Mobile communication base station energy-saving control system based on Ai

Abstract

The invention relates to the technical field of communication base station energy-saving control, in particular to an Ai-based mobile communication base station energy-saving control system, which comprises an energy supply parameter presetting module, a signal transmission quality analysis module, an equipment operation analysis module, a user state analysis module, an energy supply calibration module and a calibration control module, according to the method, mean value and fluctuation trend dual modeling is carried out on the power consumption data in the historical time period, the initial energy supply parameter in the current monitoring time period is generated by combining the weight coefficient and mapping to the predefined state level table, and then the signal transmission quality parameter, the equipment operation parameter and the connection user parameter of the base station in the current monitoring time period are monitored. A signal transmission influence value, an equipment influence value and a user connection influence value are respectively calculated, when any influence value exceeds a preset reference interval, an energy supply influence value is calculated, and an initial energy supply parameter is combined to generate a calibration energy supply parameter through difference operation, so that accurate prediction and on-demand energy supply of the energy consumption demand of the base station are realized.

Description

Technical Field

[0001] This invention relates to the field of energy-saving control technology for communication base stations, specifically to an AI-based energy-saving control system for mobile communication base stations. Background Technology

[0002] With the rapid evolution of 5G and next-generation mobile communication technologies, the coverage density of communication networks and the number of base stations deployed have increased significantly. The energy consumption of mobile communication base stations in communication networks has continued to rise, and the energy-saving control of base stations has received increasing attention. Especially in application scenarios with multiple users and high concurrent access and frequent dynamic changes in service load, the existing base station power supply strategies generally rely on fixed power configuration or manual adjustment, lacking the ability to perceive and respond dynamically to the base station's operating status in real time. Traditional power supply control methods are usually based on static configuration, which makes it difficult to effectively cope with energy consumption fluctuations caused by factors such as changes in the number of user connections, signal transmission fluctuations, and changes in equipment operating status. This results in base stations maintaining high power consumption during non-high load periods, causing a waste of power resources. At the same time, this type of control method also has an adverse impact on system operating efficiency and service quality, and cannot meet the flexible and efficient energy consumption management requirements of the next generation of intelligent communication networks.

[0003] To address the aforementioned shortcomings, a technical solution is provided. Summary of the Invention

[0004] The purpose of this invention is to solve the problems in the existing energy-saving control process of mobile communication base stations, which rely on fixed power configuration or manual adjustment, resulting in delayed response and lack of real-time perception and dynamic control of the actual operating status of the base station. Therefore, this invention proposes an AI-based energy-saving control system for mobile communication base stations.

[0005] The objective of this invention can be achieved through the following technical solutions: An AI-based energy-saving control system for mobile communication base stations includes: The power supply parameter preset module is used to obtain the power consumption data of the historical period corresponding to the current monitoring period of the base station, calculate the average value of the power consumption data at different time points in the historical period as the average value of the historical period data, determine the first trend coefficient of the power consumption data of the historical period based on the average value of the historical period data, calculate the fluctuation of the power consumption data at different time points in the historical period, determine the second trend coefficient of the power consumption data of the historical period based on the fluctuation, and determine the power supply parameters for the current monitoring period based on the first trend coefficient and the second trend coefficient of the power consumption data of the historical period. The signal transmission quality analysis module is used to monitor the signal transmission quality parameters of the base station during the current monitoring period, obtain transmission fluctuation values, throughput fluctuation values, and bit error rate fluctuation values, and determine the impact value on signal transmission. The equipment operation analysis module is used to monitor the equipment operation parameters of the base station during the current monitoring period and determine the equipment impact value. The user status analysis module is used to monitor the connected user parameters of the base station during the current monitoring period and determine the impact value of connected users. The signal transmission impact value, equipment impact value, and connected user impact value are compared with the corresponding comparison reference range. If they are outside the comparison reference range, the power supply calibration module is executed. The power supply calibration module is used to extract the values ​​of signal transmission influence, equipment influence, and connection user influence, perform normalization processing to obtain the power supply influence value, and determine the calibration power supply parameters based on the power supply influence value and power supply parameters. The calibration control module is used to control and adjust the calibration power parameters of the base station during the current monitoring period by controlling the execution device.

