AC access flywheel UPS control method for stabilizing dynamic load of data center

Abstract

The invention discloses an AC access flywheel UPS control method for stabilizing a dynamic load in a data center, and relates to the field of power supply management, and the method comprises the following steps: S1, predicting and analyzing the load fluctuation of the data center, collecting the operation data of a flywheel through a sensor, and carrying out the monitoring and modeling of the state of a flywheel UPS; according to the method, data fitting and pattern recognition can be more efficiently carried out, the training process needing a large amount of time and computing resources in a traditional method is reduced, the system response time is shortened, excessive dependence on a preset model and a static rule is avoided, more flexible strategy selection is effectively provided, and the capacity of the system for coping with changes is improved; the flywheel UPS system can quickly respond to load change, enhance the robustness and stability of the system, avoid overload and low-efficiency operation of single equipment, efficiently evaluate the state and performance of the flywheel UPS system, avoid excessive consumption or waste of energy, improve the energy utilization rate, prolong the service life of the equipment and reduce the maintenance cost.

Description

Technical Field

[0001] The present invention relates to the field of power supply management, and in particular to an AC access flywheel UPS control method for stabilizing dynamic loads in a data center. Background Art

[0002] With the continuous development of information technology and the advancement of digital transformation, data centers have become key infrastructure to support various information services in society. As the core infrastructure supporting technologies such as cloud computing, big data analysis and artificial intelligence, data centers have become an important part of modern society. The operation and maintenance of data centers need to ensure stable and efficient power supply, especially when facing dynamic load fluctuations. How to ensure the continuous supply of power has become a key issue. The surge in global data traffic and large-scale computing needs have increased the power consumption of data centers year by year, while the instability of power supply and dynamic load fluctuations have brought great challenges to the efficient operation of data centers. Traditional power supply systems often face challenges such as grid load fluctuations, uneven equipment loads and power failures, resulting in service interruptions and equipment overloads in data centers. To meet this challenge, more and more data centers have begun to introduce advanced energy storage systems, especially flywheel UPS systems, to balance the fluctuations in power demand and ensure the stability of power supply. As a physical energy storage device, flywheel UPS stores energy by rotating flywheels and provides emergency power when power demand increases suddenly. It has the advantages of fast response, long life and efficient energy conversion. However, in practical applications, how to accurately adjust the flywheel UPS under dynamic load fluctuations and how to achieve coordinated work between different energy storage units are still technical problems that need to be solved urgently.

[0003] After searching, Chinese patent number CN109560549A discloses a control method and system for a backup flywheel energy storage UPS. Although the invention uses less equipment, has a low installation and use threshold, high efficiency, fast response, and high output voltage quality, but; in addition, the existing flywheel UPS control method, for this reason, we propose an AC access flywheel UPS control method for smoothing dynamic loads in data centers. Summary of the invention

[0004] The purpose of the present invention is to solve the above-mentioned problem and to provide an AC access flywheel UPS control method for smoothing dynamic loads in a data center.

[0005] The present invention proposes an AC access flywheel UPS control method for stabilizing dynamic loads in a data center. The technical solution adopted to solve the technical problem is: S1: Predict and analyze the load fluctuations of the data center, collect the operating data of the flywheel through sensors, and monitor and model the flywheel UPS status; S2: Based on the predicted load fluctuations, simulate the collaborative work between different energy storage systems to create a load adjustment plan and provide feedback to the staff for review and adjustment; S3: Dynamically optimize the charge and discharge cycle and power adjustment strategy of the flywheel UPS based on the monitoring modeling results and the actual load conditions collected by the sensors; S4: Monitor the operating status of the flywheel and external load changes through sensors and data acquisition systems, and adaptively adjust the flywheel UPS based on real-time monitoring data; S5: Establish a digital twin model of the flywheel UPS, synchronize various data and control schemes of the flywheel UPS in real time, evaluate the energy conversion efficiency of the flywheel UPS, and adjust the control parameters; S6: Collect and analyze flywheel UPS performance data in real time, evaluate the effect of the current control strategy, and readjust the flywheel UPS control strategy based on the evaluation results.

