Power grid transmission section power control method and system

Abstract

The invention provides a power grid transmission section power control method and system, and relates to the technical field of power system operation optimization. Aiming at the problems that the traditional power transmission section power control only depends on a thermal power generating unit, the regulation and control flexibility is insufficient, and the load space-time transfer characteristic of a data center is not fully utilized, the method realizes the coordinated regulation and control through three steps of sample generation, AI modeling and constraint optimization. Firstly, a Monte Carlo method is utilized to generate a power grid operation mode sample, and a data center and thermal power generating unit cooperative control model with power transmission section power as input and a regulation and control target as output is constructed; based on a power grid regulation and control demand, calculating a preliminary regulation and control target through the model; and finally, determining a final scheme through a load space-time transfer optimization model in combination with a data center service quality constraint. The method can improve power transmission section power control precision and response speed, gives consideration to power grid safety and data center benefits, and is suitable for real-time operation optimization of a large-scale power system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system operation optimization, and in particular to a power grid transmission section power control method and system. BACKGROUND

[0002] The development and construction of data centers are driven by the progress of computing technology. Data centers in power systems have good regulation and control capabilities, can achieve comprehensive regulation and control in time and space, and have a great effect on improving the flexibility of power system operation. In addition, with the continuous growth of load demand, the power system operation often encounters the problem of limited transmission capacity of the section power due to the power over-limit of some lines, which seriously restricts the transmission capacity of the power grid transmission section and affects the economy and safety of power operation. Therefore, in order to fully utilize the flexible time-space transfer characteristics of data centers and the regulation and control capabilities of conventional thermal power units, and to realize precise control of the transmission section power, a collaborative control method of data centers and thermal power units is urgently needed. The transmission capacity of the power grid transmission section refers to the maximum power that the transmission system allows from one local system (or power plant) to another local system (or substation) in the power system. The transmission power of the transmission section is an important parameter for one region to supply power to another region, and monitoring the power flow of the transmission section is to ensure that the transmission power of the transmission section does not exceed the static stability limit, and is also an important basis for dispatching as unit regulation output.

[0003] The traditional transmission section power control method mainly focuses on the dispatching of thermal power units, ignoring the time-space transfer characteristics of data center loads. In recent years, researchers have begun to try to combine demand side management of data centers with supply side optimization of thermal power units, and through load forecasting, dynamic scheduling and combined heat and power supply technologies, to realize the collaborative control of the two. Current research mainly focuses on load forecasting models based on machine learning, real-time scheduling algorithms and the construction of collaborative optimization frameworks. However, how to realize flexible control of transmission section power through collaboration between data centers and thermal power units is still rarely reported, and further in-depth research is needed.

[0004] That is, although existing research attempts to combine demand side management of data centers with dispatching of thermal power units, it mainly focuses on load forecasting or static optimization, does not construct a dynamic collaborative model for transmission section power control, and does not fully consider the quality of service constraints of data centers (such as data transmission delay), resulting in insufficient practicability of the scheme. Therefore, a collaborative control method that takes into account the regulation and control requirements of the power grid and the benefits of data centers is urgently needed. SUMMARY

[0005] The present application aims to overcome at least one technical problem in the prior art and provides a power grid transmission section power control method and system.

[0006] In a first aspect, an embodiment of the present application provides a power grid transmission section power control method, the method comprising: step S1, obtaining a data center load fluctuation interval in a power system; step S2, based on a Monte Carlo method, randomly sampling in the data center load fluctuation interval to generate a plurality of data center load scenarios; step S3, for each data center load scenario, performing optimal power flow calculation to obtain corresponding transmission section power value and thermal power unit output value; step S4, taking the transmission section power value as an input feature, taking the corresponding data center load value and thermal power unit output value as an output target, and combining to form a training sample; step S5, repeating steps S1 to S4 until a predetermined number of training sample sets are generated; step S6, training a preset neural network base model using the training sample set to obtain a data center and thermal power unit collaborative control model; step S7, based on the power regulation and control requirements of the power system transmission section, using the data center and thermal power unit collaborative control model to calculate the theoretical operation regulation and control target of each data center and thermal power unit; step S8, receiving the theoretical operation regulation and control target of each data center output by the data center and thermal power unit collaborative control model, and based on a data center load space-time transfer optimization model, correcting the data center theoretical operation regulation and control target; wherein the data center load space-time transfer optimization model decomposes the total data center load into invariable load, transferable load and adjustable load; step S9, solving the data center load space-time transfer optimization model to obtain a final executable regulation and control scheme containing the transferable load and adjustable load power of each data center; step S10, outputting the final executable regulation and control scheme as an executable data center load regulation and control instruction, and outputting the theoretical operation regulation and control target of the thermal power unit as an executable thermal power unit output regulation and control instruction.

[0007] Further, the random sampling in the Monte Carlo method in step S2 adopts uniform distribution, and the load of each data center is randomly generated with equal probability in the fluctuation interval formed by the minimum load value and the maximum load value. The step S3 comprises: calculating by using the optimal power flow method to obtain the power of the data section of the power grid and the output of the thermal power unit. respectively represent the load value of each data center, the power of each line of the transmission section, and the output of each thermal power unit; n, m and s respectively represent the number of data centers, the number of transmission section lines, and the number of thermal power units; the step S4 comprises: combining the data section power, the data center load, and the thermal power unit output to generate a training sample, represented by

[0008] ​​​​​Furthermore, the optimal power flow calculation includes: minimizing the total power generation cost of the system as the objective function, and the constraints include: system active power balance constraints; upper and lower limits of output of each thermal power unit constraints; constraints that the power of each transmission line does not exceed its transmission capacity limit; and constraints that the voltage of each node is within the safe operating range.

