A 5G base station cooling system energy efficiency optimization method

Abstract

The present application belongs to the technical field of base station cooling system energy saving, and particularly relates to a 5G base station cooling system energy efficiency optimization method; the method comprises the following steps: calculating the system power of the cooling system running in different modes; according to the system power, the system running parameters of the cooling system under different outdoor environment temperatures and different base station loads are calculated by using a genetic algorithm, and a base station cooling system running mode partition map is constructed; the base station load and the outdoor environment temperature of the base station are acquired in real time, and the interpolation mode of the base station is judged according to the base station cooling system running mode partition map; the system running parameters of the current cooling system are calculated according to the interpolation mode of the base station by using a double-parameter linear interpolation method; the system running parameters of the cooling system are run, and the energy efficiency optimization of the cooling system is realized; the present application can realize the optimization calculation of the running parameters of the cooling system under any load and outdoor temperature conditions, reduce the system energy consumption, and solve the problem that a microcontroller unit cannot process complex codes.

Description

Technical Field

[0001] This invention belongs to the field of energy-saving technology for base station cooling systems, and specifically relates to a method for optimizing the energy efficiency of a 5G base station cooling system. Background Technology

[0002] With the continuous development of information technology, the amount of data in today's era is exploding. As a crucial hub connecting mobile clients and data centers, the number of 5G base stations is rapidly increasing. With the exponential growth of mobile data traffic, the energy consumption of 5G base stations is also increasing year by year. It is predicted that by 2025, there will be approximately 13.1 million base stations globally, consuming 200 billion kilowatt-hours of energy. Therefore, researching energy-saving technologies for 5G base stations is particularly urgent in this context.

[0003] In the energy consumption structure of 5G base stations, the main equipment and cooling system account for the highest proportions, with the two being roughly equal and their sum exceeding 90% of the total energy consumption of the base station. The main reason for the excessive energy consumption of 5G base stations is the high proportion of energy consumption in the cooling system. Therefore, energy-saving research on the cooling system of 5G base stations will effectively improve the overall energy efficiency of 5G base stations and promote the development of 5G technology towards a green and low-carbon direction.

[0004] 5G base stations are characterized by their large number, high power consumption, lack of maintenance, and complex and variable environments. Due to the dispersed nature of their deployment, the geographical environments in which base stations are located can be extremely complex, and the climate varies significantly between different locations. For example, base stations located in northern and southern my country have vastly different average daily temperatures and humidity levels. Current base station cooling systems fail to recognize these differences, operating solely according to pre-set parameters. Typically, the supply and return air temperatures are set empirically, and the cooling system's operating mode is switched based on the outdoor temperature. A PID feedback control method is used to regulate the base station temperature. While this method can meet temperature control requirements, it lacks consideration for load variations, and the switching of operating modes relies solely on outdoor temperature, resulting in imprecise switching criteria. This can lead to inappropriate operating mode selection and deviations from actual operating conditions, ultimately causing excessive energy consumption in the base station cooling system. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention proposes an energy efficiency optimization method for 5G base station cooling systems, which includes:

[0006] S1: Calculate the system power of the cooling system operating in different modes;

[0007] S2: Based on the system power, the system operating parameters of the cooling system under different outdoor ambient temperatures and different base station loads are calculated using a genetic algorithm, and a partition map of the base station cooling system operation mode is constructed.

[0008] S3: Real-time acquisition of base station load and outdoor ambient temperature, and determination of base station interpolation mode based on base station cooling system operation mode partitioning diagram;

[0009] S4: The system operating parameters of the current cooling system are calculated using a two-parameter linear interpolation method based on the interpolation mode of the base station.

[0010] S5: Cooling system operating parameters to optimize cooling system energy efficiency.

[0011] Preferably, in step S2, the process of calculating the system operating parameters using a genetic algorithm includes:

[0012] S21: Encode the system operating parameters as chromosomes in the genetic algorithm;

[0013] S22: Set the population size and initialize the chromosome population;

[0014] S23: Calculate the fitness of each individual in the current population based on the base station load and system power;

[0015] S24: Generate offspring populations by selecting, crossing over, and mutating chromosomes based on fitness;

[0016] S25: Determine whether the iteration stopping condition has been met. If it has, select the individual with the highest fitness in the current offspring population as the optimal system operating parameter. Otherwise, return to step S24.

