An efficient load power supply configuration method based on intelligent optimization of photovoltaic power generation and base station integration

Abstract

This invention relates to the field of energy storage technology and discloses a method for efficiently configuring load power supply by integrating photovoltaic power generation with base stations based on intelligent optimization. The method comprises: obtaining photovoltaic power generation forecast data, charge and discharge loss characteristics of energy storage equipment, and power consumption data of base station loads; obtaining an optimized time step based on load demand satisfaction; constructing a scheduling-energy storage dual-objective optimization model based on the Ant Lion Optimization Algorithm, with energy scheduling and energy storage management as dual optimization objectives; iteratively updating the scheduling-energy storage dual-objective optimization model to obtain an optimal solution; and configuring base station load power supply based on the optimal solution. This application can optimize the dual objectives of energy scheduling and energy storage management through the Ant Lion Optimization Algorithm to maximize energy utilization efficiency and minimize energy storage device losses.

Description

Technical Field

[0001] The present invention relates to the field of energy storage technology, and in particular to a high-efficiency load power supply configuration method based on intelligent optimization and integration of photovoltaic power generation and base stations. Background Art

[0002] With the increasing global demand for sustainable energy and the rapid expansion of communications network infrastructure, ensuring stable base station operation while achieving energy conservation and emission reduction has become a critical issue. Traditional base stations primarily rely on mains electricity, requiring diesel generators in remote areas. This not only increases operating costs but also places significant pressure on the environment. Furthermore, due to Massive MIMO technology and high frequency band characteristics, 5G base stations consume significantly more power than 4G base stations, further exacerbating the imbalance between energy supply and demand.

[0003] In recent years, the development of photovoltaic power generation technology has provided a new solution to this problem, especially in remote areas far away from the power grid, such as mountainous areas and deserts. Base stations mainly rely on photovoltaic power generation to power their base stations. During the day, the base station load directly uses photovoltaic power and stores excess power in energy storage devices. At night, the energy storage devices are used to supplement electricity to reduce dependence on municipal electricity. By integrating the photovoltaic power generation system with the base station, energy utilization efficiency can be significantly improved, carbon emissions can be reduced, and power supply reliability can be enhanced.

[0004] However, traditional PV-base station fusion systems have significant drawbacks: a mismatch between energy supply and demand. PV power generation is intermittent, mismatching with peak base station power demand. Frequent charging and discharging of energy storage batteries significantly shortens their lifespan, impacting system stability. Traditional PV systems employ a fixed power supply priority (e.g., PV power first, utility power backup), ignoring real-time electricity prices and battery status. This leads to extensive energy management, resulting in frequent charging and discharging of energy storage batteries, significantly shortening their lifespan. Furthermore, they fail to consider base station load characteristics (e.g., the periodic start and stop of air conditioners and the constant power consumption of transmission equipment), resulting in low energy utilization. Furthermore, they lack multi-source data fusion (weather forecasts, grid electricity prices, load forecasts), making advance scheduling impossible. Therefore, PV-base station fusion requires further optimization to meet demand. Summary of the Invention

[0005] The present invention aims to provide an efficient load power supply configuration method based on intelligent optimization of photovoltaic power generation and base station integration, so as to optimize the dual objectives of energy scheduling and energy storage management through the ant lion optimization algorithm to maximize energy utilization efficiency and minimize the loss of energy storage equipment.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A high-efficiency load power supply configuration method based on intelligent optimization and integration of photovoltaic power generation and base stations, comprising:

[0008] S1, obtains photovoltaic power generation forecast data, energy storage equipment charging and discharging loss characteristics, and base station load power consumption data;

[0009] S2, based on the photovoltaic power generation forecast data and the base station load power consumption data, obtains the load demand satisfaction, and obtains the optimized time step based on the load demand satisfaction;

[0010] S3, based on photovoltaic power generation forecast data, energy storage equipment charging and discharging loss characteristics, and base station load electricity consumption data, takes energy scheduling and energy storage management as dual optimization goals, builds a scheduling-energy storage dual-objective optimization model based on the Ant Lion optimization algorithm, iteratively updates the scheduling-energy storage dual-objective optimization model to obtain the optimal solution, and configures the base station load power supply based on the optimal solution.