[0006] Furthermore, the specific process for solving the first trend coefficient and the second trend coefficient is as follows: The mean values ​​of historical data for different historical periods are sorted according to time sequence to obtain a mean sequence. The mean sequence is then decomposed according to the STL time series decomposition method to obtain the slope of the trend term as the first trend coefficient. The maximum absolute value of the difference between the power consumption data at different time points within a historical period and the mean of the historical period data is determined as the maximum difference. The product of the maximum difference and the preset difference coefficient is calculated as the fluctuation coefficient. The fluctuation coefficients of each historical period are sorted according to the time sequence to obtain the fluctuation sequence. The fluctuation sequence is decomposed according to the STL time series decomposition method to obtain the slope of the trend term as the second trend coefficient.

[0007] Furthermore, the specific process for solving the energy supply parameters is as follows: The first and second trend coefficients of historical power consumption data are multiplied by their corresponding weighting coefficients and then summed to obtain the overall trend coefficient. The overall trend coefficient is then compared and matched with the stored overall trend status determination table to obtain the overall trend status level. Each overall trend coefficient corresponds to an overall trend status level. At the same time, it is matched with the energy supply parameters corresponding to the overall trend status level to obtain the energy supply parameters for the historical period, which are then used as the energy supply parameters for the current monitoring period.

[0008] Furthermore, the specific process for solving the signal transmission influence value is as follows: The values ​​of transmission fluctuation (BD1), throughput fluctuation (BD2), and bit error rate fluctuation (BD3) are extracted and normalized according to the formula: The signal transmission influence value θ1 is obtained; Among them, BD1 * BD2 *and BD3 * These are respectively represented as reference transmission fluctuation value, reference throughput fluctuation value, and reference bit error rate fluctuation value. λ1, λ2, and λ3 represent the weighting coefficients of transmission fluctuation deviation value, throughput fluctuation difference value, and bit error rate fluctuation deviation value, respectively, and λ1 > λ2 > λ3.

[0009] Furthermore, the specific process for solving the transmission fluctuation value is as follows: The signal transmission strength at each monitoring time point during the current monitoring period of the base station is obtained. The monitoring time point is used as the abscissa and the signal transmission strength is used as the ordinate. A two-dimensional rectangular coordinate system is established by connecting the signal transmission strength of adjacent monitoring time points with line segments in chronological order. The midpoint of each line segment in the two-dimensional rectangular coordinate system is obtained and used as the center of a circle. A circle is drawn with half the length of each line segment as the radius to obtain an annulus and the area of ​​the annulus is calculated. The difference between the areas of adjacent annulus is calculated to obtain the annulus surface difference value. At the same time, the annulus surface difference value is averaged to obtain the annulus surface difference average value. The annulus surface difference average value is used as the transmission fluctuation value of the base station during the current monitoring period.

[0010] Furthermore, the specific process for solving the throughput fluctuation value is as follows: The throughput of each monitoring time point in the current monitoring period of the base station is obtained. The throughput of each monitoring time point in the current monitoring period of the base station is compared and analyzed with a preset throughput threshold. If the throughput of a certain monitoring time point in the current monitoring period of the base station is greater than or equal to the preset throughput threshold, the throughput of that monitoring time point in the current monitoring period of the base station is marked as peak throughput. If the throughput of a certain monitoring time point in the current monitoring period of the base station is less than the preset throughput threshold, the throughput of that monitoring time point in the current monitoring period of the base station is marked as stable throughput. The number of peak throughput and stable throughput is counted, and the number of peak throughput is divided by the number of stable throughput to obtain the throughput fluctuation value, which is used as the throughput fluctuation value of the base station in the current monitoring period.