[0017] Furthermore, the formula for calculating fitness is:

[0018]

[0019] Where F represents fitness, Q represents base station load, and W represents system power.

[0020] Preferably, in step S2, the process of constructing the base station cooling system operation mode partition map includes: plotting the system operation parameters of the cooling system under different outdoor ambient temperatures and different base station loads, with outdoor ambient temperature as the vertical axis and base station load as the horizontal axis, fitting the system operation mode on the graph, drawing the dividing line between the two operation modes, and obtaining the base station cooling system operation mode partition map.

[0021] Preferably, in step S3, the process of determining the interpolation mode of the base station includes: obtaining the upper and lower bounds of the real-time base station load; determining whether the upper and lower bounds belong to the same partition in the base station cooling system operation mode partitioning diagram based on the real-time outdoor ambient temperature; if they do, then determining that the base station is in the first interpolation mode; otherwise, determining that the base station is in the second interpolation mode.

[0022] Furthermore, step S4 specifically includes:

[0023] When the base station is in the first interpolation mode, the first system operating parameters of the current base station are calculated based on the real-time base station load, the lower limit of the real-time base station load, and the system operating parameters corresponding to the upper and lower limits of the real-time base station load.

[0024] When the base station is in the second interpolation mode, if the real-time base station load is closer to the lower boundary, an additional lower boundary value is added; the second system operating parameters of the current base station are calculated based on the real-time base station load, the lower boundary value of the real-time base station load, the system operating parameters corresponding to the lower boundary value, and the additional lower boundary value; if the real-time base station load is closer to the upper boundary, an additional upper boundary value is added, and the third system operating parameters of the current base station are calculated based on the real-time base station load, the upper boundary value of the real-time base station load, the system operating parameters corresponding to the upper boundary value, and the additional upper boundary value.

[0025] Furthermore, the formula for calculating the operating parameters of the first system is as follows:

[0026]

[0027] Where X′ represents the first system operating parameter, X l This represents the system operating parameter corresponding to the lower limit of the real-time base station load, where Q represents the real-time base station load. l X represents the lower bound of the real-time base station load. u X represents the system operating parameter corresponding to the upper bound of the real-time base station load. l This represents the system operating parameters corresponding to the lower bound of the real-time base station load.

[0028] Furthermore, the formula for calculating the operating parameters of the second system is as follows:

[0029]

[0030] Where X″ represents the operating parameters of the second system, X l This represents the system operating parameter corresponding to the lower bound of the real-time base station load, where Q represents the real-time base station load. l X represents the lower bound of the real-time base station load. ll This represents the system operating parameters corresponding to the additional lower bound value.

[0031] Furthermore, the formula for calculating the operating parameters of the third system is as follows:

[0032]

[0033] Where X″′ represents the operating parameters of the third system, Q represents the real-time base station load, and X u Q represents the system operating parameter corresponding to the upper bound of the real-time base station load. uX represents the lower bound of the real-time base station load. uu This represents the system operating parameters corresponding to the additional upper bound value.

[0034] The beneficial effects of this invention are as follows:

[0035] Since the existing control methods for 5G base stations rely on experience to set the base station temperature and use PID feedback regulation, they cannot simultaneously consider base station load and outdoor temperature changes to make operating mode decisions and optimize operating parameters. The cooling system energy efficiency optimization method proposed in this invention can realize the optimization calculation of cooling system operating parameters under any load and outdoor temperature conditions, thus solving the problem of high energy consumption in cooling system operation.