[0011] The principles and advantages of this solution are as follows: In practical applications, obtaining photovoltaic power generation forecast data, energy storage equipment charge and discharge loss characteristics, and base station load power consumption data facilitates accurate understanding of basic supply, demand, and loss data, laying the foundation for subsequent optimization. Determining the optimized time step based on load demand satisfaction facilitates dynamic adjustment of the time step based on real-time data to ensure optimal system performance. Precise control of the time step enables more precise control of energy scheduling and energy storage management, improving overall efficiency, response speed, and stability. By optimizing energy scheduling and energy storage management as dual objectives, the system maximizes photovoltaic utilization and minimizes utility power procurement costs. It also optimizes the charging and discharging strategies of energy storage equipment, extending its service life and reducing loss costs. Due to the Ant Lion Optimization Algorithm's strong global search capabilities, a scheduling-energy storage dual-objective optimization model is constructed based on the Ant Lion Optimization Algorithm. This model is then iteratively updated to obtain the optimal solution, enabling the global optimal solution to be found in complex multidimensional spaces. Dynamic parameter adjustment allows for refined local search to improve solution quality. The Ant Lion Optimization Algorithm can also flexibly handle different constraints and objective functions, enhancing the overall performance and adaptability of the solution. Finally, based on the optimal solution, base station load power supply configuration is performed to effectively improve efficiency and reliability, while reducing operating costs and maximizing green energy utilization. For base stations in remote areas, this can alleviate grid coverage issues and reduce the frequency of diesel generator use. For base stations in high-energy-consuming urban areas, it can reduce reliance on utility power. For base stations in special environments, such as those in high-temperature and high-humidity areas or the dusty northwest, dynamic adjustments can improve weather resistance.

[0012] Preferably, as an improvement, the S2 further includes:

[0013] S21, construct the load demand satisfaction model as follows:

[0014]

[0015]

[0016] in, is the predicted photovoltaic power generation within time step t, is the predicted discharge of energy storage device i in time step t, is the predicted base station load demand within time step t, is the loss value within time step t;

[0017] S22, obtain the peak value of the load demand satisfaction, and use the t value of the most recent bottom peak as the time step.

[0018] Technical effect: It is easy to reduce the number of optimizations, reduce resource waste, and meet the needs at the same time.

[0019] Preferably, as an improvement, the base station load demand is a dynamic demand determined according to the load priority and load demand satisfaction, and the determination model is:

[0020]

[0021] in, To meet the needs of all base station loads, ~ To meet the needs of the current priority load.

[0022] Technical effect: Load demands vary in different time periods. By dynamically adjusting load demands, resources can be optimized more accurately.

[0023] Preferably, as an improvement, the S3 further includes:

[0024] S31, initialize the population and randomly generate a set of initial energy scheduling and energy storage management plans;

[0025] S32, defining a comprehensive fitness function and constraints, wherein the comprehensive fitness function includes:

[0026]

[0027] in, represents the comprehensive fitness function, and is the weight coefficient, and are the energy scheduling fitness function and energy storage management fitness function, is the penalty function;

[0028] S33, calculate the quality of each solution based on the comprehensive fitness function;

[0029] S34, build a trap based on the quality of the current solution, and let the ants move randomly in the trap, gradually moving closer to the center of the trap;

[0030] S35, outputting the optimal solution after iterative updating for a preset number of times.

[0031] Technical effect: It is convenient to simulate the hunting process of ant lions in nature through the ant lion optimization algorithm to perform dual-objective optimization.

[0032] Preferably, as an improvement, the energy scheduling fitness function includes:

[0033]

[0034] in, =[ ], represents the photovoltaic utilization ratio, Indicates the proportion of energy storage equipment used, 、 are weight coefficients used to balance the utilization rate of photovoltaic power generation and the cost of city electricity, is the photovoltaic power generation, is the base station load demand, is the capacity of the energy storage device i, It is the mains electricity price.

[0035] Technical effect: Maximize the utilization rate of photovoltaic power generation, make full use of photovoltaic electricity, and reduce dependence on mains electricity.

[0036] Preferably, as an improvement, the energy storage management fitness function includes:

[0037]

[0038] in, =[ ], Indicates the charging power ratio, Indicates the discharge power ratio, is the common sense coefficient used to balance the loss cost of energy storage equipment, is the charging power of energy storage device i, is the discharge power of energy storage device i.