[0011] Furthermore, the specific process for solving the bit error rate fluctuation value is as follows: Obtain the bit error rate (BER) at each monitoring time point during the current monitoring period of the base station. Mark the BER at the initial monitoring time point of the current monitoring period of the base station as the initial BER. Then, calculate the difference between the BER at other monitoring time points and the initial BER to obtain the BER difference value. Calculate the average of all the obtained BER differences to obtain the BER fluctuation value during the current monitoring period of the base station.

[0012] Furthermore, the specific process for solving the equipment influence value is as follows: Obtain the main equipment temperature at each monitoring time point during the current monitoring period of the base station and mark it as SW. i ; Extract the average temperature of the main equipment from the historical periods corresponding to the current monitoring period, use it as the reference main equipment temperature for the current monitoring period of the base station, and label it as SW. * ; Based on the formula: The equipment impact value θ2 is obtained, where i represents the number of each monitoring time point, m represents the total number of monitoring time point numbers, and SXW i p1 represents the temperature change value of the main equipment, p2 represents the influence factor when the temperature of the main equipment at the i-th monitoring time point of the current monitoring period of the base station is greater than the corresponding reference main equipment temperature, and p2 represents the influence factor when the temperature of the main equipment at the i-th monitoring time point of the current monitoring period of the base station is less than the corresponding reference main equipment temperature. R represents the natural constant.

[0013] Furthermore, the specific process for solving the influence value of connected users is as follows: Obtain the number of connected users at each monitoring time point during the current monitoring period of the base station, and arrange the number of connected users at each monitoring time point in descending order to obtain a connected user number sequence. Then, filter out the values ​​of the maximum connected user number, minimum connected user number, median connected user number, and mode connected user number, and label them as YH respectively. max YH min YH med YH mod Among them, the median number of connected users refers to the median of the number of connected users in the current monitoring period of the base station, and the mode number of connected users refers to the mode of the number of connected users in the current monitoring period of the base station. Based on the formula: The influence value θ3 of the connected users is obtained. The reference number of connected users is denoted as e, which represents the natural constant. μ1, μ2, μ3, and μ4 represent the influence factors corresponding to the maximum, minimum, median, and mode connected user numbers, respectively.

[0014] Furthermore, the specific process of solving for the calibration power supply parameters is as follows: Obtain the power supply parameters for the current monitoring period of the base station, match the power supply impact value of the current monitoring period of the base station with the power supply impact parameters corresponding to each set power supply impact value to obtain the power supply impact parameters for the current monitoring period of the base station, and calculate the difference between the power supply parameters and the power supply impact parameters for the current monitoring period of the base station to obtain the difference between the power supply parameters and the power supply impact parameters for the current monitoring period of the base station, which is used as the calibration power supply parameters for the current monitoring period of the base station.

[0015] The technical solution provided by this invention has the following advantages compared with the known prior art: 1. This invention, by performing dual modeling of the mean trend and fluctuation trend of historical power consumption data and combining weight coefficients to map to a predefined state level table, can automatically generate energy supply parameters that meet the needs of the current monitoring period, thereby realizing forward-looking prediction and on-demand energy supply for base station energy-saving control, and providing a quantitative and reliable initial setpoint for subsequent calibration control.

