Power consumption management method for photovoltaic power supply

Abstract

The invention discloses a photovoltaic power supply power consumption management method, which comprises the following steps: establishing a multi-source fusion prediction model to calculate and predict power consumption and photovoltaic power generation by acquiring traffic flow, weather forecast, energy storage module state and charging demand data, and calculating a priority index according to the charge state, health state, temperature and residual life of an energy storage module. And generating an energy distribution strategy based on the net load power demand and the priority index, and controlling photovoltaic, energy storage and power grid cooperative power supply. The prediction precision is improved through multi-source data fusion, the priority of the energy storage module is dynamically calculated, optimal distribution of energy is achieved, the photovoltaic self-generation and self-use rate can be improved, dependence on a power grid is reduced, the power grid access frequency and power transmission loss are reduced, the service life of the energy storage module is prolonged, and the overall operation efficiency of the system is improved. And the power supply stability and reliability are guaranteed, and good economical efficiency and practicability are achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of artificial intelligence and relates to a photovoltaic power supply power consumption management method. Background Art

[0002] Currently, charging stations equipped with photovoltaic power sources face significant challenges amid the rapid adoption of new energy vehicles. For one thing, the power generated by the photovoltaic systems within the stations is highly dependent on natural sunlight, while the demand for charging new energy vehicles is significantly uneven in both time and space. During the off-peak hours of midday, while the photovoltaic system's power generation is high, charging demand is relatively low, resulting in significant waste of photovoltaic power. During peak charging times in the evening and at night, the photovoltaic system's power generation capacity decreases while the charging load increases dramatically, forcing charging stations to rely on the main power grid for power, significantly reducing the utilization of the photovoltaic power source.

[0003] On the other hand, existing charging station energy management systems mostly follow traditional models, which can only simply allocate photovoltaic energy storage based on historical charging data. They are unable to accurately capture abnormal power demand such as sudden changes in traffic volume and users' temporary energy replenishment. When large-scale events lead to a surge in vehicles in the surrounding area, or when extreme weather triggers concentrated vehicle charging, the charging station energy storage system often lacks the integrated analysis of multi-dimensional data such as traffic volume and weather changes, making it difficult to optimize energy allocation in advance. This leads to the dilemma of insufficient energy storage capacity and ineffective utilization of photovoltaic power, which not only affects the user charging experience but also increases the operating costs of the charging station. In addition, due to the failure to integrate the aging status of photovoltaic modules and the health of energy storage batteries with real-time power generation forecasts, there is still an over-reliance on energy storage equipment in the energy scheduling process, which accelerates battery loss and shortens the service life of the equipment.

[0004] While traditional user microgrids possess basic energy management logic, they rely solely on static calculations based on fixed sunlight durations and preset electricity consumption ratios. They lack a response mechanism to dynamic environmental factors such as real-time light intensity and temperature changes, making it difficult to adapt to demand fluctuations in complex scenarios such as sudden weather changes and changes in user electricity consumption habits. Management systems based on second-life batteries use state of charge, health status, and temperature parameters to construct a priority evaluation system for energy storage modules, but they do not integrate external environmental data such as traffic flow and weather forecasts for forward-looking scheduling. Sudden charging peaks can easily lead to energy storage module overloads or grid access delays. AI-based management systems incorporate traffic flow data for electricity demand forecasting, but they do not incorporate battery health status and photovoltaic power generation forecasts into a unified decision-making framework. This results in insufficient scheduling accuracy for energy allocation strategies in photovoltaic surplus / shortage scenarios, making it difficult to achieve coordinated optimization of the grid, photovoltaics, and energy storage.

[0005] The existing technology lacks an intelligent management solution that can integrate multi-dimensional data, dynamically optimize energy distribution paths, and achieve coordinated grid-photovoltaic-energy storage scheduling. Therefore, a photovoltaic power management method is urgently needed to address the above issues. Summary of the Invention

[0006] In order to solve the problems existing in the background technology, the present invention proposes a photovoltaic power supply power consumption management method, which can quickly and accurately process complex data in the power system, adapt to the dynamic changes in the power system operating status in real time, achieve multi-objective global optimal scheduling, and improve the reliability, economy and stability of the power system operation.