[0009] Furthermore, the system active power balance constraint includes the total power generated by all thermal power units being equal to the total power consumed by all loads plus the transmission loss of the power grid; the upper and lower limit constraints of the output of each thermal power unit include the output of each thermal power unit being within the range of its maximum and minimum technical output values; the constraint that the power of each transmission line does not exceed its transmission capacity limit includes the power flowing through any line not exceeding its thermal stability limit; and the constraint that the voltage of each node is within the safe operating range includes the voltage of all nodes being maintained between preset safe upper and lower limits.

[0010] Furthermore, step S6 includes: taking the power of each line of the power grid transmission section as input features, and taking the load control target of each data center and the output control target of each thermal power unit as outputs; using an artificial neural network as the basic model, setting the number of hidden layer nodes to N1-N2, and using the sigmoid function as the activation function; dividing the generated training samples into a training set and a validation set according to a preset ratio, using the training set to train the artificial neural network model, using the validation set to verify the accuracy of the artificial neural network model, and completing the model construction when the average error between the output of the artificial neural network model and the actual value of the sample reaches a preset condition.

[0011] Furthermore, the theoretical operation and control target of the data center is the theoretical load that needs to be adjusted for each data center; the theoretical operation and control target of the thermal power unit is the theoretical output value that needs to be adjusted for each thermal power unit.

[0012] Furthermore, the mathematical expression of the data center load spatiotemporal transfer optimization model in step S8 is as follows: ; The constraints are: ; ; ; In the formula, , and These represent the fixed load power, transferable load power, and adjustable load power of data center i, respectively. The theoretical operational control target for data center i; and These are the upper and lower limits of the adjustable load power for data center i, respectively. is the data transmission delay between data centers i and j; is the upper limit of the data transmission delay between data centers affecting the quality of service of the data center.

[0013] Further, the step S9 comprises: solving the data center load space-time transfer optimization model by using a particle swarm optimization algorithm, and the specific steps comprise: step S901, initializing the particle swarm: setting the number of particles to a-b, each particle corresponding to a set of and ; step S902, calculating fitness: taking the deviation value of the objective function as the fitness function, the smaller the deviation, the higher the fitness; step S903, iterative updating: updating the speed and position of the particles, and retaining the particle with the optimal fitness; step S904, terminating iteration: when the number of iterations reaches a preset number or the optimal fitness does not change for N consecutive iterations, stopping iteration, and the and corresponding to the optimal particle is the solving result, and in combination with , the final executable control scheme of the data center i is obtained.

[0014] Further, the final executable control scheme comprises: obtaining the optimized total load power of each data center as the final executable control scheme based on the sum of the transferable load power, the adjustable load power and the invariable load by solving the data center load space-time transfer optimization model.

[0015] In a second aspect, the embodiment of the present application provides a power grid transmission section power control system, which is implemented by using the power grid transmission section power control method described above, and comprises: a training sample set generation module, which is adapted to obtain data center load fluctuation intervals in a power system; a Monte Carlo method-based random sampling module, which is adapted to generate a plurality of data center load scenarios by performing random sampling within the data center load fluctuation intervals; an optimal power flow calculation module, which is adapted to obtain corresponding transmission section power values and thermal power unit output values by performing optimal power flow calculation on each data center load scenario; a training sample combination module, which is adapted to combine the transmission section power values as input features and the corresponding data center load values and thermal power unit output values as output targets to form training samples; a training sample set generation module, which is adapted to generate a predetermined number of training sample sets; a data center and thermal power unit collaborative control model construction module, which is adapted to train a preset neural network base model by using the training sample sets to obtain a data center and thermal power unit collaborative control model; a data center and thermal power unit theoretical operation regulation and control target calculation module, which is adapted to calculate the theoretical operation regulation and control targets of each data center and thermal power unit by using the data center and thermal power unit collaborative control model based on power system transmission section power regulation and control requirements; a data center theoretical operation regulation and control target optimization module, which is adapted to receive the data center theoretical operation regulation and control targets output by the data center and thermal power unit collaborative control model, and correct the data center theoretical operation regulation and control targets based on a data center load space-time transfer optimization model; wherein the data center load space-time transfer optimization model decomposes total data center load into invariable load, transferable load and adjustable load; a space-time transfer optimization model solution module, which is adapted to solve the data center load space-time transfer optimization model to obtain a final executable regulation and control scheme containing the transferable load and adjustable load powers of each data center; and a regulation and control instruction output module, which is adapted to output the final executable regulation and control scheme as executable data center load regulation and control instructions and output the theoretical operation regulation and control targets of the thermal power unit as executable thermal power unit output regulation and control instructions.

[0016] In a third aspect, the embodiment of the present application further provides an electronic device, which comprises a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and the computer program is executed by the processor to implement the power grid transmission section power control method described above.

[0017] In a fourth aspect, the embodiment of the present application further provides a readable storage medium, and when instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the power grid transmission section power control method described above.