[0016] 2. This invention monitors the signal transmission quality parameters, equipment operating parameters, and user connection parameters of the base station during the current monitoring period. It calculates the signal transmission impact value, equipment impact value, and user connection impact value, respectively. Then, it normalizes and weights these three types of impact values ​​to obtain a power supply impact value that comprehensively reflects the base station's operational pressure. Next, it matches this power supply impact value with the set power supply impact parameters corresponding to each power supply impact value to obtain the corresponding power supply impact parameters. Finally, it subtracts these parameters from the original power supply parameters to generate refined calibration power supply parameters. Based on these calibration parameters, the energy supply of the base station is dynamically adjusted, thereby achieving on-demand power supply and effectively avoiding energy waste during off-peak and low-load periods. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0018] Figure 1 This is an overall module block diagram of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0020] like Figure 1 As shown, the AI-based mobile communication base station energy-saving control system includes: a power supply parameter preset module, a signal transmission quality analysis module, an equipment operation analysis module, a user status analysis module, a power supply calibration module, and a calibration control module; The power supply parameter preset module is used to obtain the power consumption data of the historical period corresponding to the current monitoring period of the base station, calculate the average value of the power consumption data at different time points in the historical period as the average value of the historical period data, determine the first trend coefficient of the power consumption data of the historical period based on the average value of the historical period data, calculate the fluctuation of the power consumption data at different time points in the historical period, determine the second trend coefficient of the power consumption data of the historical period based on the fluctuation, and determine the power supply parameters for the current monitoring period based on the first trend coefficient and the second trend coefficient of the power consumption data of the historical period. It should be noted that the energy supply parameter is a numerical indicator used to characterize the overall energy supply demand of a mobile communication base station during the current monitoring period. This parameter is expressed in kilowatt-hours (kWh) or as a percentage. As a key reference value for base station energy-saving control, the energy supply parameter is derived by trend modeling and calculation based on the total power consumption of all operating equipment of the base station in historical periods. It reflects the expected energy consumption level of the base station in the current period and serves as the initial set point in the execution of energy-saving control strategies. In practical applications, this energy supply parameter serves as the target value for base station power consumption, guiding power output devices (such as uninterruptible power supplies, DC power supply modules) to set appropriate output power or voltage levels. At the same time, it is also used to determine the power mode switching strategies of various subsystems (such as BBU, RRU, transmission links, heat dissipation systems, etc.), including entering a sleep state, operating at reduced frequency, or maintaining a full-power active state, thereby achieving dynamic energy consumption control and synergistic optimization of operating efficiency. Among them, the power consumption data is the sum of the power consumption data of each operating device corresponding to the base station; In this embodiment of the invention, the average of power consumption data at different points in a historical period can be calculated as the average of historical period data; Optionally, in this embodiment of the invention, determining the first trend coefficient of historical power consumption data based on the average of historical time period data includes: sorting the average of historical time period data of different historical time periods in chronological order to obtain an average sequence; performing time-series decomposition on the average sequence based on the STL time-series decomposition method to obtain the slope of the trend term as the first trend coefficient. Among them, the STL time series decomposition method is a time series decomposition method that uses robust local weighted regression as a smoothing method. It decomposes the time series into corresponding trend terms, seasonal terms and residual terms. The STL time series decomposition method is a well-known time series decomposition method in this field, and will not be described in detail here. In this embodiment of the invention, the mean sequence is decomposed by STL to obtain the corresponding trend term curve. Then, the slope of the trend term is used as the first trend coefficient. Since the trend term can represent the overall trend of the corresponding time series, the slope of the trend term can represent the changing trend of the daily average data for different days. That is to say, the first trend coefficient can represent the changing trend of the historical average data for different historical periods. Optionally, in this embodiment of the invention, calculating the fluctuation of power consumption data at different points in a historical period and determining the second trend coefficient of the historical period based on the fluctuation includes: determining the maximum absolute value of the difference between the power consumption data at different points in the historical period and the mean of the historical period data as the maximum difference; calculating the product of the maximum difference and a preset difference coefficient as the fluctuation coefficient; sorting the fluctuation coefficients of each historical period in chronological order to obtain a fluctuation sequence; and performing chronological decomposition on the fluctuation sequence based on the STL chronological decomposition method to obtain the slope of the trend term as the second trend coefficient. In this embodiment of the invention, the fluctuation coefficients of each historical period are sorted according to the time sequence to obtain the fluctuation sequence. The fluctuation sequence is then decomposed according to the STL time sequence decomposition method to obtain the slope of the trend term as the second trend coefficient. This process is similar to the STL time sequence decomposition method for the fluctuation sequence described above, and will not be repeated here. Thus, the second trend coefficient is obtained. Optionally, in this embodiment of the invention, the energy supply parameters for the current monitoring period are determined based on the first trend coefficient and the second trend coefficient of the historical power consumption data. Specifically, the first trend coefficient and the second trend coefficient of the historical power consumption data are multiplied by their corresponding weight coefficients and then added together to obtain the total trend coefficient. The total trend coefficient is compared and matched with the stored total trend status determination table to obtain the total trend status level. Each obtained total trend coefficient corresponds to a total trend status level. At the same time, it is matched with the energy supply parameters corresponding to the total trend status level to obtain the energy supply parameters for the historical period, which are used as the energy supply parameters for the current monitoring period.