[0007] A first aspect of the present invention provides a photovoltaic power supply power management method, comprising: Obtain traffic flow data, weather forecast data, energy storage module status data, and charging demand data; Based on the data, a multi-source fusion prediction model is established to calculate the predicted power consumption and photovoltaic power generation; the priority index is calculated based on the charge state, health state, temperature and remaining life of the energy storage module; Generate an energy allocation strategy based on net load power demand and priority index to control the coordinated power supply of photovoltaics, energy storage and power grid.

[0008] Optionally, the method for constructing the multi-source fusion prediction model is: using a time series model for regular measurement periods ,in is the predicted electricity consumption in time period Dj, is the average power consumption in the historical T cycles, is the forecast value of the previous period, is the adaptive weight; the traffic flow correlation model is used for the approved metering period ,in Predict traffic flow for associated routes, is the correlation coefficient, is the user's electricity consumption characteristic value, is the climate correction factor, 、 are model parameters.

[0009] Optionally, a radius is defined with the charging station as the center The coverage length within the range is greater than The one-way roads are the intended roads, and the ratio of traffic volume to electricity consumption of each intended road is calculated. , through discrete values ,choose The smallest road, where bp is the bs mean, es is the electricity consumption, vs is the traffic volume, and n is the data volume.

[0010] Optionally, the calculation model of the photovoltaic power generation is ,in is the power generation at time t, For overall efficiency, is the photovoltaic panel area, is the light intensity, is the ambient temperature, is the standard temperature, is the temperature coefficient.

[0011] Optionally, the calculation model of the energy storage module priority index is ,in is the priority of module s at time t, is the state of charge, For health status, is the temperature, is the optimal operating temperature, L(s) is the remaining life index, 、 、 、 is the weight coefficient.

[0012] Optionally, the energy allocation strategy is generated by: Calculate net load power ,in To predict power consumption, is the photovoltaic power generation power at time t; when ≤0, energy storage charging power , where Z is the full charge capacity of the energy storage, is the average state of charge, For charging efficiency; when >0, if Only energy storage is used for power supply. The grid power supply ,in, is the real-time energy storage power, 、 is the threshold ratio, The lower limit of safety.

[0013] Optionally, the step of optimizing the energy allocation strategy by reinforcement learning, wherein the reward function of the reinforcement learning is , where R is the reward value, Cost is the operating cost, Emission is the carbon emission, and Satisfaction is the user satisfaction. 、 、 is the weight coefficient.

[0014] Optionally, the energy storage module status data includes the SOC, SOH, voltage, current and temperature parameters of the cascade utilization battery pack, the vehicle flow data is collected through video recognition and RFID technology, and the weather forecast data includes the light intensity, temperature and precipitation probability in a preset time period in the future.

[0015] Optionally, the energy allocation strategy also includes an energy storage module screening mechanism based on a priority index, which selects energy storage modules from high to low according to P(s, t) to participate in charging and discharging, and exits operation when the module SOC is lower than a preset threshold or the SOH is lower than a preset threshold.

[0016] The steps of controlling photovoltaic, energy storage and grid coordinated power supply include: Instructions are sent to the switch management module via the high-speed controller area network to control the on / off status of the grid and the inverter, and to adjust the output power of the charging pile. The inverter efficiency is used as a correction factor in the power calculation.