[0036] Due to the small scale and unattended nature of 5G base stations, microcontroller units are generally used. The processor performance, memory size, and storage space of these microcontrollers limit the complexity and scale of the algorithms that can be run. This invention proposes a two-parameter linear interpolation method that simplifies complex iterative optimization algorithms into linear interpolation calculations, solving the problem that microcontroller units cannot handle complex code. Attached Figure Description

[0037] Figure 1 This is a flowchart of the energy efficiency optimization method for the 5G base station cooling system in this invention;

[0038] Figure 2 This is a partition diagram of the base station cooling system operation mode in this invention;

[0039] Figure 3 This is a partition diagram of the base station cooling system operation mode when the base station is in the first interpolation mode in this invention;

[0040] Figure 4 This is a partition diagram of the base station cooling system operation mode when the real-time base station load is closer to the lower boundary in this invention;

[0041] Figure 5 This is a partition diagram of the base station cooling system operation mode when the real-time base station load is closer to the upper boundary in this invention. Detailed Implementation

[0042] 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 embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] This invention proposes a method for optimizing the energy efficiency of a 5G base station cooling system, such as... Figure 1 As shown, the method includes the following:

[0044] In this invention, the 5G base station cooling system is an air-cooled cooling system, which consists of components such as heat exchangers, compressors and fans, and can operate in natural cooling mode, mechanical cooling mode and combined cooling mode.

[0045] S1: Calculate the system power of the cooling system operating in different modes.

[0046] Wind turbine energy consumption modeling:

[0047] W fan =h(m)

[0048] f fan =g(m)

[0049] Among them, W fan f is the power of the wind turbine. fan Let m be the fan frequency, m represent the air mass flow rate, and h and g be polynomial functions.

[0050] Therefore, the evaporator fan model is as follows:

[0051] W fan-e =h e (m e )

[0052] f fan-e =g e (m e )

[0053] Condenser fan model:

[0054] W fan-c =h c (m c )

[0055] f fan-c =g c (m c )

[0056] In this context, the subscript 'e' indicates the evaporator and the subscript 'c' indicates the condenser.

[0057] Compressor energy consumption modeling:

[0058]

[0059] Among them, W comp Let represent the compressor power, w represent the compressor's theoretical work, η represent the compressor's performance coefficient, a(T) and b(T) be linear functions of the outdoor temperature T, and f comp This refers to the compressor frequency.

[0060] When the cooling system is operating in natural cooling mode, only the natural cooling cycle is running, and the system power is:

[0061] W = W fan-e +W fan-c

[0062] At this point, the heat transfer formula for the system is k(m e ,m c ,T e,o ,T c,o ,T r ) = 0, m e The mass flow rate of the evaporator is expressed in m. c T represents the air mass flow rate of the condenser. e,o T represents the temperature at the evaporator outlet. c,o T represents the temperature at the condenser outlet. r This indicates the temperature of the heat pipe working fluid.

[0063] When the cooling system operates in mechanical refrigeration mode, only the vapor compression cycle is running, and the system power is:

[0064] W = W fan-e +W fan-c +W comp

[0065] At this point, the heat transfer formula for the system is k(m e ,m c ,T e,o ,T c,o ,T r ,T me ,T mc W comp ) = 0, T c,s The temperature between the evaporator fan and the condenser fan is represented by T. me T represents the evaporation temperature of the vapor compression cycle. mc This indicates the condensation temperature of the vapor compression cycle.

[0066] When the cooling system operates in combined cooling mode, both the natural cooling cycle and the vapor compression cycle are running, and the system power is:

[0067] W = W fan-e +W fan-c +W comp

[0068] At this point, the heat transfer formula for the system is k(m e ,m c ,T e,o ,T c,o ,T c,s ,T r ,T me ,T mc Q fc Q mc Wcomp ) = 0, Q fc Q represents the power consumption in natural cooling mode. mc This indicates the power consumption of the mechanical refrigeration mode.

[0069] In system power, the unknown parameter f fan-c ,f fan-e m can be used c m e express.

[0070] S2: Based on the system power, the system operating parameters of the cooling system under different outdoor ambient temperatures and different base station loads are calculated using a genetic algorithm, and a partition map of the base station cooling system operation mode is constructed.

[0071] S21: Encode the system operating parameters as chromosomes in the genetic algorithm.