[0018] The beneficial effects of the present application compared with the prior art are: compared with the prior art, the beneficial effects of the present application are that the artificial intelligence training sample generation method for coordinated control of the data center and the thermal power generating unit is more time-saving than the conventional sample generation method based on the iterative flow model. In addition, based on the data center and thermal power generating unit regulation and control target given by the artificial intelligence-based data center and thermal power generating unit coordinated control model, the implementability of the data center load regulation and control target is considered, and the data center load space-time transfer optimization model considering the data center service quality constraint is calculated to finally determine the regulation and control scheme of the data center load.

[0019] That is, the beneficial effects of the present application are embodied in: 1. Higher regulation and control efficiency: the neural network model replaces the traditional flow iteration calculation, and the regulation and control target response time is reduced from minutes to milliseconds, which adapts to the real-time regulation and control demand of the power grid.

[0020] 2. Greater regulation and control potential: dynamic coordination of the data center and the thermal power generating unit is realized, and the space-time transfer characteristics of the data center load are fully utilized, and compared with the traditional method, the cross-section power control accuracy is improved by more than 30%.

[0021] 3. Multiple main body interests are considered: through the data center service quality constraint, the influence of regulation and control on the data center business is avoided, and the three-way balance of power grid safety, thermal power generating unit efficient operation and data center benefit is realized. BRIEF DESCRIPTION OF DRAWINGS

[0022] The present application will be further described below in conjunction with the drawings and embodiments.

[0023] Figure 1 Fig. 1 is a power grid power transmission cross-section power control method flowchart provided by an embodiment 1 of the present application.

[0024] Figure 2 Fig. 2 is an IEEE39 node system schematic diagram adopted by the embodiment 1 of the present application, wherein nodes 3, 8, 11, 14, 23, 27, 31 and 35 are marked as data centers, and lines 14-15 and 16-17 are marked as power transmission cross-sections to be controlled.

[0025] Figure 3 Fig. 3 is a power grid power transmission cross-section power control system structure schematic diagram provided by an embodiment 2 of the present application.

[0026] Figure 4 Fig. 4 is a partial block diagram of an electronic device provided by an embodiment 3 of the present application. DETAILED DESCRIPTION

[0027] Before any examples are described in further detail, it should be noted that some examples are described as processes depicted as flow diagrams. Although each can describe the operations as a sequential process, many of the operations can be performed in parallel, concurrently or simultaneously, and / or in a different order than that described. In addition, this process can be terminated when its operations are completed but can also have additional steps not included in the figure. A process can correspond to a method, a function, a procedure, a subroutine, a subprogram, etc. When introduced herein, "example" or "exemplary" means serving as an example, instance, or illustration. Any implementation described herein as an "example" or "exemplary" is not necessarily to be construed as preferred or advantageous over other examples.

[0028] It should be understood that, although terms such as "first," "second," and the like can be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element without departing from the scope of the examples. The term "and / or" as used herein encompasses any and all combinations of one or more of the associated associated items.

[0029] The application will now be described in detail with reference to the drawings. The diagram is a simplified schematic diagram, which schematically shows the basic structure of the application, and thus only shows the components relating to the application.

[0030] Example 1

[0031] The specific implementation is as follows: As Figure 1 shown, a power control method for power transmission section of power grid provided by the application is shown in the flow chart.

[0032] As an example, the method comprises: step S1, obtaining a data center load fluctuation interval in a power system; step S2, based on a Monte Carlo method, randomly sampling in the data center load fluctuation interval to generate a plurality of data center load scenarios; step S3, for each data center load scenario, performing optimal power flow calculation to obtain corresponding power transmission section power value and thermal power unit output value; step S4, taking the power transmission section power value as an input feature, and taking the corresponding data center load value and thermal power unit output value as an output target to form a training sample; step S5, repeating steps S1 to S4 until a predetermined number of training sample sets are generated; step S6, training a preset neural network base model using the training sample set to obtain a data center and thermal power unit collaborative control model; step S7, based on the power regulation demand of the power transmission section of the power system, using the data center and thermal power unit collaborative control model to calculate the theoretical operation regulation target of each data center and thermal power unit; step S8, receiving the theoretical operation regulation target of each data center output by the data center and thermal power unit collaborative control model, and correcting the data center theoretical operation regulation target based on a data center load space-time transfer optimization model; wherein the data center load space-time transfer optimization model decomposes the total data center load into invariable load, transferable load and adjustable load; step S9, solving the data center load space-time transfer optimization model to obtain a final executable regulation scheme containing the transferable load and adjustable load power of each data center; step S10, outputting the final executable regulation scheme as an executable data center load regulation instruction, and outputting the theoretical operation regulation target of the thermal power unit as an executable thermal power unit output regulation instruction.

[0033] In some possible embodiments, the random sampling in the Monte Carlo method in step S2 adopts a uniform distribution, and the load of each data center is randomly generated with equal probability in the fluctuation interval formed by the minimum load value and the maximum load value of the respective data center; the power system sample randomly extracted based on the Monte Carlo method ; the step S3 comprises: calculating by using the optimal power flow method to obtain the power of the data section and the output of the thermal power unit; wherein , and respectively represent the load value of each data center, the power of each line of the power transmission section, and the output of each thermal power unit; n, m and s respectively represent the number of data centers, the number of power transmission section lines, and the number of thermal power units; the step S4 comprises: combining the data section power, the data center load, and the thermal power unit output to generate a training sample, denoted as .