[0006] As a further solution of the present invention, the specific steps of predicting and analyzing the load fluctuation of the data center in S1 are as follows: P1.1: The flywheel UPS control platform collects multiple groups of load data collected by the data center, removes noise and abnormal data from each group of load data collected, and fills in missing data through interpolation. Then, it uses Z-score standardization to unify each group of load data into a specified range, and then uses variance screening method to extract required feature information from each group of load data, extracts historical load data from the data center, and divides it into training set, test set and validation set, and divides the training set and validation set into multiple training subsets and validation subsets respectively; P1.2: Create a predictive analysis model based on the quantum support vector machine architecture. At the beginning of each round of training, the training subset data is input into the predictive analysis model in turn. The model maps each training set data to the quantum space through the angle encoding method and generates corresponding quantum bits. Then, through the quantum superposition effect, multiple sets of quantum bits of characteristic information exist simultaneously in one quantum state. After that, the predictive analysis model calculates the inner product between each quantum state in the quantum space through the quantum kernel function, and evaluates the similarity between each quantum state based on the calculated inner product. P1.3: Based on the calculated quantum kernel function values ​​between each set of training subset data, the corresponding kernel matrix is ​​constructed, and the corresponding objective function is constructed with the purpose of maximizing the interval between each input data. The constraints of the optimization target are set, and then the objective function is solved by the gradient descent algorithm based on the constraints to obtain the optimized prediction analysis model parameters; P1.4: After each round of model parameter optimization is completed, the validation subset is input into the predictive analysis model in turn, and the cross-validation error of the model is calculated to evaluate the average performance of the predictive analysis model. If the average performance of the model does not reach the preset threshold, the regularization parameter, number of quantum bits, and inner product calculation method of the predictive analysis model are adjusted, and the predictive analysis model is repeatedly trained and verified until the average performance of the model converges to the preset threshold. The performance of the trained predictive analysis model on unknown data is then tested using the test set data. If the test result reaches the preset expected value, the predictive analysis model is deployed on the flywheel UPS control platform of the data center. Otherwise, the predictive analysis model is retrained. P1.5: The preprocessed new load data is input into the prediction analysis model. The prediction analysis model forward propagates the new load data. The prediction analysis model converts the load data into quantum state through quantum feature mapping, calculates the similarity between the new data and the training data, and predicts the load data through the decision function. The classification decision is then made based on the decision function value to generate the corresponding category label, where the category label value is 0 and 1, 0 indicates that the load fluctuation exceeds the preset threshold, and 1 indicates that the load fluctuation is within the normal range.

[0007] It should be further explained that the specific calculation formula of the angle encoding method described in P1.2 is as follows: In the formula, Representative Input data ; Represents around Quantum manipulation of axis rotation; represents the angular parameter of the quantum bit mapping, where , Represents a constant used to adjust the amplitude of the data; represents the initial state of the quantum bit; The specific calculation formula of the quantum kernel function described in P1.2 is as follows: In the formula, represents the value of the quantum kernel function, i.e. Input data The quantum state is Input data The similarity between the quantum states; and Respectively represent Input data With Input data The quantum state after mapping; Represents the inner product between two sets of quantum states.

[0008] The specific calculation formula of the decision function described in P1.5 is as follows: In the formula, Representative New load data; Represents the total number of training data; Representative Lagrange multipliers; Representative Training data Labels; Represents new load data and training data The kernel function value between ; Represents the bias term.

[0009] As a further solution of the present invention, the specific steps of monitoring and modeling the flywheel UPS state described in S1 are as follows: P2.1: After the load fluctuation prediction is completed, various operating data of the flywheel UPS system are collected in real time through various sensors, including speed, voltage, power, charging status, torque and temperature, and the time of collection of various operating parameters is recorded. The collected data is denoised and normalized, and then the flywheel dynamic system model is constructed based on the processed flywheel operating data; P2.2: Take the current flywheel state as the root node of the operation tree, randomly execute various operations of the current flywheel according to the flywheel dynamic system model and the current flywheel state, and generate new states based on each group of operations executed, generate the lower-level child nodes of the root node of the current operation tree, and initialize the number of visits and reward values ​​of each node of the current operation tree; P2.3: Starting from the root node of the current operation tree, calculate the UCB value of each group of child nodes, and then select the child node with the highest UCB value layer by layer according to the upper confidence bound selection strategy until an unvisited child node is selected. Then stop selecting, and according to the flywheel state of the current child node and the flywheel dynamic system model, randomly execute various operations of the current flywheel to expand a new flywheel state, and add it as a new child node to the operation tree; P2.4: Randomly select an operation path for simulation. During the simulation, update the flywheel state through the flywheel dynamic system model until the set simulation time is reached, then stop the simulation, calculate the final simulation reward value based on the flywheel power consumption, energy storage and efficiency during the simulation, and then propagate the simulated reward value from the current simulation node back to the root node, and update the reward value and visit count of each child node in the same path; P2.5: Through multiple rounds of selection, expansion, simulation and backtracking, until the preset number of iterations is reached, then starting from the root node of the operation tree, traverse each child node layer by layer, and select the operation path with the maximum reward according to the reward value of each child node. After the operation path is found, make real-time decisions based on the path, control the operation of the flywheel UPS, and continuously monitor and adjust the operation path according to the real-time flywheel data.