[0017] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a photovoltaic power supply power management method. By acquiring traffic flow, weather forecast, energy storage module status, and charging demand data, a multi-source fusion prediction model is established to calculate the predicted power consumption and photovoltaic power generation. The priority index is calculated based on the charge state, health state, temperature, and remaining life of the energy storage module. An energy allocation strategy is generated based on the net load power demand and the priority index to control the coordinated power supply of photovoltaic, energy storage, and power grid. The present invention improves prediction accuracy through multi-source data fusion, dynamically calculates the priority of energy storage modules, and realizes optimal energy allocation. It can increase the photovoltaic self-generation and self-use rate, reduce dependence on the power grid, reduce the number of grid access times and power transmission losses, extend the service life of the energy storage module, improve the overall operating efficiency of the system, ensure the stability and reliability of power supply, and has good economic and practical properties. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of a photovoltaic power management method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0020] In one embodiment, if Figure 1 As shown, a photovoltaic power supply power management method is provided, which is applied in Figure 1Take the example of , and explain it, including the following specific steps: S10: Obtaining traffic flow data, weather forecast data, energy storage module status data, and charging demand data; Specifically, traffic flow data refers to the flow of motor vehicles on roads associated with photovoltaic charging stations, including parameters such as the number of vehicles passing per unit time, traffic density, and vehicle type distribution. This traffic flow data is used to predict fluctuations in charging demand and establish a correlation model between traffic flow and electricity consumption.

[0021] Weather forecast data refers to the predicted meteorological parameters at the location of the photovoltaic power generation unit, including light intensity, ambient temperature, precipitation probability, wind speed, etc. It is used to predict photovoltaic power generation and assess the impact of weather on power generation efficiency.

[0022] The energy storage module status data is based on the real-time operating parameters of each module in the energy storage system, including state of charge, health status, operating temperature, voltage, current, and remaining life index. It is used to evaluate the available capacity, health status, and priority ranking of energy storage modules.

[0023] Charging demand data refers to data related to user charging behavior, including historical charging time, charging power, charging duration, and power usage pattern feature vectors. It is used to predict future charging demand and optimize energy allocation strategies.

[0024] S20: Establish a multi-source fusion prediction model based on the data to calculate the predicted power consumption and photovoltaic power generation.

[0025] Specifically, the multi-source fusion prediction model adopts a "time period-multi-dimensional" modeling approach, dividing the charging period into regular metering periods and metering periods awaiting approval. It uses time series models and vehicle flow correlation models to predict electricity consumption, and constructs a photovoltaic power generation prediction model through weather forecast data to achieve two-way prediction of electricity consumption and power generation.

[0026] The construction method of the multi-source fusion prediction model is: using the time series model for the regular measurement period ,in is the predicted electricity consumption in time period Dj, is the average power consumption in the historical T cycles, is the forecast value of the previous period, is the adaptive weight; the traffic flow correlation model is used for the approved metering period ,in Predict traffic flow for associated routes, is the correlation coefficient, is the user's electricity consumption characteristic value, is the climate correction factor, 、 are model parameters.

[0027] Determine the radius with the charging station as the center The coverage length within the range is greater than The one-way roads are the intended roads, and the ratio of traffic volume to electricity consumption of each intended road is calculated. , through discrete values ,choose The smallest road, where bp is the bs mean, es is the electricity consumption, vs is the traffic volume, and n is the data volume.

[0028] The calculation model of photovoltaic power generation is: ,in is the power generation at time t, For overall efficiency, is the photovoltaic panel area, is the light intensity, is the ambient temperature, is the standard temperature, is the temperature coefficient.

[0029] S30: Calculate a priority index based on the state of charge, health status, temperature, and remaining life of the energy storage module.

[0030] Specifically, the priority index model quantifies the available priority of energy storage modules during the charging and discharging process through weighted calculation of multi-dimensional parameters. The core parameters include state of charge (SOC), state of health (SOH), operating temperature and remaining life index. A comprehensive assessment of the module status is achieved through weight distribution.

[0031] The calculation model of the energy storage module priority index is: ,in, is the priority of module s at time t, is the state of charge, For health status, is the temperature, is the optimal operating temperature, L(s) is the remaining life index, 、 、 、 is the weight coefficient.

