A Demand Response Optimization Method and System for the Load of 5G Base Station Clusters

Abstract

The present invention discloses a demand response optimization method and system for the load of 5G base station clusters, which relates to the technical fields of power systems and communication networks, and includes obtaining grid node information, 5G base station cluster information, and task allocation information within a region; establishing an operator revenue function according to a preset initial incentive electricity price, 5G base station cluster information, and task allocation information, and calculating the current incentive amount; establishing a grid expenditure function according to the current incentive amount, grid node information, and task allocation information, and calculating the current generator output; establishing a total operating cost function with the goal of minimizing the total operating cost, calculating the current incentive electricity price to update the initial incentive electricity price until the function converges to obtain the optimal incentive electricity price. The present invention reasonably allocates computing tasks, increases the incentive amount obtained by the operator, reduces the power generation cost of the grid, minimizes the grid expenditure cost while maximizing the operator's revenue, and improves energy efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems and communication networks, and more specifically, to a demand response optimization method and system for the load of 5G base station clusters. Background Art

[0002] A 5G base station cluster is a new type of hybrid active load group, which is connected to the load through a low-voltage distribution network and mainly includes four parts: one is the electric power for supporting signal transmission and reception, such as the power consumption of active antenna units; the second is the electric power for supporting various protocol calculations and resource scheduling calculations of the communication network, such as the power consumption of baseband processing units; the third is the electric power for supporting the maintenance of the base station working environment, such as air conditioners, lighting, etc.; the fourth is the battery energy storage inside the base station. There is a high-speed communication network, such as an optical fiber network, between 5G base stations. When the computing load is not high and the quality of user services is guaranteed, the computing tasks can be transferred from local allocation to the background cloud computing platform or other surrounding base stations for calculation, and the relevant electrical loads generated along with the computing load are also synchronously transferred between the base station and the background cloud service, and between the base station clusters. Since each base station and the background cloud service in the base station cluster are connected to different power grid nodes, and even different power supply areas, when different incentives are provided for different base stations, it will affect the computing task transfer situation of the base station management party, and then realize the transfer of relevant electrical loads between different power grid nodes or regions, and realize the adjustable automation of the base station electrical load.

[0003] The demand-side response based on 5G base station clusters can not only reduce the energy consumption of 5G base stations, but also be transformed into one of the support and regulation units for power grid operation. Demand Response (DR) is the abbreviation of power demand response, which refers to when the wholesale power market price rises or the system reliability is threatened, after receiving a direct compensation notice or a power price increase signal for inducing load reduction sent by the power supply side, power users change their inherent habitual power consumption patterns to reduce or postpone the power load during a certain period to respond to the power supply, so as to ensure the stability of the power grid and suppress the rise of electricity prices. Demand response includes time-based demand response and incentive-based demand response. Time-based demand response is also called price-based demand response, which means that users adjust their power demand according to the received price signals, including time-of-use electricity price response, real-time electricity price response, peak electricity price response, etc.; in incentive-based demand response, there are generally two types of incentives for participating users: one is direct compensation independent of the existing electricity price policy; the other is a discount on the existing electricity price. Before the implementation of the demand response plan, usually the implementing agency of the demand response needs to sign a contract with the participating users in advance, and stipulate the content of the demand response in the contract, such as the size and accounting standard of the reduced power load, the response duration, the maximum number of responses during the contract period, etc., and also notify the time in advance, the compensation or electricity price discount standard, the penalty measures for breach of contract, etc.

[0004] The prior art discloses a price incentive method applied to a virtual power plant, including: constructing a fourth model reflecting the revenue of the virtual power plant through a first model reflecting the total incentive obtained by the virtual power plant from the power grid, a second model reflecting the total incentive of the virtual power plant for aggregated resources, and a third model reflecting the capital consumption of the virtual power plant, and calculating the internal pricing parameters of each aggregated resource with the maximization of the revenue of the virtual power plant as the goal through the fourth model and the constraint conditions of the fourth model. This solution considers the dispatching resources of indirect incentives, but when facing the demand response of a 5G base station group, the task allocation is unreasonable, and it is impossible to maximize the operator's revenue while minimizing the grid expenditure cost. Summary of the Invention

[0005] In order to overcome the defect of unreasonable task allocation of the prior art for the demand response of a 5G base station group, the present invention provides a demand response optimization method and system for the load of a 5G base station group, rationally allocating computing tasks, increasing the incentive amount obtained by the operator, reducing the power generation cost of the power grid, maximizing the operator's revenue while minimizing the grid expenditure cost, and improving energy efficiency.

[0006] To solve the above technical problems, the technical solution of the present invention is as follows:

[0007] The present invention provides a demand response optimization method for the load of a 5G base station group, including:

[0008] S1: Obtain the power grid node information, 5G base station group information, and task allocation information participating in the demand response within the region;

[0009] S2: Based on the preset initial incentive electricity price, 5G base station group information, and task allocation information, establish an operator revenue function with the maximization of the operator's revenue as the goal, and set the revenue function constraint conditions to calculate the current incentive amount;

[0010] S3: Based on the current incentive amount, power grid node information, and task allocation information, establish a power grid expenditure function with the minimization of the power grid expenditure as the goal, and set the expenditure function constraint conditions to calculate the current generator output;

[0011] S4: Based on the current incentive amount and the current generator output, establish a total operating cost function with the minimization of the total operating cost as the goal, and set the operating cost constraint conditions to calculate the current incentive electricity price;

[0012] S5: Update the initial incentive electricity price with the current incentive electricity price, repeat steps S2 - S4 until the total operating cost function meets the convergence condition, output the corresponding current incentive electricity price as the optimal incentive electricity price, and apply it to the power grid nodes and 5G base station groups participating in the demand response within the region.