[0010] It should be further explained that the specific expression of the flywheel dynamic system model described in P2.1 is as follows: In the formula, Representative time Angular momentum of the flywheel at ; represents the moment of inertia of the flywheel; Representative time The speed of the flywheel at 1: Representative time Energy storage of the flywheel; Representative time Input power at ; Representative time The external torque on the flywheel at the time; Representative time Energy storage of the flywheel; Represents the discrete time step.

[0011] As a further solution of the present invention, the specific steps of simulating the collaborative work between different energy storage systems to create a load regulation solution in S2 are as follows: P3.1: According to the design requirements of the flywheel UPS system and the initial load fluctuation, the energy state of each energy storage unit of the flywheel UPS is set, and the generated optimal operation path is used as the initial collaborative work strategy. At the same time, with the goal of maximizing energy utilization, minimizing power loss and extending equipment service life, the corresponding fitness function is constructed, and according to the dynamic changes of energy in each storage unit of the flywheel UPS system and the interaction between different energy storage units, the corresponding virtual ecological model is generated; P3.2: Initialize a group of populations according to the number of storage units in the flywheel UPS system, simulate the dynamic energy changes and interactions of each storage unit under the collaborative working strategy based on the virtual ecological model, calculate the fitness value of each storage unit through the fitness function, and then screen out the storage units below the preset selection threshold; P3.3: Copy the remaining storage units, and adjust the collaborative working strategy of the copied storage units by adjusting the charge and discharge power and the energy flow path according to the energy state of the copied storage units. Then, combine the adjusted storage units with the remaining storage units into a new population, and use the newly generated population as the population for the next iteration; P3.4: The population is updated repeatedly until the charging and discharging strategies of each storage unit are no longer adjusted and the fitness value converges to the preset optimization target. After that, the virtual ecological model outputs the final charging and discharging strategy, energy flow path and energy status of each storage unit, and implements it in the actual flywheel UPS system while performing real-time control.

[0012] As a further solution of the present invention, the specific calculation formula of the fitness function described in P3.1 is as follows: In the formula, Representative Energy loss of each storage unit; Representative The charging efficiency of each storage unit; Representative The input energy of each storage unit; Representative The discharge efficiency of each storage unit; Representative Output energy of each storage unit; Representative Energy utilization of each storage unit; Represents the overall energy loss of the flywheel UPS system and; Representative Energy loss of each storage unit; Represents the overall energy utilization of the flywheel UPS system; Representative Energy utilization of each storage unit; Representative The service life of each storage unit; Representative The number of charge and discharge cycles of a storage unit; Representative Energy loss during each charge and discharge process of a storage unit; Represents the overall service life of the flywheel UPS system; Representative The service life of each storage unit; , as well as Both represent the total number of storage units in the flywheel UPS system; Represents the overall fitness value of the flywheel UPS system; represents the weight of energy utilization; represents the weight of the total power loss; The weight representing the useful life of the equipment.