[0072] The individual chromosome construction described in this invention refers to the encoding of the solution space, transforming the solution to the energy efficiency optimization problem from the physical space to the gene-encoded data structure of a genetic algorithm. Firstly, the solution space is constructed. The solution to the cooling system energy efficiency optimization problem refers to the system operating parameters that achieve optimal overall system energy efficiency under given base station load and outdoor ambient temperature conditions. These parameters include the operating frequencies of the terminal fans (evaporator fans), outdoor fans (condenser fans), and compressors. The solution space can be defined as {f}. fan-c ,f fan-e ,f comp}, when f comp = 0 indicates natural cooling mode; otherwise, it indicates other operating modes. The individual chromosome structure described in this invention uses Gray encoding for solution space transformation, with each individual being a string consisting of 0s and 1s.

[0073] S22: Set the population size and initialize the chromosome population.

[0074] To set the number of individuals in the population, in order to ensure that the initial population is as evenly distributed as possible across the search domain, f is set... fan-e and f comp The regulation range is divided into multiple regions, and multiple individuals are randomly generated in equal numbers in each region to form the initial population.

[0075] S23: Calculate the fitness of each individual in the current population based on the base station load and system power.

[0076] As can be seen from the aforementioned steps, the system power can be calculated by using the system operating parameters and the outdoor ambient temperature to determine the system power in different operating modes.

[0077] This invention uses the energy efficiency of the base station cooling system as the fitness factor, and the formula is:

[0078]

[0079] Where F represents fitness, Q represents base station load (system heat exchange), and W represents system power.

[0080] A higher fitness value indicates that the individual carries superior genes. It's important to note that the fitness function solution process involves obtaining not only the fitness value but also intermediate parameters of the cooling system and the outdoor fan frequency f. fan-c The solution will also be found. When intermediate parameters do not conform to physical common sense or the outdoor fan speed exceeds the adjustable range (i.e., there is no physical solution), such individuals need to be identified and their fitness value reduced so that such individuals are eliminated as much as possible during the calculation process.

[0081] S24: Select, crossover, and mutate chromosomes based on fitness to generate offspring populations.

[0082] Selection: A roulette wheel selection method is used, a replay-based random sampling method. It selects individuals based on their fitness value and the sum of the fitness values ​​of all individuals in the population. The selection probability is the ratio of an individual's fitness value to the sum of the fitness values ​​of all individuals in the population. Therefore, individuals with higher fitness values ​​have a higher probability of being selected, and their superior genes are more likely to spread in the offspring. Conversely, individuals with lower fitness values ​​are more likely to be eliminated during the population's reproduction process, and their undesirable genes gradually decrease or even disappear in the offspring. For a population N = {a1, a2, ..., an} containing n individuals, the fitness value of individual Ai is denoted as Fit(Ai), and its probability of being selected can be expressed as follows:

[0083]

[0084] Crossover: Single-point crossover. The operation process is to pair up the individuals selected by the selection operator, randomly select a crossover position on the chromosome, and exchange chromosomes at the crossover point to generate two new individuals. It is worth noting that the crossover operator is not executed every time. It is also performed according to probability, and the value range is generally 0.4 to 0.99.

[0085] Mutation: Mutation operators can generate new genes for the population, enhancing the robustness of the algorithm. Similar to biological evolution in nature, mutation always occurs with an extremely low probability, typically set to 0.001–0.1.

[0086] S25: Determine whether the iteration stopping condition has been met. If it has, select the individual with the highest fitness in the current offspring population as the optimal system operating parameter. Otherwise, return to step S24.

[0087] Determine if the iteration stopping condition has been met. Preferably, the iteration stopping condition is a preset maximum number of iterations or a preset maximum execution time. If met, select the individual with the highest fitness in the current offspring population and decode it to obtain the solution to the energy efficiency optimization problem {f}. fan-e ,f fan-c ,f comp}

[0088] By selecting different integer base station loads, preferably ranging from 3000W to 10000W with an interval of 1000, the optimal system operating parameters of the base station under different outdoor temperatures can be calculated using the above genetic algorithm under each base station load condition, thus obtaining all system operating parameters of the cooling system under different outdoor ambient temperatures and different base station loads.

