0092] The anti-backflow strategy module monitors the anti-backflow meter in real time and adjusts the photovoltaic power generation or battery charging and discharging strategy to ensure that no excess power flows back to the grid.

[0093] like Figure 2 As shown, the EMS monitors the backflow prevention meter in real time. If it detects that power is flowing back into the grid, it adjusts the photovoltaic power generation or battery charging and discharging strategy to ensure that no excess power flows back into the grid. The data is uploaded to the cloud, and the cloud calculates the revenue based on the real-time data. At the same time, when the photovoltaic power and battery SOC are above the minimum threshold and can meet the load demand, the change in photovoltaic power is input into the fuzzy logic control model, and the output of the model is used as the output power of the inverter. This enables rapid adjustment of the inverter output power, ensures power balance between the photovoltaic system and the load, and prevents power from being drawn from the grid.

[0094] When inputting the change value of photovoltaic power into the fuzzy logic control model, the fuzzy logic is removed and the classic PID algorithm is used for adjustment, that is, the proportional, integral and differential method, that is, the output is input into the PID algorithm.

[0095]

[0096] Among them, u t is the control output e t =P b，t+1 -P b，t , K p K i K d They are proportional, integral, and differential gains respectively; this PID is an incremental PID. The advantage of using incremental PID is that it prevents excessive system oscillation, which reduces grid power consumption and reverse flow when energy storage and photovoltaics can meet the load.

[0097] This application uses energy management logic to comprehensively consider battery SOC, rechargeable capacity, photovoltaic predicted power and power generation, and superimposes peak-valley arbitrage and safety anti-backflow logic. First, the battery management module obtains the battery state (SOC) and the remaining capacity that the battery can currently charge; photovoltaic predicted power is the predicted power of photovoltaic power generation in a period of time in the future, and photovoltaic current power generation is the actual power generation of the current photovoltaic system; local peak-valley electricity prices are the fluctuations of grid electricity prices in different time periods; the anti-backflow strategy is to prevent excess electricity from flowing back into the grid and protect the grid and equipment.

[0098] Reference Figure 4 As shown, the second aspect of the present application provides an energy management method based on a cloud-edge integrated photovoltaic storage system, the method comprising:

[0099] S1. Collect information of optical storage device modules and load modules through the EMS controller and send the data to the cloud module.

[0100] S2. The cloud module predicts future photovoltaic power generation and power generation through the photovoltaic power generation prediction model, predicts future load through the load prediction model, and analyzes peak and valley periods of electricity prices through the electricity price management model.

[0101] S3. Execute the control logic of energy management in the EMS controller based on the battery SOC, rechargeable capacity, future photovoltaic power and power generation, and superimposed peak-valley arbitrage and safety anti-reverse flow logic.

[0102] S4,EMS controller monitors photovoltaic power generation, battery SOC and backflow prevention in real time, and adjusts photovoltaic power generation and battery charging and discharging strategies according to the instructions of the cloud module.

[0103] By deploying photovoltaic power generation prediction models, electricity price management models and load prediction models in the cloud module; deploying EMS controllers in the edge module to collect information from the photovoltaic storage device module and the load module, and sending the data to the cloud; the cloud module predicts the future photovoltaic power generation power / power through the photovoltaic power generation prediction model, predicts the future load through the load prediction model, and analyzes the peak and valley periods of electricity prices through the electricity price management model; according to the battery SOC, rechargeable capacity, photovoltaic predicted power and power generation, superimposed peak and valley arbitrage and safety anti-backflow logic, the energy management control logic is executed in the EMS controller. The EMS controller executes battery charging management, discharge management and anti-backflow strategies according to the instructions and prediction data issued by the cloud module, improves energy utilization efficiency, and makes the entire energy management system highly intelligent and automated, and can make fast and accurate decisions based on real-time data.

[0104] In the above embodiment, the present application also includes S5, where the cloud module calculates revenue based on real-time data and optimizes the energy management strategy according to the revenue situation.