[0013] As a further solution of the present invention, the specific steps of adaptively adjusting the flywheel UPS according to the real-time monitoring data in S4 are as follows: P4.1: Real-time monitoring of the operating status of the flywheel UPS system and external load changes, extracting the various states of the flywheel UPS system and the corresponding operations under each state from each set of data received, and constructing a state set based on the collected state information and operation information. And action set , then calculate the probability of transferring to the next random state after collecting any action in the action set in each state in the state set; P4.2: All state-action pairs The initial value of is set to 0, where , , collect all state-action pairs and build a corresponding data repository to store each state-action pair and its corresponding value, and then at each moment , according to the current state and the information in the data repository, select Action implement; P4.3: Selecting an action When a random number is generated, if the random number is higher than the preset selection threshold , then from the action set Randomly select an action, otherwise select The action with the largest value, After the selection is completed, the simulated flywheel UPS system executes the action and enters the next state ; P4.4: Calculate the instant reward in the corresponding state after executing any action based on energy efficiency, power loss and equipment life. , use the Q value update formula to update the current state-action pair , and update the data repository, and then repeat the action selection and data repository update until the state-action pair The value changes converge to the preset range; P4.5: After the iteration, traverse the state-action pairs in the data repository value, and select the state-action pair with the highest value as the optimal response strategy, and at the same time, according to the feedback and The value is updated to pre-set the future operation strategy of the flywheel UPS system.

[0014] Beneficial effects of the present invention: 1. The present invention can perform data fitting and pattern recognition more efficiently, and reduce the training process that requires a lot of time and computing resources in traditional methods, shortening the system response time. At the same time, through quantum machine learning, the flywheel UPS system can adjust the control strategy in real time according to load fluctuations, avoid over-reliance on preset models and static rules, effectively provide more flexible strategy selection, and enhance the system's ability to cope with changes.

[0015] 2. The present invention uses dynamic system modeling to enable the flywheel UPS system to quickly respond to load changes, enhance the robustness and stability of the system, and efficiently evaluate the status and performance of the flywheel UPS system. It also avoids excessive consumption or waste of energy and improves energy utilization by precisely controlling the charge and discharge strategies of the flywheel UPS. It also ensures load balancing and optimal resource allocation among the units within the flywheel UPS system by simulating the interaction between the various energy units and intelligent optimization strategies, thereby avoiding overloading and inefficient operation of a single device, extending equipment life, and reducing maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described below in conjunction with the accompanying drawings.

[0017] Figure 1 Framework diagram of the AC access flywheel UPS control method for smoothing dynamic loads in data centers. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0020] Example 1 The embodiment of the present invention provides an AC access flywheel UPS control method for stabilizing dynamic loads in a data center. Figure 1 , Figure 1 A framework diagram of an AC access flywheel UPS control method for stabilizing dynamic loads in a data center provided by an embodiment of the present invention. The method comprises the following steps: S1: Predict and analyze the load fluctuations of the data center, collect the operating data of the flywheel through sensors, and monitor and model the flywheel UPS status.

[0021] Specifically, the flywheel UPS control platform collects multiple groups of load data collected by the data center, removes noise and abnormal data from each group of load data collected, and fills in missing data through interpolation. After that, it uses Z-score standardization to unify each group of load data within the specified range, and then uses variance screening method to extract required feature information from each group of load data. It extracts historical load data from the data center and divides it into training set, test set and validation set, and divides the training set and validation set into multiple training subsets and validation subsets respectively. A predictive analysis model is created based on the quantum support vector machine architecture. At the beginning of each round of training, the training subsets are divided into The data of each training set is input into the prediction and analysis model. The model maps each training set data to the quantum space through the angle encoding method and generates corresponding quantum bits. Then, through the quantum superposition effect, the quantum bits of multiple sets of characteristic information exist in one quantum state at the same time. After that, the prediction and analysis model calculates the inner product between each quantum state in the quantum space through the quantum kernel function, and evaluates the similarity between each quantum state based on the calculated inner product. According to the calculated quantum kernel function value between each set of training subset data, the corresponding kernel matrix is ​​constructed, and the corresponding objective function is constructed with the purpose of maximizing the interval between each input data, and the constraints of the optimization objective are set. Then, based on Constraints are set, and the objective function is solved by the gradient descent algorithm to obtain the optimized predictive analysis model parameters. After each round of model parameter optimization is completed, the verification subset is input into the predictive analysis model in turn, and the cross-validation error of the model is calculated to evaluate the average performance of the predictive analysis model. If the average performance of the model does not reach the preset threshold, the regularization parameter of the predictive analysis model, the number of quantum bits, and the inner product calculation method of the hyperparameters are adjusted, and the predictive analysis model is repeatedly trained and verified until the average performance of the model converges to the preset threshold. The test set data is then used to detect the performance of the trained predictive analysis model on unknown data. If the test result If the preset expected value is reached, the predictive analysis model is deployed to the flywheel UPS control platform of the data center. Otherwise, the predictive analysis model is retrained and the preprocessed new load data is input into the predictive analysis model. The predictive analysis model forward propagates the new load data. The predictive analysis model converts the load data into a quantum state through quantum feature mapping, calculates the similarity between the new data and the training data, and predicts the load data through a decision function. Then, a classification decision is made based on the decision function value to generate a corresponding category label, where the category label value is 0 and 1, 0 indicates that the load fluctuation exceeds the preset threshold, and 1 indicates that the load fluctuation is within the normal range.