[0089] The process of constructing a base station cooling system operation mode zoning diagram includes:

[0090] The operating parameters of the cooling system under different outdoor ambient temperatures and base station loads are plotted on the vertical axis (outdoor ambient temperature) and the horizontal axis (base station load). The operating modes of the system parameters are obtained, and these modes are fitted onto the graph to draw two boundary lines between the two operating modes (the boundary lines represent the outdoor ambient temperatures separating the two different operating modes). This results in a zoning map of the base station cooling system's operating modes, as shown below. Figure 2 As shown, Zone I below the blue line indicates that the system is in natural cooling mode, Zone II between the blue and red lines indicates that the system is in combined cooling mode, and Zone III above the red line indicates that the system is in mechanical cooling mode.

[0091] S3: Real-time acquisition of base station load and outdoor ambient temperature, and determination of base station interpolation mode based on base station cooling system operation mode partitioning diagram.

[0092] Obtain the upper and lower bounds of the real-time base station load; based on the real-time outdoor ambient temperature, determine whether the upper and lower bounds belong to the same partition in the base station cooling system operation mode partitioning diagram; if they do, determine that the base station is in the first interpolation mode; otherwise, determine that the base station is in the second interpolation mode. For example: Figure 3 As shown, the intersection points of the upper and lower bounds of the real-time outdoor ambient temperature and the real-time base station load are both located in the middle region of the two lines, meaning that the upper and lower bounds belong to the same zone. Figure 4 As shown, the intersection of the upper limit of the real-time outdoor ambient temperature and the real-time base station load is located in the middle area of ​​the two lines, while the intersection of the lower limit of the real-time outdoor ambient temperature and the real-time base station load is located below the two lines, meaning that the upper and lower limits do not belong to the same zone.

[0093] S4: The system operating parameters of the current cooling system are calculated using a two-parameter linear interpolation method based on the interpolation mode of the base station.

[0094] When the base station is in the first interpolation mode, the first system operating parameters of the current base station are calculated based on the real-time base station load, the lower bound of the real-time base station load, and the system operating parameters corresponding to the upper and lower bounds of the real-time base station load; the calculation formula is:

[0095]

[0096] Where X′ represents the first system operating parameter, X l This represents the system operating parameter corresponding to the lower limit of the real-time base station load, where Q represents the real-time base station load. l X represents the lower bound of the real-time base station load. u X represents the system operating parameter corresponding to the upper bound of the real-time base station load. l This represents the system operating parameters corresponding to the lower bound of the real-time base station load.

[0097] When the base station is in the second interpolation mode, such as Figure 4 As shown, if the real-time base station load is closer to the lower boundary, that is... Then add an additional lower bound value, preferably Q. ll =Q l -1000; The second system operating parameters of the current base station are calculated based on the real-time base station load, the lower bound of the real-time base station load, the system operating parameters corresponding to the lower bound, and the additional lower bound; the calculation formula is:

[0098]

[0099] Where X″ represents the operating parameters of the second system, X ll This represents the system operating parameters corresponding to the additional lower bound value.

[0100] like Figure 5 As shown, if the real-time base station load is closer to the upper boundary, that is... Then an additional upper bound value is added; preferably, this additional upper bound value is Q. uu =Q u +1000; The third system operating parameter of the current base station is calculated based on the real-time base station load, the upper limit of the real-time base station load, the system operating parameters corresponding to the upper limit, and the additional upper limit; the calculation formula is:

[0101]

[0102] Where X″′ represents the operating parameters of the third system, Q u X represents the lower bound of the real-time base station load.uu This represents the system operating parameters corresponding to the additional upper bound value.

[0103] S5: Cooling system operating parameters to optimize cooling system energy efficiency.

[0104] The base station's cooling system obtains all system operating parameters calculated by the genetic algorithm, and runs the above-mentioned two-parameter linear interpolation method on the base station. The cooling system operates by obtaining the corresponding system operating parameters through the above calculation process, which can realize the energy efficiency optimization of the cooling system.