[0039] Technical effect: Based on the above objective function, it is easy to minimize operating costs, reduce electricity procurement costs and losses of energy storage equipment.

[0040] Preferably, as an improvement, the constraint condition is:

[0041]

[0042]

[0043]

[0044]

[0045] in, is the photovoltaic power generation, is the discharge capacity of i energy storage devices, is the base station load demand, and are the minimum and maximum power of energy storage device i, () indicates the priority of energy storage equipment, is the charging power of energy storage device i, is the discharge power of energy storage device i, is the energy storage loss of device i in time step t.

[0046] Technical effect: Through constraints, the load demand of the base station is ensured to be always met, and the power of the energy storage device is within the threshold range, thereby ensuring the stable operation of the base station, ensuring that the load demand is always met, and avoiding power outages.

[0047] Preferably, as an improvement, the penalty function is:

[0048]

[0049] in, represents the penalty function, m is the number of constraints, is the penalty coefficient, is the degree of violation of the mth constraint.

[0050] Technical effect: It is convenient to increase the corresponding penalty value when the solution does not meet the constraints.

[0051] Preferably, as an improvement, the and For dynamic weights, the dynamic adjustment model includes:

[0052]

[0053]

[0054] in, is the weight adjustment function.

[0055] Technical effect: Facilitates prioritizing different objectives at different stages.

[0056] Preferably, as an improvement, the dynamic adjustment parameters of the ant lion optimization algorithm include:

[0057]

[0058]

[0059] Among them, a and b are dynamically adjusted parameters. and is the initial value, is the current iteration number, and N is the maximum iteration number.

[0060] Technical effect: Balances the ability of global search and local search, and gradually reduces the search range by two dynamically adjusted parameters with the number of iterations. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 Schematic diagram of the structure of an embodiment of the present invention. DETAILED DESCRIPTION

[0062] The following is further described in detail through specific implementation methods:

[0063] The embodiment is basically as shown in the attached Figure 1 As shown, a high-efficiency load power supply configuration method based on intelligent optimization and integration of photovoltaic power generation and base stations includes:

[0064] S1, obtains photovoltaic power generation forecast data, energy storage equipment charging and discharging loss characteristics, and base station load power consumption data;

[0065] For photovoltaic power generation forecast data, parameters such as light intensity and temperature are usually obtained through meteorological stations, satellite images, and historical data, and predictions are made using time series analysis methods (such as ARIMA and LSTM). The charging and discharging loss characteristics of energy storage equipment include the technical specifications provided by the energy storage equipment manufacturer, including maximum charging power, maximum discharge power, charging and discharging efficiency, etc. Based on the technical parameters and actual operating data of the energy storage equipment, a charging and discharging loss model is established. For example, an empirical formula is used to fit the charging and discharging loss function. The power consumption data of base station loads includes historical power consumption data recorded by the base station management system (BMS). Through statistical analysis of historical data, the load change pattern is identified, and future load is predicted based on business needs.

[0066] S2, based on the load demand satisfaction, obtains the optimized time step, that is, analyzes the load demand satisfaction under different time steps and selects the optimal time step to ensure that the system is always in the best state, thereby more accurately controlling energy scheduling and energy storage management and improving the overall efficiency of the system. S2 also includes:

[0067] S21, construct the load demand satisfaction model as follows:

[0068]

[0069]

[0070] in, is the predicted photovoltaic power generation within time step t, is the predicted discharge of energy storage device i in time step t, is the predicted base station load demand within time step t, is the loss value within time step t.

[0071] The base station load demand is a dynamic demand determined according to the load priority and load demand satisfaction, and the determination model is:

[0072]

[0073] in, To meet the needs of all base station loads, ~ To meet the needs of the current priority load.

[0074] Since the photovoltaic power generation data and load demand are both forecast data, the longer the time, the lower the accuracy.

[0075] Adjust the error, for:

[0076]

[0077] in, is a constant.

[0078] S22, obtaining the peak value of the load demand satisfaction, and using the t value of the most recent bottom peak as the time step. That is, when the load demand satisfaction is at a local minimum, the t value is used as the time step.