[0013] The 5G base station group refers to a group of base stations directly controlled by a communication operator, equipped with a baseband processing unit and capable of undertaking communication computing task processing. It is characterized by a large number and is distributed in various nodes of the power grid in the form of low-voltage distribution network electrical loads. The 5G base stations transfer and transmit computing tasks through optical fiber communication and other means. The 5G base station groups connected to the same power grid node are regarded as a whole. When the computing tasks are transferred between 5G base station groups belonging to different power grid nodes, the electrical load corresponding to the task processing by the baseband processing unit is also regarded as having been transferred between the power grid nodes. The operator and the power grid manager determine the power grid node information, 5G base station group information, and task allocation information participating in demand response within the region, and negotiate the initial incentive electricity price. An operator revenue function is established based on the initial incentive electricity price, 5G base station group information, and task allocation information to calculate the current incentive amount. Combining the power grid node information and task allocation information, a power grid expenditure function is established to calculate the current generator output. Finally, based on the current incentive amount and the current generator output, a total operating cost function is established with the goal of minimizing the total operating cost to obtain the current incentive electricity price. The initial incentive electricity price is updated using the current incentive electricity price, and the above process is repeated to iteratively calculate the next incentive electricity price until the total operating cost function meets the convergence condition. At this time, the corresponding incentive electricity price enables the operator to obtain the maximum incentive amount after responding, achieving the maximum transfer of the electrical load of the power grid nodes; at the same time, it minimizes the power generation cost of the power grid nodes, minimizes the power grid expenditure cost, and improves energy efficiency.

[0014] Preferably, in the step S1, the power grid node information includes the number of power grid nodes, the number of generator sets, the maximum output of the generator sets, the minimum output of the generator sets, the number of transmission lines, the power flow of the transmission lines, the maximum power flow of the transmission lines, and the admittance of the transmission lines.

[0015] The 5G base station group information includes the radio unit electrical load, the static power consumption of the baseband processing unit, the dynamic power consumption of the baseband processing unit, the transfer power consumption of the baseband processing unit, the upper limit of the processing capacity of the baseband processing unit, the electrical load per unit task volume of optical fiber transmission, and the optical fiber transmission speed.

[0016] The initial task allocation information includes the task sequence, the initial task calculation scheme, the initial task transfer scheme, and the upper limit of the average upload delay of a single task.

[0017] Preferably, the task sequence, the initial task calculation scheme, and the initial task transfer scheme are all represented in matrix form. Among them, the task sequence matrix is:

[0018]

[0019] The task calculation matrix is:

[0020]

[0021] The task transfer matrix is as follows:

[0022]

[0023] In the formula, L represents the task sequence matrix, l j represents the j-th task, NTASK represents the total number of tasks; B represents the initial task calculation matrix, DB represents the task transfer matrix, and NB represents the number of power grid nodes; the elements in the task calculation matrix B and the task transfer matrix D are all 0, 1 variables, b m,j =1 indicates that the j-th task is initially calculated by the 5G base station group of the m-th power grid node, db m,j =1 indicates that the j-th task is transferred to the 5G base station group of the m-th power grid node for calculation; j ∈ [1, NTASK], m ∈ [1, NB].

[0024] Preferably, in the step S2, the specific method for establishing the operator revenue function according to the 5G base station group information and the task allocation information is as follows:

[0025] Set the initial incentive electricity price CP = [cp1 … cp m … cp NB T , where cp m represents the initial incentive electricity price of the m-th power grid node;

[0026] Calculate the electric load matrix LD of the 5G base station group of the power grid node operating with the task transfer matrix DB:

[0027]

[0028] In the formula, ld m represents the electric load of the 5G base station group of the m-th power grid node operating with the task transfer matrix DB:

[0029] ld m =paau m +pbbu m

[0030] pbbu m =pst m +pdy m +pmi m

[0031] In the formula, paau m represents the radio unit electric load of the 5G base station group of the m-th power grid node, pbbu m represents the baseband processing unit electric load of the 5G base station group of the m-th power grid node, pst m represents the static power consumption of the baseband processing unit of the 5G base station group of the m-th power grid node, pdy​m The dynamic power consumption and pmi of the baseband processing unit of the 5G base station group representing the mth power grid node m represents the transfer power consumption of the baseband processing unit of the 5G base station group of the mth power grid node;

[0032] pst m = η m ·uf m + θ m

[0033] pdy m = (pbu m - pst m )·uf m

[0034]

[0035]

[0036] In the formula, η m represents the static power consumption proportionality coefficient of the baseband processing unit of the 5G base station group of the mth power grid node, uf m represents the task occupancy rate of the baseband processing unit of the 5G base station group of the mth power grid node, θ m represents the minimum fixed power consumption of the static power consumption of the baseband processing unit of the 5G base station group of the mth power grid node, pbu m represents the maximum power consumption when the baseband processing unit of the 5G base station group of the mth power grid node operates at full power, β otp represents the electrical load per unit task of optical fiber transmission, cap m represents the upper limit of the processing capacity of the baseband processing unit of the 5G base station group of the mth power grid node;

[0037] Then the operator's revenue function is:

[0038] max Obj1(DB) = CP T ×(LD - LD ori )

[0039] In the formula, Obj1(DB) represents the current incentive amount when the 5G base station group of the power grid node operates with the task transfer matrix DB, CP T represents the transpose matrix of the initial incentive electricity price CP, LD ori represents the electrical load when the 5G base station group of the power grid node operates with the task calculation matrix, that is, the electrical load when there is no task transfer in the 5G base station group of the power grid node.