[0032] The present invention takes an energy storage system containing three second-life battery modules as an example. Given the charge state, health state, operating temperature, remaining life index and weight coefficient of each module, the priority index of each module can be calculated and sorted by substituting them into the formula. Therefore, high-priority modules are activated first during discharge, and the current of low-priority modules is limited during charging. In this way, the difference in module lifespan is reduced and the overall operating efficiency of the system is improved.

[0033] S40: Generate an energy allocation strategy based on the net load power demand and priority index to control the coordinated power supply of photovoltaics, energy storage and the grid.

[0034] Specifically, the energy allocation strategy is generated as follows: Calculate net load power ,in To predict power consumption, is the photovoltaic power generation power at time t; when ≤0, energy storage charging power , where Z is the full charge capacity of the energy storage, is the average state of charge, For charging efficiency; when >0, if Only energy storage is used for power supply. The grid power supply ,in, is the real-time energy storage power, 、 is the threshold ratio, The lower limit of safety.

[0035] The optimization strategy also includes the step of optimizing the energy allocation strategy through reinforcement learning, and the reward function of the reinforcement learning is , where R is the reward value, Cost is the operating cost, Emission is the carbon emission, and Satisfaction is the user satisfaction. 、 、 is the weight coefficient.

[0036] The energy storage module status data includes the state of charge, health status, voltage, current and temperature parameters of the cascade utilization battery pack. The vehicle flow data is collected through video recognition and RFID technology. The weather forecast data includes the light intensity, temperature and precipitation probability in the future preset time period.

[0037] Energy storage module status data covers the state of charge (SOC), state of health (SOH), voltage, current, and temperature parameters of the second-use battery pack. SOC reflects the percentage of the battery's current charge to rated capacity, SOH indicates the battery's health and remaining service life, voltage and current parameters monitor the battery's charge and discharge status, and temperature parameters are related to battery safety and performance stability. Traffic flow data is collected using video recognition and RFID technology. Specifically, high-definition cameras deployed on roads, combined with computer vision algorithms (such as the YOLO object detection model), identify vehicles and count traffic. RFID readers read the electronic tags of new energy vehicles, capturing information such as vehicle model and driving trajectory, providing data support for charging demand forecasts. Weather forecast data includes light intensity, temperature, and precipitation probability for a preset time period (e.g., 72 hours). Light intensity directly affects photovoltaic power generation, temperature parameters are used to calibrate the photovoltaic panel efficiency model, and precipitation probability is incorporated into electricity demand forecasts as a climate-influencing factor. Data accuracy and timeliness are ensured by accessing a national meteorological data API and real-time monitoring at on-site weather stations.

[0038] The energy allocation strategy also includes a storage module screening mechanism based on a priority index, which selects storage modules from high to low according to P(s, t) to participate in charging and discharging. When the state of charge of the module is lower than a preset threshold or the health state is lower than a preset threshold, the module exits operation.

[0039] The steps of controlling photovoltaic, energy storage and grid coordinated power supply include: Instructions are sent to the switch management module via the high-speed controller area network to control the on / off status of the grid and the inverter, and to adjust the output power of the charging pile. The inverter efficiency is used as a correction factor in the power calculation.

[0040] Specifically, energy allocation strategy instructions are transmitted to the switch management module via a high-speed controller area network to achieve precise control of the on / off status of the grid and inverter. When the system requires grid intervention for power supply, the switch management module receives the instruction and closes the corresponding circuit breaker, connecting the grid to the power supply circuit. If photovoltaic and energy storage can meet the power demand, the grid connection is disconnected, and clean energy is used first. At the same time, the output power of the charging pile is adjusted according to the energy allocation strategy to ensure that each charging pile charges the vehicle at the preset power. In the power calculation process, the inverter efficiency is included as a correction factor. For example, when calculating the power that can be supplied by photovoltaic power generation, "photovoltaic power generation × inverter efficiency" is used as the actual available power to avoid power calculation deviations caused by ignoring inverter losses, thereby achieving coordinated and efficient power supply of photovoltaic, energy storage and grid, and improving the overall energy utilization of the system.