[0079] S3, with energy scheduling and energy storage management as dual optimization objectives, constructs a scheduling-energy storage dual-objective optimization model based on the Ant Lion optimization algorithm, iteratively updates the scheduling-energy storage dual-objective optimization model to obtain the optimal solution, and configures the base station load power supply based on the optimal solution. S3 also includes:

[0080] S31, initialize the population and randomly generate a set of initial energy scheduling and energy storage management plans.

[0081] S32, defining a comprehensive fitness function and constraints, wherein the comprehensive fitness function includes:

[0082]

[0083] in, represents the comprehensive fitness function, and is the weight coefficient, and are the energy scheduling fitness function and energy storage management fitness function, is the penalty function;

[0084] The energy scheduling fitness function includes:

[0085]

[0086] in, =[ ], represents the photovoltaic utilization ratio, Indicates the proportion of energy storage equipment used, 、 are weight coefficients used to balance the utilization rate of photovoltaic power generation and the cost of city electricity, is the photovoltaic power generation, is the base station load demand, is the capacity of the energy storage device i, It is the mains electricity price.

[0087] The energy storage management fitness function includes:

[0088]

[0089] in, =[ ], Indicates the charging power ratio, Indicates the discharge power ratio, is the common sense coefficient used to balance the loss cost of energy storage equipment, is the charging power of energy storage device i, is the discharge power of energy storage device i.

[0090] The constraints are:

[0091]

[0092]

[0093]

[0094]

[0095] in, is the photovoltaic power generation, is the discharge capacity of i energy storage devices, is the base station load demand, and are the minimum and maximum power of energy storage device i, () indicates the priority of energy storage equipment, is the charging power of energy storage device i, is the discharge power of energy storage device i, is the energy storage loss of device i in time step t.

[0096] By adding a penalty term, it is convenient to increase the corresponding penalty value when the solution does not meet the constraint conditions. The penalty function is:

[0097]

[0098] in, represents the penalty function, m is the number of constraints, is the penalty coefficient, is the degree of violation of the mth constraint.

[0099] described and For dynamic weights, the dynamic adjustment model includes:

[0100]

[0101]

[0102] in, is the weight adjustment function.

[0103] S33, calculates the quality of each solution based on the comprehensive fitness function. The comprehensive fitness function is an indicator used to quantify the quality of each solution. In the dual-objective optimization problem of energy scheduling and energy storage management, this process determines which solutions are closer to the optimal solution. For example, consider three different solutions 、 、 ,in, of =0.8, =0.9, =0.1; of =0.7, =0.85, no constraint violation, i.e. =0; of =0.6, =0.95, =0.05; and are 0.6 and 0.4 respectively, then the solution is The score is 0.6×0.8+0.4×0.9-0.1=0.74, solution The score is 0.6×0.7+0.4×0.85-0=0.76, solution The score is 0.6×0.6+0.4×0.95-0.05=0.71, so in this example, the solution is It is considered to be the solution with the best quality in this iteration due to its high comprehensive fitness score.

[0104] S34: Build a trap based on the quality of the current solution, and let the ants move randomly inside the trap, gradually moving closer to the center of the trap. This includes:

[0105] Constructing traps: Each trap corresponds to an antlion (high-quality solution). The position of the antlion determines the center of the trap. The center of the trap is:

[0106]

[0107] in, is the weight of the cth solution, which is inversely proportional to its fitness, is the trap center location of the vth decision variable.

[0108] Hunting: The ants move randomly in the trap, gradually moving closer to the center of the trap, and the position of the ants is updated after each movement:

[0109]

[0110] Among them, r is a random number, a and b are dynamically adjusted parameters, specifically:

[0111]

[0112]

[0113] in, and is the initial value, is the current iteration number, and N is the maximum iteration number.

[0114] S35: After a preset number of iterations, the optimal solution is output. Using an antlion optimization algorithm, the system simulates the natural hunting process of ant lions, constructing traps to capture ants and gradually finding the optimal solution. Based on this optimal solution, the base station load power supply is configured to achieve a reasonable ratio of photovoltaic, energy storage, and utility power, maximizing photovoltaic utilization and ensuring stable base station operation. The charging and discharging strategies of the energy storage equipment are also optimized to extend its service life and reduce losses.