[0034] Preferably, the optimal power flow calculation comprises: taking the total generation cost minimization of the system as the objective function, and the constraint conditions comprising: system active power balance constraint; each thermal power unit output upper and lower limit constraint; each transmission line power not exceeding its transmission capacity limit constraint; each node voltage being within the safe operation range constraint. The system active power balance constraint comprises the total power emitted by all thermal power units being equal to the total power consumed by all loads plus the transmission loss of the power grid; the each thermal power unit output upper and lower limit constraint comprises the output of each thermal power unit being within the range composed of its maximum and minimum technical output values; the each transmission line power not exceeding its transmission capacity limit constraint comprises the power flowing through any line being unable to exceed its thermal stability limit; and the each node voltage being within the safe operation range constraint comprises the voltage of all nodes being maintained between the preset safe upper and lower limits.

[0035] Specifically, the operating states of the power grid are infinite, and we cannot enumerate them. Therefore, by random sampling, the Monte Carlo method can cover typical, boundary, and even extreme operating states of the power grid with a very high probability, thereby enabling the neural network model trained subsequently to have strong generalization ability. Specifically, for each data center, random values are taken within its possible load range (such as 0-20 MW) to generate thousands of vectors. Each vector represents a possible power consumption scenario. The optimal power flow is the "truth model" of the power system. Based on physical laws (Kirchhoff's law), it calculates the optimal generation distribution and the resulting cross-section power under the given load with the lowest total generation cost or the minimum network loss as the target. Key role: the optimal power flow calculation ensures that each training sample , ( , ) is not only physically feasible but also economically optimal. This means that the neural network model learns not just any control method, but the optimal or near-optimal control strategy. Among them, the essence of the sample: each sample is "the correspondence between the cross-section power and the source and load when the power grid is optimally operated under a certain power consumption condition". What the neural network model learns is this "optimal relationship".

[0036] More specifically, model training comprises: letting the neural network model learn and summarize the rules, and the deep meaning of input / output design: the input is : this represents the operating state and control demand of the power grid. The dispatchers see the cross-section power, so this design is very close to the actual application scenario. The output is , : this represents the action plan that needs to be taken on the source and load sides to maintain or achieve the operating state.

[0037] More specifically, the selection of neural network model includes: artificial neural network: especially refers to deep neural network. It has multiple hidden layers like human brain, which can extract features layer by layer and learn extremely complex nonlinear relationships. It can fit complex mapping from cross-section power to numerous regulation targets. Extreme learning machine: a fast learning algorithm for single hidden layer neural network. Its training speed is extremely fast, although the model capacity may not be as good as deep network, but for some specific problems, it can reach acceptable accuracy in a shorter time, suitable for scenarios with extremely high real-time requirements.

[0038] More specifically, the final product of training includes: a trained neural network model, which becomes a "seasoned super dispatcher". It remembers the optimal regulation mode in countless situations and can give an answer in a moment.

[0039] In some feasible embodiments, the step S6 includes: taking the power of each line of the power grid transmission section as the input feature, and taking the load regulation target of each data center and the output regulation target of each thermal power unit as the output; adopting artificial neural network as the basic model, setting the number of hidden layer nodes to N1-N2 (such as 100), and adopting sigmoid function as the activation function; dividing the generated training samples into training set and validation set according to the preset proportion, training the artificial neural network model with the training set, verifying the accuracy of the artificial neural network model with the validation set, and completing the model construction when the average error between the output of the artificial neural network model and the actual value of the sample reaches the preset condition.

[0040] In some feasible embodiments, the theoretical operation regulation target of the data center is the theoretical load amount that each data center needs to adjust; and the theoretical operation regulation target of the thermal power unit is the theoretical output value that each thermal power unit needs to adjust.

[0041] Specifically, as shown in Figure 2 The specific operation mode of step S7 includes: application scenario: the dispatching center monitors that the power of lines 14-15 and 16-17 needs to be controlled to [50MW, 224MW]. Process: input the target value into the neural network model. Output: the model immediately outputs the preliminary theoretical operation regulation target of the data center and the thermal power unit. It should be noted that the preliminary theoretical operation regulation target of the data center and the thermal power unit obtained at this time is an ideal value. It only considers the optimization of the power grid level and has not considered the business constraints of the data center itself. Therefore, it is only a "target", not an "executable instruction".

[0042] That is, according to the actual demand of the power grid (such as adjusting the power of a certain transmission section line from 35 MW to 50 MW), the target power is input into the constructed data center and thermal power unit cooperative control model, and the model will output the theoretical load regulation target of each data center and the theoretical output regulation target of each thermal power unit within milliseconds. For example, the model may output "data center 1 load regulation to 322.0 MW, thermal power unit 1 output regulation to 161.76 MW", etc., to provide preliminary basis for subsequent regulation.