[0040] Preferably, in the step S2, the specific method for setting the operating cost constraint condition is:

[0041] To ensure the transmission quality and processing quality of tasks among 5G base station clusters at power grid nodes, income function constraint conditions are set, including:

[0042] The baseband processing unit equipped in the 5G base station cluster of each power grid node has a certain upper limit on the computing task processing capacity per unit time period, and the allocated computing task volume cannot exceed this upper limit. Then, the processing capacity constraint:

[0043]

[0044] When computing tasks are transmitted between 5G base station clusters through optical fibers, transmission delay will be generated. The increase in delay will affect the usage experience of communication users. Therefore, it is necessary to consider controlling the average delay of each transferred computing task within a threshold range. Then, the transfer delay constraint:

[0045]

[0046] In the formula, cap m represents the upper limit of the processing capacity of the baseband processing unit of the 5G base station cluster at the m-th power grid node, RA represents the optical fiber transmission speed, and τ j represents the upper limit of the average upload delay of a single task.

[0047] Preferably, in the step S3, the specific method for establishing the power grid expenditure function according to the current incentive amount, power grid node information, and task allocation information is as follows:

[0048] Calculate the maximum incentive amount Bill t and the maximum load response TLD according to the current incentive amount:

[0049] Bill t = CP T ×(LDF(arg max{Obj1(DB),[Cond1,CP]}) - LD ori )

[0050]

[0051] In the formula, Cond1 represents the first boundary parameter; LDF(*) represents the power consumption load calculation function;

[0052] Cond1 = {LD ori , η, θ, L, CAP, B, β otp , RA}

[0053] Among them, η represents the static power consumption proportional coefficient matrix of the baseband processing unit, and θ represents the minimum fixed power consumption matrix of the static power consumption of the baseband processing unit;

[0054] Record the output of the generator set as G = [g1 … gi … g NG T , where g i represents the output of the i-th generator set, and NG represents the number of generator sets;

[0055] Then the power grid expenditure function is:

[0056]

[0057] In the formula, Obj2(G) represents the power grid expenditure, a i , b i , c i represent the first, second, and third output curve cost parameters of the i-th generator set.

[0058] Preferably, in the step S3, the specific method for setting the expenditure function constraint conditions is:

[0059] To ensure the power generation safety and quality of the generator sets at the power grid nodes, set the expenditure function constraint conditions, including:

[0060] Power balance constraint:

[0061]

[0062] Among them,

[0063]

[0064] Maximum and minimum power generation constraints:

[0065] G min ≤ G ≤ G max

[0066] Line power flow constraint:

[0067] -PL max ≤ LF ≤ PL max

[0068] Among them

[0069] LF = SF × [KP × G - KD × (LD opt - LD ori )]

[0070] SF = XB × KL T × (KL × XB × KL T ) -1

[0071] In the formula, ld ori,m represents the electrical load of the 5G base station group at the m-th power grid node operating with the task calculation matrix, loss nl ​Denote the network loss of the \(n_l\)-th transmission line, \(U\) represents the power grid node voltage matrix, and \(R\) nl represents the impedance of the \(n_l\)-th transmission line, and \(NL\) represents the number of transmission lines; \(G\) max represents the maximum output of the generating unit, and \(G\) min represents the minimum output of the generating unit; \(LF\) represents the power flow of the transmission line, and \(PL\) max represents the maximum power flow of the transmission line, \(LF = [lf_1 \ldots lf\) nl \(\ldots lf\) NL \( T , and \(lf\) nl represents the power flow of the \(n_l\)-th transmission line; \(SF\) is the transfer factor matrix, \(XB\) is the line admittance matrix, \(KL\) is the incidence matrix of the line, \(KP\) is the incidence matrix of the generating unit, and \(KD\) is the incidence matrix of the load.

[0072] Preferably, the specific method of step S4 is:

[0073] The total operating cost is equal to the power grid expenditure minus the operator's income, so the total operating cost function is:

[0074] \(\min Obj3(CP)=Obj2(G)-Obj1(DB)\)

[0075] To ensure the enthusiasm of the operator and the power grid to participate in demand response, set the cost function constraint conditions, that is, the minimum incentive amount constraint:

[0076] \(CP\) T \(\times (LDF(\arg\max\{Obj1(DB),[Cond1,CP]\}) - LD\) ori )\geq Ben\) min

[0077] In the formula, \(Ben\) min represents the preset minimum incentive amount.

[0078] Preferably, in step S5, the specific method for obtaining the optimal incentive electricity price is:

[0079] Update the initial incentive electricity price using the current incentive electricity price. When the power grid expenditure function \(Obj2(G)\) obtains the minimum value and the operator's income function \(Obj1(DB)\) obtains the maximum value, the total cost operation function satisfies the convergence condition; substitute the optimal unit output \(G\) corresponding to \(\min Obj2(G)\) opt and the optimal task transfer matrix \(DB\) corresponding to \(\max Obj1(DB)\) opt into the total cost operation function, calculate the minimum total operating cost, and obtain the corresponding optimal incentive electricity price \(CP\) opt ; where:

[0080] \(G\) opt= arg min{Obj2(G), [Cond2, CP]}

[0081] DB opt = arg max{Obj1(DB), [Cond1, CP]}

[0082] Wherein, Cond2 represents the second boundary parameter.