[0105] By obtaining real-time revenue information and optimizing energy management strategies based on the revenue information, we can design solutions from the perspective of maximizing revenue and optimize energy management strategies.

[0106] In the above embodiment, the photovoltaic power generation prediction model adopts linear regression, the load prediction model uses the state space model, and the electricity price management model uses the K-means clustering algorithm; the control logic of energy management includes battery management, photovoltaic power generation management, charging management, discharge management and anti-backflow strategy; the EMS controller dynamically adjusts the photovoltaic power generation and battery charging and discharging strategies based on the cloud module prediction data and real-time data to achieve energy optimization management.

[0107] This application is applicable to the energy management of photovoltaic energy storage systems, and is also applicable to the energy management of expanded systems based on photovoltaic storage, such as adding combined wind turbines, diesel generators and other systems.

[0108] This application involves aspects such as photovoltaic power generation prediction, load energy consumption prediction, and peak-valley arbitrage based on predicted energy consumption. It involves the independent subsystems in the above part, and this application is also applicable.

[0109] The above describes the specific embodiments of the present application. It should be understood that the present application is not limited to the specific embodiments described above, and those skilled in the art may make various modifications or variations within the scope of the claims, which do not affect the substantive content of the present application. The above preferred features may be used in any combination as long as they do not conflict with each other.

Claims

1. An energy management system based on a cloud-edge integrated photovoltaic storage system, characterized in that: include: Cloud module and edge module; The cloud module includes: photovoltaic power generation prediction model, electricity price management model and load prediction model; The edge module includes: an EMS controller with energy management control logic and information collection functions; The EMS controller is used to collect information of the optical storage device module and the load module, and send the collected data to the cloud module; The cloud module receives data collected by the EMS controller, predicts future photovoltaic power generation and power generation through the photovoltaic power generation prediction model; predicts future load through the load prediction model; analyzes peak and valley periods of electricity prices through the electricity price management model, and sends the predicted data to the edge module; The edge module performs energy management through the EMS controller based on battery SOC, rechargeable capacity, future photovoltaic power generation and generation, superimposed peak-valley arbitrage and safety anti-backflow logic; The EMS controller monitors the photovoltaic power generation in real time and reports it to the cloud module, helping the cloud module to correct the predicted value of photovoltaic power generation in real time; The EMS controller dynamically adjusts photovoltaic power generation and battery charging and discharging strategies based on the cloud module's predicted data and real-time data to achieve energy optimization management; The photovoltaic power generation prediction model adopts a linear regression method, the load prediction model uses a state space model, and the electricity price management model uses a K-means clustering algorithm.

2. The energy management system based on the cloud-edge integrated solar-storage system according to claim 1, characterized in that: The photovoltaic power generation prediction model includes: The first data collection unit is used to collect meteorological data of the equipment location, historical power generation data of the equipment, equipment status data, and equipment geographic information; A first data preprocessing unit, used for cleaning the collected data and preprocessing feature engineering data; The first prediction unit uses a linear regression equation to input the pre-processed data into the photovoltaic power generation prediction model to predict the future photovoltaic power generation power and power generation.

3. The energy management system based on the cloud-edge integrated photovoltaic storage system according to claim 2 is characterized in that: The linear regression equation is: p=β+β1T+β2H+β3S+β4C+β5S pv +β6S inv +β7L lat +β8L lon +β9A+∈; Among them, β, β1, β2, β3, β4, β5, β6, β7, β8, and β9 are coefficients to be fitted, which need to be estimated through training data; ∈ is the correction deviation; p is the photovoltaic power generation prediction data; S pv is the photovoltaic panel status; S inv is the inverter state; L lat is the latitude of the geographical location; L lon is the longitude of the geographical location; C is the cloud cover; S is the radiation intensity; T is the temperature; H is the humidity; and A is the altitude.