[0021] The signal transmission quality analysis module is used to monitor the signal transmission quality parameters of the base station during the current monitoring period, and thereby analyze the impact of the signal transmission on the base station during the current monitoring period. The specific analysis process is as follows: By acquiring the signal transmission strength at each monitoring time point during the current monitoring period of the base station, the signal transmission strength at each monitoring time point during the current monitoring period is obtained. With each monitoring time point as the abscissa and the signal transmission strength as the ordinate, a two-dimensional rectangular coordinate system is established by connecting the signal transmission strength of adjacent monitoring time points in chronological order with line segments. The midpoint of each line segment in the two-dimensional rectangular coordinate system is obtained and used as the center of a circle. A circle is drawn with half the length of each line segment as the radius to obtain an annulus and the area of ​​the annulus is calculated. The difference between the areas of adjacent annulus is calculated to obtain the annulus surface difference value. At the same time, the annulus surface difference value is averaged to obtain the annulus surface difference average value. The annulus surface difference average value is used as the transmission fluctuation value of the base station during the current monitoring period. By acquiring the throughput at each monitoring time point during the current monitoring period of the base station, the throughput at each monitoring time point during the current monitoring period of the base station is obtained. The throughput at each monitoring time point during the current monitoring period of the base station is compared and analyzed with a preset throughput threshold. If the throughput at a certain monitoring time point during the current monitoring period of the base station is greater than or equal to the preset throughput threshold, the throughput at that monitoring time point during the current monitoring period of the base station is marked as peak throughput. If the throughput at a certain monitoring time point during the current monitoring period of the base station is less than the preset throughput threshold, the throughput at that monitoring time point during the current monitoring period of the base station is marked as stable throughput. The number of peak throughput and stable throughput is counted, and the number of peak throughput is divided by the number of stable throughput to obtain the throughput fluctuation value, which is used as the throughput fluctuation value of the base station during the current monitoring period. By acquiring the bit error rate at each monitoring time point during the current monitoring period of the base station, the bit error rate at each monitoring time point during the current monitoring period of the base station is obtained. The bit error rate at the initial monitoring time point during the current monitoring period of the base station is marked as the initial bit error rate. Then, the bit error rate at other monitoring time points is calculated with the initial bit error rate to obtain the bit error rate difference. The average of all the obtained bit error rate differences is calculated to obtain the bit error rate fluctuation value during the current monitoring period of the base station. The signal transmission quality parameters are composed of the transmission fluctuation value, throughput fluctuation value, and bit error rate fluctuation value during the current monitoring period of the base station; Extract the transmission fluctuation value, throughput fluctuation value, and bit error rate fluctuation value from the signal transmission quality parameters, and label them as BD1, BD2, and BD3 respectively. Then, normalize these three values ​​according to the formula: The signal transmission influence value θ1 is obtained; Among them, BD1 * BD2 * and BD3 * These are respectively represented as reference transmission fluctuation value, reference throughput fluctuation value, and reference bit error rate fluctuation value, and λ1, λ2, and λ3 represent the weighting coefficients of transmission fluctuation deviation value, throughput fluctuation difference value, and bit error rate fluctuation deviation value, respectively, and λ1 > λ2 > λ3; The signal transmission impact value is compared and analyzed with a preset comparison reference range. If the signal transmission impact value is outside the preset comparison reference range, the power supply calibration module is executed, and the obtained signal transmission impact value is sent to the power supply calibration module.