[0083] Cond2 = {A, B′, C, DB opt , LD ori , PL max , SF, KP, KD, G max , G min}

[0084] Wherein, A, B, and C respectively represent the first, second, and third output curve cost parameter matrices of the generator set.

[0085] The present invention also provides a demand response optimization system for the load of a 5G base station group. Based on the above-mentioned demand response optimization method for the load of a 5G base station group, the system includes:

[0086] A data acquisition module, configured to acquire power grid node information, 5G base station group information, and task allocation information participating in the demand response within the region;

[0087] An operator revenue optimization module, configured to establish an operator revenue function with the maximization of operator revenue as the goal according to the preset initial incentive electricity price, 5G base station group information, and task allocation information, and set revenue function constraint conditions to calculate the current incentive amount;

[0088] A power grid expenditure optimization module, configured to establish a power grid expenditure function with the minimization of power grid expenditure as the goal according to the current incentive amount, power grid node information, and task allocation information, and set expenditure function constraint conditions to calculate the current generator set output;

[0089] A total operating cost optimization module, configured to establish a total operating cost function with the minimization of the total operating cost as the goal according to the current incentive amount and the current generator set output, and set operating cost constraint conditions to calculate the current incentive electricity price;

[0090] An optimal incentive electricity price deployment module, configured to update the initial incentive electricity price in the operator revenue optimization module by using the current incentive electricity price, and repeat the steps of the operator revenue optimization module - total operating cost optimization module until the total operating cost function meets the convergence condition, and output the corresponding current incentive electricity price as the optimal incentive electricity price, and apply it to the power grid nodes and 5G base station groups participating in the demand response within the region.

[0091] Compared with the prior art, the beneficial effects of the technical solution of the present invention are:

[0092] The present invention first determines the grid node information, 5G base station group information, and task allocation information participating in demand response within the region, and presets an initial incentive electricity price; then establishes an operator revenue function based on the initial incentive electricity price, 5G base station group information, and task allocation information, and calculates the current incentive amount; combines the current incentive amount, grid node information, and task allocation information to establish a grid expenditure function and calculates the current generator output; finally, based on the current incentive amount and the current generator output, establishes a total operating cost function with the goal of minimizing the total operating cost to obtain the current incentive electricity price; uses the current incentive electricity price to update the initial incentive electricity price, repeats the above process, and iteratively calculates the next incentive electricity price until the total operating cost function meets the convergence condition. At this time, the corresponding incentive electricity price is output and applied as the optimal incentive electricity price, the obtained incentive amount reaches the maximum, and the electric load transfer of the grid nodes is realized to the greatest extent; at the same time, the power generation cost of the grid nodes is minimized, the grid expenditure cost is minimized, and the energy efficiency is improved. Description of the Drawings

[0093] Figure 1 FIG. is a flowchart of a demand response optimization method for a 5G base station group load according to Embodiment 1;

[0094] Figure 2 FIG. is a schematic structural diagram of a demand response optimization system for a 5G base station group load according to Embodiment 3. Detailed Embodiments

[0095] The drawings are only for illustrative purposes and should not be construed as a limitation of this patent;

[0096] To better illustrate this embodiment, some components in the drawings are omitted, enlarged, or reduced, and do not represent the dimensions of the actual product;

[0097] For those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0098] The technical solutions of the present invention will be further described below with reference to the drawings and embodiments.

[0099] Embodiment 1

[0100] This embodiment provides a demand response optimization method for a 5G base station group load, as Figure 1 shown, including:

[0101] S1: Obtain the grid node information, 5G base station group information, and task allocation information participating in demand response within the region;

[0102] S2: Based on the preset initial incentive electricity price, 5G base station group information, and task allocation information, establish an operator revenue function with the goal of maximizing the operator's revenue, set the constraint conditions for the revenue function, and calculate the current incentive amount.

[0103] S3: According to the current incentive amount, power grid node information, and task allocation information, establish a power grid expenditure function with the goal of minimizing the power grid expenditure, set the constraint conditions for the expenditure function, and calculate the current generator output.

[0104] S4: Based on the current incentive amount and the current generator output, establish a total operating cost function with the goal of minimizing the total operating cost, set the constraint conditions for the operating cost, and calculate the current incentive electricity price.

[0105] S5: Use the current incentive electricity price to update the initial incentive electricity price in step S2, repeat steps S2 - S4 until the total operating cost function meets the convergence condition, output the corresponding current incentive electricity price as the optimal incentive electricity price, and apply it to the power grid nodes and 5G base station groups participating in demand response within the region.

[0106] In the specific implementation process, the operator and the power grid manager determine the power grid node information, 5G base station group information, and task allocation information of the region participating in demand response, and negotiate the initial incentive electricity price; establish an operator revenue function based on the initial incentive electricity price, 5G base station group information, and task allocation information, and calculate the current incentive amount; combine the power grid node information and task allocation information to establish a power grid expenditure function and calculate the current generator output; finally, based on the current incentive amount and the current generator output, establish a total operating cost function with the goal of minimizing the total operating cost to obtain the current incentive electricity price; use the current incentive electricity price to update the initial incentive electricity price, repeat the above process, iteratively calculate the next incentive electricity price until the total operating cost function meets the convergence condition. At this time, the corresponding incentive electricity price enables the operator to obtain the maximum incentive amount after response, achieving the maximum transfer of the electrical load of the power grid nodes; at the same time, the power generation cost of the power grid nodes is the lowest, minimizing the power grid expenditure cost and improving energy efficiency.