4. The energy management system based on the cloud-edge integrated photovoltaic storage system according to claim 1, characterized in that: The load prediction model includes: The second data collection unit is used to collect historical electricity consumption data and electricity consumption plans; a second data preprocessing unit, configured to perform normalization processing on the collected data; The second prediction unit is used to construct a state equation and an observation equation to predict future load based on the data processed by the second data preprocessing unit and the state space model.

5. The energy management system based on the cloud-edge integrated photovoltaic storage system according to claim 4 is characterized in that: The state equation is: t =Ax t-1 +Bu t +w t ; The observation equation is: t =Cx t +Du t +v t ; Among them, X t is the state vector at time t, Y t is the prediction vector at time t, u t is the control input at time t, A, B, C, D are the system matrices, and w t and v t are process noise and observation noise.

6. The energy management system based on the cloud-edge integrated photovoltaic storage system according to claim 4, characterized in that: The predicting of future load specifically includes: estimating parameters of a state space model using a Bayesian method; and performing state estimation and prediction using a Kalman filter.

7. The energy management system based on the cloud-edge integrated photovoltaic storage system according to claim 1, characterized in that: The electricity price management model includes: a third data collection unit, configured to obtain electricity price data from a website of an electricity company or an electricity market operator; The third data preprocessing unit fills or deletes missing values, ensures that the electricity price data has a consistent timestamp format, and aligns the data; Data visualization unit to show how electricity prices change over time; The peak and valley period recognition unit uses the K-means clustering algorithm to automatically identify the peak period, normal period, and valley period in electricity prices.

8. The energy management system based on the cloud-edge integrated photovoltaic storage system according to claim 1, characterized in that: The EMS controller includes: The battery management module is used to monitor the battery's SOC and the remaining capacity of the battery that can be charged; Photovoltaic power generation management module, used to monitor photovoltaic power generation in real time and report it to the cloud module, helping the cloud module to correct the predicted values ​​of photovoltaic power generation and power generation in real time; The charging management module calculates the SOC value of the battery pre-charge at night and controls the battery charging based on the load power consumption, photovoltaic power generation and peak and valley time periods of electricity prices in the next day; The discharge management module prioritizes battery discharge or reduces grid power purchases based on electricity price periods and battery SOC; The anti-backflow strategy module monitors the anti-backflow meter in real time and adjusts the photovoltaic power generation or battery charging and discharging strategy to ensure that no excess power flows back to the grid.

9. An energy management method based on a cloud-edge integrated photovoltaic storage system, characterized in that: The method comprises: The EMS controller collects information from the optical storage device module and the load module and sends the data to the cloud module; The cloud module predicts future photovoltaic power generation and power generation through a photovoltaic power generation prediction model, predicts future load through a load prediction model, and analyzes peak and valley periods of electricity prices through an electricity price management model; Execute the control logic of energy management in the EMS controller based on the battery SOC, rechargeable capacity, future photovoltaic power and power generation, superimposed peak-valley arbitrage and safety anti-reverse flow logic; The EMS controller monitors photovoltaic power generation, battery SOC, and backflow prevention in real time, and adjusts photovoltaic power generation and battery charging and discharging strategies according to the instructions of the cloud module; The cloud module calculates revenue based on real-time data and optimizes energy management strategies according to revenue; The EMS controller monitors the photovoltaic power generation in real time and reports it to the cloud module, helping the cloud module to correct the predicted value of photovoltaic power generation in real time; The EMS controller dynamically adjusts photovoltaic power generation and battery charging and discharging strategies based on the cloud module's predicted data and real-time data to achieve energy optimization management; The photovoltaic power generation prediction model adopts a linear regression method, the load prediction model uses a state space model, and the electricity price management model uses a K-means clustering algorithm.

10. The energy management method based on the cloud-edge integrated photovoltaic storage system according to claim 9, characterized in that: The control logic of the energy management includes battery management, photovoltaic power generation management, charging management, discharging management and anti-backflow strategy.

Images