[0022] Specifically, after the load fluctuation prediction is completed, various operating data of the flywheel UPS system are collected in real time through various sensors, including speed, voltage, power, charging status, torque and temperature, and the time of collection of various operating parameters is recorded. The collected data is denoised and normalized, and then the flywheel dynamic system model is constructed based on the processed flywheel operation data. The current flywheel state is used as the root node of the operation tree. According to the flywheel dynamic system model and the current flywheel state, various operations of the current flywheel are randomly executed, and new states are generated based on each group of operations executed. The lower-level child nodes of the root node of the current operation tree are generated, and the number of visits and reward values ​​of each node of the current operation tree are initialized. Starting from the root node of the current operation tree, the UCB value of each group of child nodes is calculated, and then the child node with the highest UCB value is selected layer by layer according to the upper confidence bound selection strategy until an unvisited child node is selected, and the selection is stopped, and the flywheel state and flywheel state of the current child node are selected. The flywheel dynamic system model is used to randomly execute various operations of the current flywheel to expand a new flywheel state, and add it as a new child node to the operation tree. The operation path is randomly selected for simulation. During the simulation, the flywheel state is updated through the flywheel dynamic system model until the set simulation time is reached, and then the simulation is stopped. The final simulation reward value is calculated based on the power consumption, energy storage and efficiency of the flywheel during the simulation, and then the simulated reward value is back-propagated from the current simulation node back to the root node, and the reward value and the number of visits of each child node in the same path are updated. Through multiple rounds of selection, expansion, simulation and backtracking, until the preset number of iterations is reached, then starting from the root node of the operation tree, each child node is traversed layer by layer downward, and the operation path with the maximum reward is selected according to the reward value of each child node. After the operation path is found, real-time decisions are made based on the path, and the operation of the flywheel UPS is controlled. The operation path is continuously monitored and adjusted according to real-time flywheel data.

[0023] In addition, it should be noted that the specific calculation formula of the angle encoding method is as follows: In the formula, Representative Input data ; Represents around Quantum manipulation of axis rotation; represents the angular parameter of the quantum bit mapping, where , Represents a constant used to adjust the amplitude of the data; represents the initial state of the quantum bit; The specific calculation formula of the quantum kernel function is as follows: In the formula, represents the value of the quantum kernel function, i.e. Input data The quantum state is Input data The similarity between the quantum states; and Respectively represent Input data With Input data The quantum state after mapping; Represents the inner product between two sets of quantum states.

[0024] The specific calculation formula of the decision function is as follows: In the formula, Representative New load data; Represents the total number of training data; Representative Lagrange multipliers; Representative Training data Labels; Represents new load data and training data The kernel function value between ; represents the bias term; The specific expression of the flywheel dynamic system model is as follows: In the formula, Representative time Angular momentum of the flywheel at ; represents the moment of inertia of the flywheel; Representative time The speed of the flywheel at 1: Representative time Energy storage of the flywheel; Representative time Input power at ; Representative time The external torque on the flywheel at the time; Representative time Energy storage of the flywheel; Represents the discrete time step.

[0025] S2: Based on the predicted load fluctuations, simulate the collaborative work between different energy storage systems to create a load adjustment plan and provide feedback to the staff for review and adjustment.