[0107] Embodiment 2

[0108] This embodiment provides a demand response optimization method for the load of a 5G base station group, including:

[0109] S1: Obtain the power grid node information, 5G base station group information, and task allocation information of the region participating in demand response.

[0110] The power grid node information includes the number of power grid nodes, the number of generators, the maximum output of the generators, the minimum output of the generators, the number of transmission lines, the power flow of the transmission lines, the maximum power flow of the transmission lines, and the admittance of the transmission lines.

[0111] The 5G base station group information includes the electrical load of the radio unit, the static power consumption of the baseband processing unit, the dynamic power consumption of the baseband processing unit, the transfer power consumption of the baseband processing unit, the upper limit of the processing capacity of the baseband processing unit, the electrical load per unit task volume of optical fiber transmission, and the optical fiber transmission speed;

[0112] The initial task allocation information includes the task sequence, the initial task calculation scheme, the initial task transfer scheme, and the upper limit of the average upload delay of a single task; the task sequence, the initial task calculation scheme, and the initial task transfer scheme are all represented in matrix form, where the task sequence matrix is:

[0113]

[0114] The task calculation matrix is:

[0115]

[0116] The task transfer matrix is:

[0117]

[0118] In the formula, L represents the task sequence matrix, l j represents the j-th task, NTASK represents the total number of tasks; B represents the initial task calculation matrix, DB represents the task transfer matrix, and NB represents the number of power grid nodes; the elements in the task calculation matrix B and the task transfer matrix D are all 0, 1 variables, b m,j =1 indicates that the j-th task is initially calculated by the 5G base station group of the m-th power grid node, db m,j =1 indicates that the j-th task is transferred to the 5G base station group of the m-th power grid node for calculation; j∈[1,NTASK], m∈[1,NB].

[0119] S2: According to the preset initial incentive electricity price, 5G base station group information, and task allocation information, establish an operator revenue function with the goal of maximizing operator revenue, set the revenue function constraint conditions, and calculate the current incentive amount;

[0120] Set the initial incentive electricity price CP = [cp1 … cp m … cp NB T , where cp m represents the initial incentive electricity price of the m-th power grid node;

[0121] Calculate the electrical load matrix LD of the 5G base station group of the power grid node operating with the task transfer matrix DB:

[0122]

[0123] In the formula, ld m ​The electrical load of the 5G base station group at the m-th power grid node operating with the task transfer matrix DB:

[0124] ld m = paau m + pbbu m

[0125] pbbu m = pst m + pdy m + pmi m

[0126] Wherein, paau m represents the radio unit electrical load of the 5G base station group at the m-th power grid node, pbbu m represents the baseband processing unit electrical load of the 5G base station group at the m-th power grid node, pst m represents the static power consumption of the baseband processing unit of the 5G base station group at the m-th power grid node, pdy m represents the dynamic power consumption of the baseband processing unit of the 5G base station group at the m-th power grid node, pmi m represents the transfer power consumption of the baseband processing unit of the 5G base station group at the m-th power grid node;

[0127] pst m = η m · uf m + θ m

[0128] pdy m = (pbu m - pst m )· uf m

[0129]

[0130] Wherein, η m represents the static power consumption proportionality coefficient of the baseband processing unit of the 5G base station group at the m-th power grid node, uf m represents the task occupancy rate of the baseband processing unit of the 5G base station group at the m-th power grid node, θ m represents the minimum fixed power consumption of the static power consumption of the baseband processing unit of the 5G base station group at the m-th power grid node, pbu m represents the maximum power consumption when the baseband processing unit of the 5G base station group at the m-th power grid node operates at full power, β otp represents the electrical load per unit task of optical fiber transmission, cap m represents the upper limit of the processing capacity of the baseband processing unit of the 5G base station group at the m-th power grid node;

[0131] Then the operator's revenue function is:

[0132] The maximum Obj1(DB) of the 5G base station group at the power grid node when operating with the task transfer matrix DB is CP T ×(LD - LD ori )

[0133] In the formula, Obj1(DB) represents the current incentive amount of the 5G base station group at the power grid node when operating with the task transfer matrix DB, and CP T represents the transposed matrix of the initial incentive electricity price CP, and LD ori represents the electrical load of the 5G base station group at the power grid node when operating with the task calculation matrix, that is, the electrical load of the 5G base station group at the power grid node without task transfer;

[0134] To ensure the transmission quality and processing quality of tasks among the 5G base station groups at the power grid node, set the constraint conditions of the revenue function, including:

[0135] The baseband processing unit equipped in the 5G base station group of each power grid node has a certain upper limit of computing task processing capacity per unit time period, and the allocated computing task volume cannot exceed this upper limit, so the processing capacity constraint:

[0136]

[0137] When the computing tasks are transmitted between the 5G base station groups through optical fibers, transmission delay will be generated. The increase in delay will affect the usage experience of communication users. Therefore, it is necessary to consider controlling the average delay of each transferred computing task within a threshold range, so the transfer delay constraint:

[0138]

[0139] In the formula, cap m represents the upper limit of the processing capacity of the baseband processing unit of the 5G base station group at the mth power grid node, RA represents the optical fiber transmission speed, and τ j represents the upper limit of the average upload delay of a single task.