[0022] The equipment operation analysis module is used to monitor the equipment operation parameters of the base station during the current monitoring period, and thereby analyze the equipment impact status of the base station during the current monitoring period. The specific analysis process is as follows: By acquiring the main equipment temperature at each monitoring time point during the current monitoring period of the base station, the main equipment temperature at each monitoring time point during the current monitoring period of the base station is obtained and marked as SW.i ; Extract the average temperature of the main equipment from the historical periods corresponding to the current monitoring period, use it as the reference main equipment temperature for the current monitoring period of the base station, and label it as SW. * ; Based on the formula: The equipment impact value θ2 is obtained, where i represents the number of each monitoring time point, m represents the total number of monitoring time point numbers, and SXW i p1 represents the temperature change value of the main equipment, p2 represents the influence factor when the temperature of the main equipment at the i-th monitoring time point of the current monitoring period of the base station is greater than the corresponding reference main equipment temperature, and p2 represents the influence factor when the temperature of the main equipment at the i-th monitoring time point of the current monitoring period of the base station is less than the corresponding reference main equipment temperature. R represents the natural constant.

[0023] The device impact value is compared and analyzed with a preset comparison reference range. If the device impact value is outside the preset comparison reference range, the power supply calibration module is executed, and the obtained device impact value is sent to the power supply calibration module.

[0024] The user status analysis module is used to monitor the connected user parameters of the base station during the current monitoring period, and thereby analyze the impact of connected users on the status of the base station during the current monitoring period. The specific analysis process is as follows: The number of connected users at each monitoring time point during the current monitoring period of the base station is obtained. Arrange the number of connected users at each monitoring time point during the current monitoring period of the base station in descending order to obtain the sequence of connected user numbers for the current monitoring period. Then, select the values ​​of the maximum, minimum, median, and mode connected user numbers and label them as YH respectively. max YH min YH med YH mod Among them, the median number of connected users refers to the median of the number of connected users in the current monitoring period of the base station, and the mode number of connected users refers to the mode of the number of connected users in the current monitoring period of the base station. Based on the formula: The influence value θ3 of the connected users is obtained. The reference number of connected users is represented by , e represents the natural constant, and μ1, μ2, μ3, and μ4 represent the influence factors corresponding to the maximum number of connected users, minimum number of connected users, median number of connected users, and mode number of connected users, respectively. The connection user impact value is compared and analyzed with a preset comparison reference range. If the connection user impact value is outside the preset comparison reference range, the power supply calibration module is executed, and the obtained connection user impact value is sent to the power supply calibration module.

[0025] The power supply calibration module receives the signal transmission impact value, equipment impact value, and connected user impact value of the base station during the current monitoring period. It then analyzes the power supply impact value of the base station during the current monitoring period to obtain the power supply impact value. The specific analysis process is as follows: The values ​​of signal transmission impact θ1, equipment impact θ2, and connected user impact θ3 during the current monitoring period of the base station are extracted and normalized according to the formula: The power supply impact value GNP of the base station during the current monitoring period is obtained, and β1, β2, and β3 represent the impact factors corresponding to the signal transmission impact value, equipment impact value, and connected user impact value, respectively. The power supply parameters of the base station during the current monitoring period are obtained, and the calibration power supply parameters of the base station during the current monitoring period are analyzed based on the power supply impact value and power supply parameters of the base station during the current monitoring period. The specific analysis is as follows: The power supply impact value of the current monitoring period of the base station is matched with the power supply impact parameters corresponding to each set power supply impact value to obtain the power supply impact parameters of the current monitoring period of the base station. The difference between the power supply parameters of the current monitoring period of the base station and the power supply impact parameters is obtained as the calibration power supply parameters of the current monitoring period of the base station.

[0026] The calibration control module is used to control and adjust the calibration power parameters of the base station during the current monitoring period by controlling the execution device.