[0043] In some possible embodiments, the mathematical expression of the data center load space-time transfer optimization model in step S8 is as follows: ; The constraint conditions are: ; ; ; In the formula, 、 and are the invariable load power, transferable load power and adjustable load power of the data center i; is the theoretical operation regulation target of the data center i; and are the upper and lower limits of the adjustable load power of the data center i; is the data transmission delay between the data centers i and j; is the upper limit of the data transmission delay between data centers affecting the service quality of the data center. The data center load space-time transfer optimization model considering the service quality constraint has variables to be solved, which are the transferable load power and the adjustable load power of each data center. The intelligent optimization algorithm can be used for calculation and solution, so as to determine the final regulation scheme of the data center load.

[0044] Specifically, the constraint conditions include: (1) adjustable load upper and lower limit constraint: ; The physical meaning is that the adjustable load power cannot exceed its technical capability range, and the adjustable load upper limit is preferably 10% to 30% of the total capacity of the data center, and the adjustable load lower limit is preferably 0 MW. It should be noted that the specific numerical value of the adjustable load upper and lower limit is not limited here, and the specific numerical value can be adjusted by relevant technical personnel according to the equipment characteristics and operating state of the data center; (2) service quality constraint-data transmission delay constraint: ; Data transmission delay between data centers i and j; Upper bound of data transmission delay between data centers that affects the quality of service of data centers; The physical meaning is that if the network delay between two data centers is too high (for example, more than 50 ms required by online games), task transfer cannot be performed between them. This constraint tightly couples the quality of the information network with the regulation of the power network.

[0045] (3) Transferrable load conservation constraint: ; The physical meaning is that the total amount of transferrable load is conserved in the entire system. The reduction of transferrable load in one data center must be compensated by another data center (spatial transfer) or be delayed in time (temporal transfer). This prevents the model from simply satisfying the target by "disappearing" tasks.

[0046] Specifically, non-variable load: definition: the power consumed to maintain the basic operation of the data center and rigid tasks that must be processed in real time. Examples: power consumption of basic cooling systems, operation of core databases, real-time online interactive services (such as video conferencing, financial transactions). Characteristics: non-adjustable, a "bottom line" that must be met. Transferrable load: definition: the power consumed by computing tasks that can be migrated in time and / or space. Temporal transfer: delaying the execution of tasks for a few hours (such as non-urgent data backup, large-scale scientific computing). Spatial transfer: transferring computing tasks through the network to another data center with surplus processing capacity for execution (such as sending rendering tasks from a Beijing data center to a Guiyang data center). Characteristics: this is the core of the flexibility of "time-space transfer". Adjustable load: definition: power that can be continuously adjusted within a certain range through technical means. Examples: IT equipment: dynamic voltage frequency adjustment, power saving by reducing CPU frequency. Cooling system: appropriately increasing the supply and return water temperature setting value under the premise of ensuring the safety of equipment. Characteristics: it can be smoothly and finely adjusted like a "knob". Among them, non-variable load usually accounts for 40%-60% of the total load of a data center. For example, a data center with a total load of 20 MW consists of: non-variable load: 10 MW, of which 3 MW is for infrastructure protection (cooling, lighting, network), and 7 MW is for core computing services (real-time transactions, core databases, video conferencing); transferrable load: 6 MW (scientific computing, data backup, video rendering, etc.); adjustable load: 4 MW (power saved by reducing frequency or turning off non-core servers, optimizing cooling parameters).

[0047] In some possible implementations, the step S9 comprises: solving the data center load space-time transfer optimization model by using a particle swarm optimization algorithm, and the specific steps comprise: step S901, initializing the particle swarm: setting the number of particles to a-b (such as 50-100), and each particle corresponds to a set of In combination with Step S902, calculating the fitness: taking the deviation value of the objective function as the fitness function, and the smaller the deviation, the higher the fitness; step S903, iteratively updating: updating the speed and position of the particles, and retaining the particle with the optimal fitness; step S904, terminating iteration: when the number of iterations reaches a preset number or the optimal fitness does not change for N consecutive iterations, stopping iteration, and the In combination with corresponding to the optimal particle is the solving result, and in combination with the final executable regulation scheme of the data center i is obtained.

[0048] Preferably, the final executable regulation scheme comprises: obtaining the total load power of each data center after optimization as the final executable regulation scheme based on the sum of the transferable load power, the adjustable load power and the invariable load power by solving the data center load space-time transfer optimization model. That is, the actual regulation value = invariable load power + transferable load power + adjustable load power.

[0049] Through the above steps S7-S9, the landing of "source-load interaction" is realized: it makes the regulation of the data center load no longer a simple "pulling the switch to limit power", but a fine and business-aware "resource scheduling". The core contradiction is solved: the contradiction between the rigid demand of power grid safety regulation and the rigid requirement of data center business quality is successfully balanced. The information-physical fusion is embodied: taking "data transmission delay", a core indicator in the information field, as a key constraint of power grid optimization control, which is a model of cross-field technology integration. That is, without this optimization process, the whole scheme is only a neural network model staying at the theoretical level; with this optimization process, it becomes a complete and robust technical solution that can be applied to the real world.

[0050] In order to facilitate the understanding of the above embodiments, specific examples are described here: in combination with Figure 2As shown, based on IEEE39 node system, it is assumed that data centers are configured at nodes 3, 8, 11, 14, 23, 27, 31 and 35, and each data center is 20MW, wherein the proportion of non-variable load, transferable load and adjustable load is 50%, 30% and 20% respectively, the task transfer delay between each data center is 10ms, and the upper limit of data center service quality requirement delay is 20ms. Lines 14-15 and 16-17 are selected as the power control transmission sections. Under the initial working condition, the power flow of the transmission section lines 14-15 and 16-17 is 35MW and 240MW respectively.