[0140] S3: According to the current incentive amount, power grid node information, and task allocation information, establish a power grid expenditure function with the goal of minimizing the power grid expenditure, set the constraint conditions of the expenditure function, and calculate the current generator output;

[0141] Calculate the maximum incentive amount Bill t and the maximum load response TLD according to the current incentive amount:

[0142] Bill t = CP T ×(LDF(arg max{Obj1(DB), [Cond1, CP]}) - LD ori )

[0143]

[0144] In the formula, Cond1 represents the first boundary parameter; LDF(*) represents the electricity load calculation function;

[0145] Cond1 = {LD ori , η, θ, L, CAP, B, β otp , RA}

[0146] Among them, η represents the static power consumption proportionality coefficient matrix of the baseband processing unit, and θ represents the minimum fixed power consumption matrix of the static power consumption of the baseband processing unit;

[0147] Denote the output of the generator set as G = [g1 … g i … g NG T , where g i represents the output of the i-th generator set, and NG represents the number of generator sets;

[0148] Then the grid expenditure function is:

[0149]

[0150] In the formula, Obj2(G) represents the grid expenditure, a i , b i , c i represent the first, second, and third output curve cost parameters of the i-th generator set;

[0151] To ensure the power generation safety and quality of the generator sets at the grid nodes, set the constraint conditions of the expenditure function, including:

[0152] Power balance constraint:

[0153]

[0154] Among them,

[0155]

[0156] Maximum and minimum power generation constraints:

[0157] G min ≤ G ≤ G max

[0158] Line power flow constraint:

[0159] -PL max ≤ LF ≤ PL max

[0160] Among them

[0161] LF = SF × [KP × G - KD × (LD​opt -LD ori )]

[0162] SF = XB × KL T ×(KL × XB × KL T ) -1

[0163] where ld ori,m represents the electrical load of the 5G base station group at the m-th power grid node operating with the task calculation matrix, loss nl represents the network loss of the nl-th transmission line, U represents the power grid node voltage matrix, R nl represents the impedance of the nl-th transmission line, NL represents the number of transmission lines; G max represents the maximum output of the generator set, G min represents the minimum output of the generator set; LF represents the power flow of the transmission line, PL max represents the maximum power flow of the transmission line, LF = [lf1 … lf nl … lf NL ) T , lf nl represents the power flow of the nl-th transmission line; SF is the transfer factor matrix, reflecting the matrix of the node voltage phase angle in the DC power flow calculation process; XB is the line admittance matrix, reflecting the diagonal matrix of the transmission line admittance; KL is the incidence matrix of the line, with rows corresponding to power grid nodes and columns corresponding to transmission lines, reflecting the connection topology of the power grid lines; KP is the incidence matrix of the generator set, with rows corresponding to power grid nodes and columns corresponding to generator sets, reflecting the connection between the generator sets and the power grid nodes; KD is the incidence matrix of the load, with rows corresponding to power grid nodes and columns corresponding to loads, reflecting the connection between the loads and the power grid nodes. Specifically:

[0164]

[0165] S4: Based on the current incentive amount and the current output of the generator set, establish a total operating cost function with the goal of minimizing the total operating cost, set the operating cost constraint conditions, and calculate the current incentive electricity price;

[0166] The total operating cost is equal to the power grid expenditure minus the operator's income. Then the total operating cost function is:

[0167] min Obj3(CP) = Obj2(G) - Obj1(DB)

[0168] To ensure the enthusiasm of the operator and the power grid to participate in demand response, set the cost function constraint conditions, that is, the minimum incentive amount constraint:

[0169] CP T×(LDF(arg max{Obj1(DB),[Cond1, CP]}) - LD ori ) ≥ Ben min

[0170] In the formula, Ben min represents the preset minimum incentive amount. Ben min cannot be too small, otherwise it will lead to insufficient enthusiasm of communication operators to participate in demand response; nor can it be too large, which will lead to insufficient increased profit of the power grid company and it is difficult to attract the power grid company to participate. In this embodiment, Ben min is set to 0.01 - 0.02 times of the electricity fee payable by users.

[0171] S5: Update the initial incentive electricity price in step S2 with the current incentive electricity price, and repeat steps S2 - S4 until the total operating cost function meets the convergence condition. Output the corresponding current incentive electricity price as the optimal incentive electricity price and apply it to the power grid nodes and 5G base station groups participating in demand response in the region.

[0172] Update the initial incentive electricity price with the current incentive electricity price. When the power grid expenditure function Obj2(G) obtains the minimum value and the operator revenue function Obj1(DB) obtains the maximum value, the convergence condition is met in the total cost operation function; substitute the optimal unit output G opt corresponding to min Obj2(G) and the optimal task transfer matrix DB opt corresponding to maxObj1(DB) into the total cost operation function, calculate the minimum total operating cost, and obtain the corresponding optimal incentive electricity price CP opt ; where:

[0173] G opt = arg min{Obj2(G), [Cond2, CP]}

[0174] DB opt = arg max{Obj1(DB), [Cond1, CP]}

[0175] In the formula, Cond2 represents the second boundary parameter.