[0105] The above-described embodiments further illustrate the purpose, technical solution, and advantages of the present invention. It should be understood that the above-described embodiments are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made to the present invention within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for optimizing the energy efficiency of a 5G base station cooling system, characterized in that, include: S1: Calculate the system power of the cooling system operating in different modes; S2: Based on the system power, the system operating parameters of the cooling system under different outdoor ambient temperatures and different base station loads are calculated using a genetic algorithm, and a partition map of the base station cooling system operation mode is constructed. S3: Real-time acquisition of base station load and outdoor ambient temperature, and determination of base station interpolation mode based on base station cooling system operation mode partitioning diagram; The process of determining the interpolation mode of a base station includes: obtaining the upper and lower bounds of the real-time base station load; determining whether the upper and lower bounds belong to the same partition in the base station cooling system operation mode partitioning diagram based on the real-time outdoor ambient temperature; if they do, the base station is determined to be in the first interpolation mode; otherwise, the base station is determined to be in the second interpolation mode. S4: The system operating parameters of the current cooling system are calculated using a two-parameter linear interpolation method based on the interpolation mode of the base station. When the base station is in the first interpolation mode, the first system operating parameters of the current base station are calculated based on the real-time base station load, the lower limit of the real-time base station load, and the system operating parameters corresponding to the upper and lower limits of the real-time base station load. When the base station is in the second interpolation mode, if the real-time base station load is closer to the lower boundary, an additional lower boundary value is added; the second system operating parameters of the current base station are calculated based on the real-time base station load, the lower boundary value of the real-time base station load, the system operating parameters corresponding to the lower boundary value, and the additional lower boundary value; if the real-time base station load is closer to the upper boundary, an additional upper boundary value is added; the third system operating parameters of the current base station are calculated based on the real-time base station load, the upper boundary value of the real-time base station load, the system operating parameters corresponding to the upper boundary value, and the additional upper boundary value. S5: Cooling system operating parameters to optimize cooling system energy efficiency.

2. The energy efficiency optimization method for a 5G base station cooling system according to claim 1, characterized in that, In step S2, the process of calculating the system operating parameters using a genetic algorithm includes: S21: Encode the system operating parameters as chromosomes in the genetic algorithm; S22: Set the population size and initialize the chromosome population; S23: Calculate the fitness of each individual in the current population based on the base station load and system power; S24: Generate offspring populations by selecting, crossing over, and mutating chromosomes based on fitness; S25: Determine whether the iteration stopping condition has been met. If it has, select the individual with the highest fitness in the current offspring population as the optimal system operating parameter. Otherwise, return to step S24.

3. The energy efficiency optimization method for a 5G base station cooling system according to claim 2, characterized in that, The formula for calculating fitness is: ； in, Indicates fitness. Indicates real-time base station load. This indicates the system power.

4. The energy efficiency optimization method for a 5G base station cooling system according to claim 1, characterized in that, In step S2, the process of constructing the base station cooling system operation mode partition map includes: plotting the system operation parameters of the cooling system under different outdoor ambient temperatures and different base station loads, with outdoor ambient temperature as the vertical axis and base station load as the horizontal axis, fitting the system operation mode on the graph, drawing the dividing line between the two operation modes, and obtaining the base station cooling system operation mode partition map.

5. The energy efficiency optimization method for a 5G base station cooling system according to claim 1, characterized in that, The formula for calculating the operating parameters of the first system is: ； in, Indicates the operating parameters of the first system. This represents the system operating parameters corresponding to the lower limit of real-time base station load. Indicates real-time base station load. This represents the lower bound of the real-time base station load. This represents the system operating parameters corresponding to the upper limit of the real-time base station load.

6. The energy efficiency optimization method for a 5G base station cooling system according to claim 1, characterized in that, The formula for calculating the operating parameters of the second system is: ； in, Indicates the operating parameters of the second system. This represents the system operating parameters corresponding to the lower bound of the real-time base station load. Indicates real-time base station load. This represents the lower bound of the real-time base station load. This represents the system operating parameters corresponding to the additional lower bound value.

7. The energy efficiency optimization method for a 5G base station cooling system according to claim 1, characterized in that, The formula for calculating the operating parameters of the third system is: ； in, Indicates the operating parameters of the third system. Indicates real-time base station load. This represents the system operating parameters corresponding to the upper bound of the real-time base station load. This represents the lower bound of the real-time base station load. This represents the system operating parameters corresponding to the additional upper bound value.

Images