[0051] Firstly, according to the load power of each data center, it is assumed that the load power of each data center is evenly distributed in [0, 20], and the Monte Carlo method is used to generate samples, and the line power of the transmission section and the output data of the thermal power generator set at this time are recorded. Set the sample number to 10000 groups, form the power grid operation mode sample set with the transmission section line power as the input, and the thermal power generator set output data and data center load data as the output; Secondly, based on the power grid operation mode sample set, a single-layer artificial neural network model is trained, the number of hidden layer nodes is set to 100, the input features are the power of the two transmission section lines, and the output results are the regulation and control targets of 8 data centers and 10 thermal power generators. Generate an artificial neural network model of the data center and thermal power generator cooperative control model; Then, based on the generated data center and thermal power generator cooperative control artificial intelligence model, the power control targets of the two transmission section lines 14-15 and 16-17 are 50MW and 224MW, and the output of 8 data centers and 10 thermal power generators is shown in Table 1.

[0052] Table 1: Data center and thermal power generator regulation and control target table (unit: MW):

[0053] Finally, a data center load space-time transfer optimization model considering the service quality constraint of the data center is constructed, and a particle swarm optimization algorithm is used for calculation to obtain the actual regulation and control scheme of the data center load as shown in Table 2.

[0054] Table 2: Data center load actual regulation and control scheme (unit: MW):

[0055] As can be seen from Table 1 and Table 2, due to the influence of data center non-variable load, the load power of some data centers cannot be reduced to 0, and due to the influence of data center service quality requirement, there is a difference between some regulation and control values and actual values, which verifies the effect of the data center load space-time transfer optimization model.

[0056] In addition, since it is assumed that the thermal power unit can fully respond to the power regulation demand, the control target of the thermal power unit is the actual regulation scheme of the thermal power unit. Further based on the actual regulation scheme of the data center and the thermal power unit, the power flow of the 39-node system is calculated to verify the power control effect of the transmission section. The power flow calculation results show that the power flow of lines 14-15 and 16-17 is 46.3 MW and 226.1 MW, respectively, and the control error is 3.7 MW and 2.1 MW, respectively. Although there is an error between the results and the control target, the power control effect of the transmission section of the data center and the thermal power unit is still reflected.

[0057] That is, in engineering practice, absolute accuracy does not exist. The error of 3.7 MW and 2.1 MW relative to the control target (50 MW and 224 MW) is very small (7.4% and 0.9%), which is within an acceptable range. This result proves the necessity of introducing steps S7 to S9. If the scheme in Table 1 is directly executed, it may cause business interruption due to the neglect of QoS constraints, or it may not be executed at all. The current scheme is a "best scheme that can be achieved under real constraints".

[0058] In summary, the technical scheme described in the present application builds a solid technical pyramid: the tower foundation (data and physics): Monte Carlo + optimal power flow, which ensures the scientificity and optimality of the method. The tower body (intelligent core): the collaborative control model provides super-fast decision-making ability. The tower top (landing guarantee): considering the space-time transfer optimization of QoS ensures the actual feasibility and business friendliness of the scheme. This three-layer structure, from theory to data, from data to intelligence, and from intelligence back to physical constraints, forms a complete, rigorous and highly innovative technical closed loop, providing a valuable "source-load interaction" model for future high-proportion new energy access power systems.

[0059] Embodiment 2

[0060] Please refer to Figure 3 The embodiment provides a power grid transmission section power control system structure schematic diagram.

[0061] As an example, the system adopts the power grid transmission section power control method described in embodiment 1, and is characterized in that the system comprises: The training sample set generation module 30 is suitable for obtaining a data center load fluctuation interval in a power system; based on a Monte Carlo method, random sampling is performed in the data center load fluctuation interval to generate a plurality of data center load scenarios; for each data center load scenario, optimal power flow calculation is performed to obtain corresponding power transmission section power values and thermal power unit output values; the power transmission section power values are used as input features, and the corresponding data center load values and thermal power unit output values are used as output targets to form training samples; until a predetermined number of training sample sets are generated.

[0062] The data center and thermal power unit cooperative control model construction module 31 is suitable for training a preset neural network base model using the training sample set to obtain a data center and thermal power unit cooperative control model.

[0063] The data center and thermal power unit theoretical operation regulation target calculation module 32 is suitable for calculating the theoretical operation regulation targets of each data center and thermal power unit based on the power regulation requirements of the power transmission section of the power system and using the data center and thermal power unit cooperative control model.

[0064] The data center theoretical operation regulation target optimization module 33 is suitable for receiving the data center theoretical operation regulation targets output by the data center and thermal power unit cooperative control model, and correcting the data center theoretical operation regulation targets based on a data center load space-time transfer optimization model; wherein the data center load space-time transfer optimization model decomposes the total data center load into invariable load, transferable load, and adjustable load.

[0065] The space-time transfer optimization model solving module 34 is suitable for solving the data center load space-time transfer optimization model to obtain a final executable regulation scheme containing the transferable load and adjustable load power of each data center.

[0066] The regulation instruction output module 35 is suitable for outputting the final executable regulation scheme as an executable data center load regulation instruction and outputting the theoretical operation regulation targets of the thermal power unit as an executable thermal power unit output regulation instruction.