[0176] Cond2 = {A, B′, C, DB opt , LD ori , PL max , SF, KP, KD, G max , G min}

[0177] Among them, A, B, and C respectively represent the first, second, and third output curve cost parameter matrices of the generator set.

[0178] Embodiment 3

[0179] This embodiment provides a demand response optimization system for the load of 5G base station clusters. Based on the demand response optimization method for the load of 5G base station clusters described in Embodiment 1 or 2, as Figure 2 shown, the system includes:

[0180] A data acquisition module, which is used to acquire the power grid node information, 5G base station cluster information, and task allocation information of the region participating in the demand response, send the 5G base station cluster information and task allocation information to the operator revenue optimization module, and send the power grid node information and task allocation information to the power grid expenditure optimization module;

[0181] An operator revenue optimization module, which is used to establish an operator revenue function with the goal of maximizing the operator's revenue according to the preset initial incentive electricity price, 5G base station cluster information, and task allocation information, set the constraint conditions of the revenue function, calculate the current incentive amount, and send it to the power grid expenditure optimization module;

[0182] A power grid expenditure optimization module, which is used to establish a power grid expenditure function with the goal of minimizing the power grid expenditure according to the current incentive amount, power grid node information, and task allocation information, set the constraint conditions of the expenditure function, and calculate the current generator output;

[0183] A total operating cost optimization module, which is used to establish a total operating cost function with the goal of minimizing the total operating cost according to the current incentive amount and the current generator output, set the constraint conditions of the operating cost, and calculate the current incentive electricity price;

[0184] An optimal incentive electricity price deployment module, which uses the current incentive electricity price to update the initial incentive electricity price in the operator revenue optimization module, repeats the steps of the operator revenue optimization module - total operating cost optimization module until the total operating cost function meets the convergence condition, outputs the corresponding current incentive electricity price as the optimal incentive electricity price, and applies it to the power grid nodes and 5G base station clusters in the region participating in the demand response.

[0185] The same or similar reference numerals correspond to the same or similar components;

[0186] The terms describing the positional relationship in the drawings are only for illustrative purposes and should not be construed as a limitation of this patent;

[0187] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, rather than limitations on the implementation manners of the present invention. For those of ordinary skill in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to enumerate all the implementation manners here. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the claims of the present invention.

Claims

1. A demand response optimization method for the load of a 5G base station group, characterized in that Including: S1: Obtain the power grid node information, 5G base station group information, and task assignment information participating in demand response within the area; The task assignment information includes a task sequence, an initial task calculation scheme, an initial task transfer scheme, and an upper limit of the average upload delay for a single task; The task sequence, the initial task calculation scheme, and the initial task transfer scheme are all represented in matrix form, where the task sequence matrix is: The task calculation matrix is: The task transfer matrix is: where L represents the task sequence matrix, and l j represents the j-th task, NTASK represents the total number of tasks; B represents the initial task calculation matrix, DB represents the task transfer matrix, and NB represents the number of power grid nodes; The elements in the task calculation matrix B and the task transfer matrix DB are all 0, 1 variables. When b m,j = 1, it means that the j-th task is initially calculated by the 5G base station group of the m-th power grid node. When db m,j = 1, it means that the j-th task is transferred to the 5G base station group of the m-th power grid node for calculation; j ∈ [1, NTASK], m ∈ [1, NB]; S2: According to the preset initial incentive electricity price, 5G base station group information, and task assignment information, establish an operator revenue function with the goal of maximizing the operator's revenue, set the constraint conditions of the revenue function, and calculate the current incentive amount; The specific method for establishing the operator revenue function is: Set the initial incentive electricity price CP = [cp1…cp m …cp NB T , where cp m represents the initial incentive electricity price of the m-th power grid node;​ Calculate the electrical load matrix LD of the 5G base station group of the power grid node operating with the task transfer matrix DB: where ld m represents the electrical load of the 5G base station group at the m-th power grid node operating with the task transfer matrix DB: ld m = paau m + pst m + pdy m + pmi m where, paau m represents the radio unit electrical load of the 5G base station group at the mth power grid node, pst m represents the static power consumption of the baseband processing unit of the 5G base station group at the mth power grid node, pdy m represents the dynamic power consumption of the baseband processing unit of the 5G base station group at the mth power grid node, pmi m represents the transfer power consumption of the baseband processing unit of the 5G base station group at the mth power grid node; Then the operator revenue function is: max Obj1(DB) = CP T ×(LD - LD ori ) Where, Obj1(DB) represents the current incentive amount for the 5G base station group of the power grid node operating with the task transfer matrix DB, and CP T represents the transposed matrix of the initial incentive electricity price CP, and LD ori represents the electrical load of the 5G base station group of the power grid node operating with the task calculation matrix, that is, the electrical load when there is no task transfer in the 5G base station group of the power grid node; The specific method for setting the constraint conditions of the revenue function is: To ensure the transmission quality and processing quality of tasks among the 5G base station groups of the power grid nodes, set the constraint conditions of the revenue function, including: Processing capacity constraint: Transfer delay constraint: where cap m represents the upper limit of the baseband processing capacity of the 5G base station group at the m-th power grid node, RA represents the optical fiber transmission speed, and τ j represents the upper limit of the average upload delay of a single task; S3: According to the current incentive amount, power grid node information, and task assignment information, establish a power grid expenditure function with the goal of minimizing the power grid expenditure, set the constraint conditions of the expenditure function, and calculate the current generator output; S4: According to the current incentive amount and the current generator output, establish a total operating cost function with the goal of minimizing the total operating cost, set the constraint conditions of the operating cost, and calculate the current incentive electricity price; S5: Update the initial incentive electricity price with the current incentive electricity price, repeat steps S2 - S4 until the total operating cost function meets the convergence condition, output the corresponding current incentive electricity price as the optimal incentive electricity price, and apply it to the power grid nodes and 5G base station groups participating in demand response within the area.