[0041] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A photovoltaic power supply management method, characterized in that: Includes: Obtain traffic flow data, weather forecast data, energy storage module status data, and charging demand data; Based on the above data, a multi-source fusion prediction model is established to calculate the predicted power consumption and photovoltaic power generation; Calculate the priority index based on the state of charge, health status, temperature and remaining life of the energy storage module; Generate an energy allocation strategy based on net load power demand and priority index to control the coordinated power supply of photovoltaics, energy storage and power grid.

2. The photovoltaic power management method according to claim 1, characterized in that: The construction method of the multi-source fusion prediction model is: using the time series model for the regular measurement period ,in is the predicted electricity consumption in time period Dj, is the average power consumption in the historical T cycles, is the predicted value of the previous period, α is the adaptive weight; Use the traffic flow correlation model for the approved metering period ,in Predict traffic flow for associated routes, is the correlation coefficient, is the user's electricity consumption characteristic value, is the climate correction coefficient, and β and γ are model parameters.

3. The photovoltaic power management method according to claim 2, characterized in that: The method for determining the associated line is: taking the charging station as the center to define the radius The coverage length within the range is greater than The one-way roads are the intended roads, and the ratio of traffic volume to electricity consumption of each intended road is calculated. , through discrete values ,choose The smallest road, where bp is the bs mean, es is the electricity consumption, vs is the traffic volume, and n is the data volume.

4. The photovoltaic power management method according to claim 1, characterized in that: The calculation model of photovoltaic power generation is: ,in is the photovoltaic power generation power at time t, For overall efficiency, is the photovoltaic panel area, is the light intensity, is the ambient temperature, is the standard temperature, is the temperature coefficient.

5. The photovoltaic power management method according to claim 1, characterized in that: The calculation model of the energy storage module priority index is: ,in is the priority of module s at time t, is the state of charge, For health status, is the temperature, is the optimal operating temperature, L(s) is the remaining life index, 、 、 、 is the weight coefficient.

6. The photovoltaic power management method according to claim 1, characterized in that: The energy allocation strategy is generated as follows: Calculate net load power ,in To predict power consumption, is the photovoltaic power generation power; when ≤0, energy storage charging power , where Z is the full charge capacity of the energy storage, is the average state of charge, For charging efficiency; when >0, if Only energy storage is used for power supply. The grid power supply ,in, is the real-time energy storage power, 、 is the threshold ratio, The lower limit of safety.

7. The photovoltaic power management method according to claim 1, characterized in that: The step of optimizing the energy allocation strategy by reinforcement learning is also included, wherein the reward function of the reinforcement learning is , where R is the reward value, Cost is the operating cost, Emission is the carbon emission, and Satisfaction is the user satisfaction. 、 、 is the weight coefficient.

8. The photovoltaic power management method according to claim 1, characterized in that: The energy storage module status data includes the state of charge, health status, voltage, current and temperature parameters of the cascade utilization battery pack. The vehicle flow data is collected through video recognition and RFID technology. The weather forecast data includes the light intensity, temperature and precipitation probability in the future preset time period.

9. The photovoltaic power management method according to claim 6, characterized in that: The energy allocation strategy also includes an energy storage module screening mechanism based on a priority index, which selects energy storage modules from high to low according to P(s, t) to participate in charging and discharging, and exits operation when the state of charge is lower than a preset threshold or the health state is lower than a preset threshold.

10. The photovoltaic power management method according to claim 1, characterized in that: The steps of controlling photovoltaic, energy storage and grid coordinated power supply include: Instructions are sent to the switch management module via the high-speed controller area network to control the on / off status of the grid and the inverter, and to adjust the output power of the charging pile. The inverter efficiency is used as a correction factor in the power calculation.

Images