[0027] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to specific implementations. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. An AI-based energy-saving control system for mobile communication base stations, characterized in that, include: The power supply parameter preset module is used to obtain the power consumption data of the historical period corresponding to the current monitoring period of the base station, calculate the average value of the power consumption data at different time points in the historical period as the average value of the historical period data, determine the first trend coefficient of the power consumption data of the historical period based on the average value of the historical period data, calculate the fluctuation of the power consumption data at different time points in the historical period, determine the second trend coefficient of the power consumption data of the historical period based on the fluctuation, and determine the power supply parameters for the current monitoring period based on the first trend coefficient and the second trend coefficient of the power consumption data of the historical period. The signal transmission quality analysis module is used to monitor the signal transmission quality parameters of the base station during the current monitoring period, obtain transmission fluctuation values, throughput fluctuation values, and bit error rate fluctuation values, and determine the impact value on signal transmission. The equipment operation analysis module is used to monitor the equipment operation parameters of the base station during the current monitoring period and determine the equipment impact value. The user status analysis module is used to monitor the connected user parameters of the base station during the current monitoring period and determine the impact value of connected users. The signal transmission impact value, equipment impact value, and connected user impact value are compared with the corresponding comparison reference range. If they are outside the comparison reference range, the power supply calibration module is executed. The power supply calibration module is used to extract the values ​​of signal transmission influence, equipment influence, and connection user influence, perform normalization processing to obtain the power supply influence value, and determine the calibration power supply parameters based on the power supply influence value and power supply parameters. The calibration control module is used to control and adjust the calibration power parameters of the base station during the current monitoring period by controlling the execution device.

2. The AI-based mobile communication base station energy-saving control system according to claim 1, characterized in that, The specific process for calculating the first and second trend coefficients is as follows: The mean values ​​of historical data for different historical periods are sorted according to time sequence to obtain a mean sequence. The mean sequence is then decomposed according to the STL time series decomposition method to obtain the slope of the trend term as the first trend coefficient. The maximum absolute value of the difference between the power consumption data at different time points within a historical period and the mean of the historical period data is determined as the maximum difference. The product of the maximum difference and the preset difference coefficient is calculated as the fluctuation coefficient. The fluctuation coefficients of each historical period are sorted according to the time sequence to obtain the fluctuation sequence. The fluctuation sequence is decomposed according to the STL time series decomposition method to obtain the slope of the trend term as the second trend coefficient.

3. The AI-based mobile communication base station energy-saving control system according to claim 1, characterized in that, The specific process for solving the energy supply parameters is as follows: The first and second trend coefficients of historical power consumption data are multiplied by their corresponding weighting coefficients and then summed to obtain the overall trend coefficient. The overall trend coefficient is then compared and matched with the stored overall trend status determination table to obtain the overall trend status level. Each overall trend coefficient corresponds to an overall trend status level. At the same time, it is matched with the energy supply parameters corresponding to the overall trend status level to obtain the energy supply parameters for the historical period, which are then used as the energy supply parameters for the current monitoring period.

4. The AI-based mobile communication base station energy-saving control system according to claim 1, characterized in that, The specific process for solving the signal transmission influence value is as follows: The values ​​of transmission fluctuation (BD1), throughput fluctuation (BD2), and bit error rate fluctuation (BD3) are extracted and normalized according to the formula: The signal transmission influence value θ1 is obtained; Wherein, BD1*, BD2* and BD3* represent the reference transmission fluctuation value, the reference throughput fluctuation value and the reference bit error rate fluctuation value, respectively, and λ1, λ2 and λ3 represent the weighting coefficients of the transmission fluctuation deviation value, the throughput fluctuation difference value and the bit error rate fluctuation deviation value, respectively, and λ1 > λ2 > λ3.

5. The AI-based mobile communication base station energy-saving control system according to claim 4, characterized in that, The specific process for solving the transmission fluctuation value is as follows: The signal transmission strength at each monitoring time point during the current monitoring period of the base station is obtained. The monitoring time point is used as the abscissa and the signal transmission strength is used as the ordinate. A two-dimensional rectangular coordinate system is established by connecting the signal transmission strength of adjacent monitoring time points with line segments in chronological order. The midpoint of each line segment in the two-dimensional rectangular coordinate system is obtained and used as the center of a circle. A circle is drawn with half the length of each line segment as the radius to obtain an annulus and the area of ​​the annulus is calculated. The difference between the areas of adjacent annulus is calculated to obtain the annulus surface difference value. At the same time, the annulus surface difference value is averaged to obtain the annulus surface difference average value. The annulus surface difference average value is used as the transmission fluctuation value of the base station during the current monitoring period.