[0067] It is not difficult to find that the present embodiment is a system embodiment corresponding to the first embodiment, and the present embodiment can be implemented in cooperation with the first embodiment. The related technical details mentioned in the first embodiment are still valid in the present embodiment. In order to reduce repetition, they will not be described here. Correspondingly, the related technical details mentioned in the present embodiment can also be applied in the first embodiment.

[0068] It is worth mentioning that each module involved in the embodiment is a logic unit, which can be a physical unit, a part of a physical unit, or a combination of multiple physical units in actual application. In addition, in order to highlight the innovative part of the present application, units not closely related to solving the technical problems proposed by the present application are not introduced in the embodiment, but this does not mean that there are no other units in the embodiment.

[0069] Embodiment 3

[0070] Please refer to Figure 4 The embodiment of the present application also provides an electronic device, comprising a memory and a processor; the memory stores at least one program instruction; and the processor implements the power control method of the power transmission section of the power grid by loading and executing the at least one program instruction.

[0071] The memory 402 and the processor 401 are connected in a bus manner, and the bus can include any number of interconnected buses and bridges, which connect one or more processors 401 and various circuits of the memory 402 together. The bus can also connect various other circuits such as peripheral devices, voltage stabilizers, and power management circuits together, which are well known in the art, and therefore, further description is not given herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be one element or multiple elements such as multiple receivers and transmitters, which provide a unit for communicating with various other devices on the transmission medium. The data processed by the processor 401 is transmitted on the wireless medium through the antenna, and further, the antenna also receives data and transmits the data to the processor 401.

[0072] The processor 401 is responsible for managing the bus and general processing, and can also provide various functions including timing, peripheral interface, voltage regulation, power management, and other control functions. The memory 402 can be used to store data used by the processor 401 in performing operations.

[0073] Embodiment 4

[0074] The embodiment of the present application also provides a storage medium, which stores the power control method of the power transmission section of the power grid, and the power grid transmission section power control program is executed by the processor to realize the steps of the power grid transmission section power control method as described above. Since the storage medium adopts all the technical solutions of all the embodiments described above, it at least has all the beneficial effects brought by the technical solutions of the above embodiments, which will not be described here.

[0075] The above-mentioned are only embodiments of the present application, and the common knowledge of specific structures and characteristics in the scheme is not described too much herein. The ordinary skilled person in the art knows all the ordinary technical knowledge in the field of the present application before the application date or the priority date, can know all the prior art in the field, and has the ability to apply conventional experimental means before that date. The ordinary skilled person in the art can perfect and implement the present scheme under the guidance of the present application, combined with their own ability. Some typical known structures or known methods should not be an obstacle for the ordinary skilled person in the art to implement the present application. It should be noted that, for those skilled in the art, without departing from the structure of the present application, a number of modifications and improvements can be made, which should also be considered as the protection scope of the present application. The protection scope of the present application should be subject to the content of its claims, and the specific implementation mode and the like in the specification can be used to explain the content of the claims.

Claims

1. A power control method for power transmission sections in a power grid, characterized in that, The method includes: Step S1: Obtain the load fluctuation range of the data center in the power system; Step S2: Based on the Monte Carlo method, random sampling is performed within the data center load fluctuation range to generate multiple data center load scenarios; Step S3: For each data center load scenario, perform optimal power flow calculation to obtain the corresponding power value of the transmission section and the output value of the thermal power unit; Step S4: Use the power value of the power transmission section as the input feature, and the corresponding data center load value and thermal power unit output value as the output target to form a training sample; Step S5: Repeat steps S1 to S4 until a predetermined number of training sample sets are generated. Step S6: Use the training sample set to train the preset neural network basic model to obtain the collaborative control model of data center and thermal power unit; Step S7: Based on the power regulation requirements of the power system transmission section, the theoretical operation regulation targets of each data center and thermal power unit are calculated using the collaborative control model of the data center and thermal power unit. Step S8: Receive the theoretical operation and control targets of each data center output by the collaborative control model of the data center and thermal power unit, and modify the theoretical operation and control targets of the data center based on the data center load spatiotemporal transfer optimization model; wherein, the data center load spatiotemporal transfer optimization model decomposes the total load of the data center into immutable load, transferable load and adjustable load; Step S9: Solve the data center load spatiotemporal transfer optimization model to obtain the final executable control scheme that includes the transferable load and adjustable load power of each data center; Step S10: Output the final executable control scheme as an executable data center load control instruction, and output the theoretical operation control target of the thermal power unit as an executable thermal power unit output control instruction.

2. The power control method for power transmission sections in a power grid according to claim 1, characterized in that, In step S2, the random sampling in the Monte Carlo method adopts a uniform distribution, and the load of each data center is generated randomly with equal probability within the fluctuation range formed by its minimum load value and maximum load value. Power system samples randomly drawn using the Monte Carlo method ; Step S3 includes: calculating the power of the power grid data section using the optimal power flow method. and the output of thermal power units ;in , and These represent the load values ​​of each data center, the power of each line in the transmission section, and the output of each thermal power unit, respectively; n, m, and s represent the number of data centers, the number of transmission lines, and the number of thermal power units, respectively. Step S4 includes: combining data section power, data center load, and thermal power unit output to generate training samples, so as to... express.