2. The demand response optimization method for the load of a 5G base station group according to claim 1, wherein In step S1, the power grid node information includes the number of power grid nodes, the number of generators, the maximum output of the generators, the minimum output of the generators, the number of transmission lines, the power flow of the transmission lines, the maximum power flow of the transmission lines, and the admittance of the transmission lines; The 5G base station group information includes the electrical load of the radio unit, the static power consumption of the baseband processing unit, the dynamic power consumption of the baseband processing unit, the transfer power consumption of the baseband processing unit, the upper limit of the processing capacity of the baseband processing unit, the electrical load per unit task volume of optical fiber transmission, and the optical fiber transmission speed.

3. The demand response optimization method for the load of a 5G base station group according to claim 2, wherein In step S3, the specific method for establishing the power grid expenditure function according to the current incentive amount, power grid node information, and task assignment information is: Calculate the maximum incentive amount Bill based on the current incentive amount t and the maximum load response TLD: Bill t = CP T × (LDF(argmax{Obj1(DB), [Cond1, CP]}) - LD ori ) In the formula, Cond1 represents the first boundary parameter; LDF(*) represents the electrical load calculation function; Record the output of the generator set as \(G = [g_1 \ldots g i \ldots g NG \) T , where \(g i \) represents the output of the \(i\)-th generator set, and \(N_G\) represents the number of generator sets; Then the power grid expenditure function is: In the formula, Obj2(G) represents the grid expenditure, a i , b i , c i represent the first, second, and third output curve cost parameters of the i-th generator set.

4. The demand response optimization method for the load of a 5G base station group according to claim 3, wherein In step S3, the specific method for setting the constraint conditions of the expenditure function is: To ensure the power generation safety and quality of the power grid node generators, set the constraint conditions of the expenditure function, including: Power balance constraint: Maximum and minimum power generation constraints: G min ≤G≤G max Line power flow constraint: -PL max ≤LF≤PL max where ld ori,m represents the electrical load of the 5G base station group at the m-th power grid node operating with a task calculation matrix, and loss nl represents the network loss of the nl-th transmission line, and NL represents the number of transmission lines; G max represents the maximum output of the generator set, and G min represents the minimum output of the generator set; LF represents the power flow of the transmission line, and PL max represents the maximum power flow of the transmission line, LF = [lf1…lf nl …lf NL T , and lf nl represents the power flow of the nl-th transmission line.​ 5. The demand response optimization method for the load of a 5G base station group according to claim 4, wherein The specific method of step S4 is: The total operating cost is equal to the power grid expenditure minus the operator's revenue, so the total operating cost function is: min Obj3(CP) = Obj2(G) - Obj1(DB) Set the cost function constraint, i.e., the minimum incentive amount constraint: CP T ×(LDF(argmax{Obj1(DB),[Cond1,CP]}) - LD ori ) ≥ Ben min where Ben min represents a preset minimum incentive amount.

6. The demand response optimization method for the load of 5G base station clusters according to claim 5, characterized in that, In the step S5, the specific method for obtaining the optimal incentive electricity price is as follows: Update the initial incentive electricity price using the current incentive electricity price. When the grid expenditure function Obj2(G) reaches the minimum value and the operator's revenue function Obj1(DB) reaches the maximum value, the total cost operation function satisfies the convergence condition; substitute the optimal unit output G corresponding to min Obj2(G) opt and the optimal task transfer matrix DB corresponding to max Obj1(DB) opt into the total cost operation function to calculate the minimum total operation cost and obtain the corresponding optimal incentive electricity price CP opt ; Where: G opt = argmin{Obj2(G), [Cond2, CP]} DB opt = argmax{Obj1(DB), [Cond1, CP]} In the formula, Cond2 represents the second boundary parameter.

7. A demand response optimization system for the load of a 5G base station group, characterized in that, Based on the demand response optimization method for the 5G base station group load according to any one of claims 1-6, the system includes: A data acquisition module for acquiring power grid node information, 5G base station group information, and task allocation information participating in the demand response within the region; An operator revenue optimization module for establishing an operator revenue function with the goal of maximizing the operator's revenue according to the preset initial incentive electricity price, 5G base station group information, and task allocation information, setting revenue function constraints, and calculating the current incentive amount; A power grid expenditure optimization module for establishing a power grid expenditure function with the goal of minimizing the power grid expenditure according to the current incentive amount, power grid node information, and task allocation information, setting expenditure function constraints, and calculating the current generator output; A total operating cost optimization module for establishing a total operating cost function with the goal of minimizing the total operating cost according to the current incentive amount and the current generator output, setting operating cost constraints, and calculating the current incentive electricity price; An optimal incentive electricity price deployment module updates the initial incentive electricity price in the operator revenue optimization module using the current incentive electricity price, repeats the steps of the operator revenue optimization module - total operating cost optimization module until the total operating cost function meets the convergence condition, outputs the corresponding current incentive electricity price as the optimal incentive electricity price, and applies it to the power grid nodes and 5G base station groups participating in the demand response within the region.

Images