[0115] The above is only an embodiment of the present invention, and the common knowledge such as the specific technical solutions and / or characteristics in the solution are not described in detail here. It should be pointed out that for those skilled in the art, without departing from the technical solution of the present invention, several variations and improvements can be made, which should also be regarded as the scope of protection of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the description can be used to interpret the content of the claims.

Claims

1. A high-efficiency load power supply configuration method based on intelligent optimization of photovoltaic power generation and base station integration, characterized in that: include: S1, obtains photovoltaic power generation forecast data, energy storage equipment charging and discharging loss characteristics, and base station load power consumption data; S2, obtaining load demand satisfaction based on photovoltaic power generation forecast data and base station load power consumption data, and obtaining an optimized time step based on the load demand satisfaction; S2 also includes: S21, construct the load demand satisfaction model as follows: in, is the predicted photovoltaic power generation within time step t, is the predicted discharge of energy storage device i in time step t, is the predicted base station load demand within time step t, is the loss value within time step t; S22, obtain the peak value of load demand satisfaction and use the t value of the most recent bottom peak as the time step; S3, based on photovoltaic power generation forecast data, energy storage equipment charge and discharge loss characteristics, and base station load power consumption data, with energy scheduling and energy storage management as dual optimization objectives, constructs a scheduling-energy storage dual-objective optimization model based on the Ant Lion optimization algorithm, iteratively updates the scheduling-energy storage dual-objective optimization model to obtain an optimal solution, and configures base station load power supply based on the optimal solution; S3 also includes: S31, initialize the population and randomly generate a set of initial energy scheduling and energy storage management plans; S32, defining a comprehensive fitness function and constraints, wherein the comprehensive fitness function includes: in, represents the comprehensive fitness function, and is the weight coefficient, and are the energy scheduling fitness function and energy storage management fitness function respectively, is the penalty function; S33, calculate the quality of each solution based on the comprehensive fitness function; S34, build a trap based on the quality of the current solution, and let the ants move randomly in the trap, gradually moving closer to the center of the trap; S35, outputting the optimal solution after iterative updating for a preset number of times; The dynamic adjustment parameters of the Ant Lion optimization algorithm include: Among them, a and b represent the dynamically adjusted parameters. and is the initial value, is the current iteration number, and N is the maximum iteration number.

2. The method for configuring efficient load power supply based on intelligent optimization of photovoltaic power generation and base station integration according to claim 1, characterized in that: The base station load demand is a dynamic demand determined according to the load priority and load demand satisfaction, and the determination model is: in, To meet the needs of all base station loads, ~ To meet the needs of the current priority load.

3. The method for configuring efficient load power supply based on intelligent optimization of photovoltaic power generation and base station integration according to claim 1, characterized in that: The energy scheduling fitness function includes: in, =[ ], represents the photovoltaic utilization ratio, Indicates the proportion of energy storage equipment used, 、 are weight coefficients used to balance the utilization rate of photovoltaic power generation and the cost of city electricity, is the photovoltaic power generation, is the base station load demand, is the capacity of the energy storage device i, It is the mains electricity price.

4. The method for configuring efficient load power supply based on intelligent optimization of photovoltaic power generation and base station integration according to claim 1, characterized in that: The energy storage management fitness function includes: in, =[ ], Indicates the charging power ratio, Indicates the discharge power ratio, is the common sense coefficient used to balance the loss cost of energy storage equipment, is the charging power of energy storage device i, is the discharge power of energy storage device i.

5. The method for configuring efficient load power supply based on intelligent optimization of photovoltaic power generation and base station integration according to claim 1, characterized in that: The constraints are: in, is the photovoltaic power generation, is the discharge capacity of i energy storage devices, is the base station load demand, and are the minimum and maximum power of energy storage device i, () indicates the priority of energy storage equipment, is the charging power of energy storage device i, is the discharge power of energy storage device i, is the energy storage loss of device i in time step t.

6. The method for configuring efficient load power supply based on intelligent optimization of photovoltaic power generation and base station integration according to claim 1, characterized in that: The penalty function is: in, represents the penalty function, m is the number of constraints, is the penalty coefficient, is the degree of violation of the mth constraint.

7. The method for configuring efficient load power supply based on intelligent optimization of photovoltaic power generation and base station integration according to claim 1, characterized in that: described and For dynamic weights, the dynamic adjustment model includes: in, is the weight adjustment function.

Images