3. The power control method for power transmission sections in a power grid according to claim 1, characterized in that, The optimal power flow calculation includes: minimizing the total power generation cost of the system as the objective function, and the constraints include: system active power balance constraints; upper and lower limits of output of each thermal power unit constraints; constraints that the power of each transmission line does not exceed its transmission capacity limit; and constraints that the voltage of each node is within the safe operating range.

4. The power control method for power transmission sections in a power grid according to claim 3, characterized in that, The system's active power balance constraint includes the requirement that the total power generated by all thermal power units equals the total power consumed by all loads plus the transmission loss of the power grid. The upper and lower limits of the output of each thermal power unit include the range of the output of each thermal power unit within the range of its maximum and minimum technical output values; The constraint that the power of each transmission line does not exceed its transmission capacity limit includes that the power flowing through any line cannot exceed its thermal stability limit. The constraint on the voltage of each node within the safe operating range includes maintaining the voltage of all nodes between preset safe upper and lower limits.

5. The power control method for power transmission sections of a power grid according to claim 1, characterized in that, Step S6 includes: The power of each line in the power grid transmission section is taken as the input characteristic, and the load control target of each data center and the output control target of each thermal power unit are taken as the output. An artificial neural network is used as the basic model, with the number of hidden layer nodes set to N1-N2, and the sigmoid function is used as the activation function. The generated training samples are divided into training set and validation set according to a preset ratio. The artificial neural network model is trained using the training set and the accuracy of the artificial neural network model is verified using the validation set. When the average error between the output of the artificial neural network model and the actual value of the sample reaches the preset condition, the model construction is completed.

6. The power control method for power transmission sections of a power grid according to claim 1, characterized in that, The theoretical operation and control target of the data center is the theoretical load that needs to be adjusted for each data center. The theoretical operation and control target of the thermal power unit is the theoretical output value that needs to be adjusted for each thermal power unit.

7. The power control method for power transmission sections in a power grid according to claim 2, characterized in that, The mathematical expression of the data center load spatiotemporal transfer optimization model in step S8 is as follows: ； The constraints are: ； ； ； In the formula, , and These represent the fixed load power, transferable load power, and adjustable load power of data center i, respectively. The theoretical operational control target for data center i; and These are the upper and lower limits of the adjustable load power for data center i, respectively. The data transmission latency between data centers i and j; The upper limit of inter-data center data transmission latency that affects the quality of data center services.

8. The power control method for power transmission sections in a power grid according to claim 7, characterized in that, Step S9 includes: using a particle swarm optimization algorithm to solve the data center load spatiotemporal transfer optimization model, specifically including the following steps: Step S901: Initialize the particle swarm: Set the number of particles to ab, with each particle corresponding to a group. and The combination; Step S902: Calculate fitness: Use the deviation of the objective function as the fitness function. The smaller the deviation, the higher the fitness. Step S903, Iterative Update: Update the velocity and position of the particles, and retain the particles with the best fitness. Step S904, Termination of Iteration: When the number of iterations reaches the preset number or the optimal fitness remains unchanged for N consecutive iterations, the iteration stops. At this time, the optimal particle corresponds to... and That is, the solution result, combined with The final executable control scheme for data center i is obtained.

9. The power control method for power transmission sections of a power grid according to claim 8, characterized in that, The final executable control scheme includes: By solving the data center load spatiotemporal transfer optimization model, the total optimized load power of each data center is obtained by summing the transferable load power, the adjustable load power, and the invariable load, which serves as the final executable control scheme.

10. A power control system for power transmission sections in a power grid, wherein the system is implemented using the power control method for power transmission sections in any one of claims 1-9, characterized in that, The system includes: The training sample set generation module is suitable for obtaining the load fluctuation range of data centers in a power system. Based on the Monte Carlo method, it randomly samples within the load fluctuation range of the data centers to generate multiple data center load scenarios. For each data center load scenario, it performs optimal power flow calculation to obtain the corresponding transmission section power value and thermal power unit output value. It uses the transmission section power value as input features and the corresponding data center load value and thermal power unit output value as output targets to combine them to form training samples. This process continues until a predetermined number of training sample sets are generated. The data center and thermal power unit collaborative control model construction module is suitable for using the training sample set to train a preset neural network basic model to obtain a data center and thermal power unit collaborative control model. The theoretical operation and control target calculation module for data centers and thermal power units is applicable to power control requirements based on power transmission sections of the power system. Using the collaborative control model of data centers and thermal power units, the theoretical operation and control targets of each data center and thermal power unit are calculated. The data center theoretical operation control target optimization module is suitable for receiving the theoretical operation control targets of each data center output by the collaborative control model of the data center and thermal power unit, and for correcting the theoretical operation control targets of the data center based on the data center load spatiotemporal transfer optimization model; wherein, the data center load spatiotemporal transfer optimization model decomposes the total load of the data center into immutable load, transferable load and adjustable load; The spatiotemporal transfer optimization model solving module is suitable for solving the spatiotemporal transfer optimization model of the data center load, and obtaining the final executable control scheme that includes the transferable load and adjustable load power of each data center; The control command output module is suitable for outputting the final executable control scheme as an executable data center load control command, and outputting the theoretical operation control target of the thermal power unit as an executable output control command of the thermal power unit.

Images