6. The AI-based mobile communication base station energy-saving control system according to claim 4, characterized in that, The specific process for solving the throughput fluctuation value is as follows: The throughput of each monitoring time point in the current monitoring period of the base station is obtained. The throughput of each monitoring time point in the current monitoring period of the base station is compared and analyzed with a preset throughput threshold. If the throughput of a certain monitoring time point in the current monitoring period of the base station is greater than or equal to the preset throughput threshold, the throughput of that monitoring time point in the current monitoring period of the base station is marked as peak throughput. If the throughput of a certain monitoring time point in the current monitoring period of the base station is less than the preset throughput threshold, the throughput of that monitoring time point in the current monitoring period of the base station is marked as stable throughput. The number of peak throughput and stable throughput is counted, and the number of peak throughput is divided by the number of stable throughput to obtain the throughput fluctuation value, which is used as the throughput fluctuation value of the base station in the current monitoring period.

7. The AI-based mobile communication base station energy-saving control system according to claim 4, characterized in that, The specific process for calculating the bit error rate fluctuation value is as follows: Obtain the bit error rate (BER) at each monitoring time point during the current monitoring period of the base station. Mark the BER at the initial monitoring time point of the current monitoring period of the base station as the initial BER. Then, calculate the difference between the BER at other monitoring time points and the initial BER to obtain the BER difference value. Calculate the average of all the obtained BER differences to obtain the BER fluctuation value during the current monitoring period of the base station.

8. The AI-based mobile communication base station energy-saving control system according to claim 1, characterized in that, The specific process for solving the equipment influence value is as follows: Obtain the main equipment temperature at each monitoring time point during the current monitoring period of the base station and mark it as SWi; Extract the average temperature of the main equipment corresponding to the historical period of the current monitoring period, use it as the reference temperature of the main equipment for the current monitoring period of the base station, and mark it as SW*; Based on the formula: The equipment influence value θ2 is obtained, where i represents the number of each monitoring time point, m represents the total number of monitoring time point numbers, SXWi represents the temperature change value of the main equipment, p1 represents the influence factor when the temperature of the main equipment at the i-th monitoring time point of the current monitoring period of the base station is greater than the corresponding reference main equipment temperature, p2 represents the influence factor when the temperature of the main equipment at the i-th monitoring time point of the current monitoring period of the base station is less than the corresponding reference main equipment temperature, and R represents the natural constant.

9. The AI-based mobile communication base station energy-saving control system according to claim 1, characterized in that, The specific process for solving the influence value of connected users is as follows: The number of connected users at each monitoring time point during the current monitoring period of the base station is obtained, and the number of connected users at each monitoring time point is arranged in descending order to obtain a sequence of connected users. The maximum number of connected users, the minimum number of connected users, the median number of connected users, and the mode number of connected users are selected from the sequence and labeled as YHmax, YHmin, YHmed, and YHmod, respectively. The median number of connected users refers to the median of the number of connected users in the sequence of connected users during the current monitoring period of the base station, and the mode number of connected users refers to the mode of the number of connected users in the sequence of connected users during the current monitoring period of the base station. Based on the formula: The influence value θ3 of the connected users is obtained. The reference number of connected users is denoted as e, which represents the natural constant. μ1, μ2, μ3, and μ4 represent the influence factors corresponding to the maximum, minimum, median, and mode connected user numbers, respectively.

10. The AI-based mobile communication base station energy-saving control system according to claim 1, characterized in that, The specific process for solving the calibration power supply parameters is as follows: Obtain the power supply parameters for the current monitoring period of the base station, match the power supply impact value of the current monitoring period of the base station with the power supply impact parameters corresponding to each set power supply impact value to obtain the power supply impact parameters for the current monitoring period of the base station, and calculate the difference between the power supply parameters and the power supply impact parameters for the current monitoring period of the base station to obtain the difference between the power supply parameters and the power supply impact parameters for the current monitoring period of the base station, which is used as the calibration power supply parameters for the current monitoring period of the base station.

Images