[0026] Specifically, according to the design requirements of the flywheel UPS system and the initial load fluctuation, the energy state of each energy storage unit of the flywheel UPS is set, and the generated optimal operation path is used as the initial collaborative work strategy. At the same time, with the goal of maximizing energy utilization, minimizing power loss and extending equipment service life, a corresponding fitness function is constructed, and according to the dynamic changes of energy of each storage unit in the flywheel UPS system and the interaction between different energy storage units, a corresponding virtual ecological model is generated, and a group of populations is initialized according to the number of storage units in the flywheel UPS system. The dynamic changes of energy and the interaction of each storage unit under the collaborative work strategy are simulated based on the virtual ecological model, and the fitness function is used to calculate the energy consumption of each storage unit. The fitness value of the unit is then screened out, and the storage units below the preset selection threshold are copied. According to the energy state of the copied storage unit, the collaborative working strategy of the copied storage unit is adjusted by adjusting the charging and discharging power and the energy flow path. The adjusted storage unit is then combined with the remaining storage units into a new population, and the newly generated population is used as the population for the next iteration. The population update is repeated until the charging and discharging strategy of each storage unit is no longer adjusted and the fitness value converges to the preset optimization target. After that, the virtual ecological model outputs the final charging and discharging strategy, energy flow path and energy state of each storage unit, and implements it in the actual flywheel UPS system, while performing real-time control.

[0027] The specific calculation formula of the fitness function is as follows: In the formula, Representative Energy loss of each storage unit; Representative The charging efficiency of each storage unit; Representative The input energy of each storage unit; Representative The discharge efficiency of each storage unit; Representative Output energy of each storage unit; Representative Energy utilization of each storage unit; Represents the overall energy loss of the flywheel UPS system and; Representative Energy loss of each storage unit; Represents the overall energy utilization of the flywheel UPS system; Representative Energy utilization of each storage unit; Representative The service life of each storage unit; Representative The number of charge and discharge cycles of a storage unit; Representative Energy loss during each charge and discharge process of a storage unit; Represents the overall service life of the flywheel UPS system; Representative The service life of each storage unit; , as well as Both represent the total number of storage units in the flywheel UPS system; Represents the overall fitness value of the flywheel UPS system; represents the weight of energy utilization; represents the weight of the total power loss; The weight representing the useful life of the equipment.

[0028] Example 2 The embodiment of the present invention provides an AC access flywheel UPS control method for stabilizing dynamic loads in a data center. Figure 1 , Figure 1 A framework diagram of an AC access flywheel UPS control method for stabilizing dynamic loads in a data center provided by an embodiment of the present invention. The method comprises the following steps: According to the monitoring modeling results and the actual load conditions collected by sensors, the charge and discharge cycle and power adjustment strategy of the flywheel UPS are dynamically optimized.

[0029] The operating status of the flywheel and external load changes are monitored through sensors and data acquisition systems, and the flywheel UPS is adaptively adjusted based on real-time monitoring data.

[0030] Specifically, the operating status of the flywheel UPS system and the external load changes are monitored in real time, the various states of the flywheel UPS system and the corresponding operations in each state are extracted from each set of received data, and a state set is constructed based on the collected state information and operation information. And action set , and then calculate the probability of transferring to the next random state after collecting any action in the action set under each state in the state set, and .... The initial value of is set to 0, where , , collect all state-action pairs and build a corresponding data repository to store each state-action pair and its corresponding value, and then at each moment , according to the current state and the information in the data repository, select Action Execute, select action When a random number is generated, if the random number is higher than the preset selection threshold , then from the action set Randomly select an action, otherwise select The action with the largest value, After the selection is completed, the simulated flywheel UPS system executes the action and enters the next state , calculate the instant reward in the corresponding state after executing any action based on energy efficiency, power loss and equipment life, and calculate the instant reward and next state based on the instant reward obtained , use the Q value update formula to update the current state-action pair , and update the data repository, and then repeat the action selection and data repository update until the state-action pair The value change converges to the preset range. After the iteration is completed, the state-action pairs in the data repository are traversed. value, and select the state-action pair with the highest value as the optimal response strategy, and at the same time, according to the feedback and The value is updated to pre-set the future operation strategy of the flywheel UPS system.

[0031] A digital twin model of the flywheel UPS is established to synchronize various data and control schemes of the flywheel UPS in real time, evaluate the energy conversion efficiency of the flywheel UPS, and adjust the control parameters.

[0032] Collect and analyze flywheel UPS performance data in real time, evaluate the effectiveness of the current control strategy, and readjust the flywheel UPS control strategy based on the evaluation results.

[0033] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. An AC access flywheel UPS control method for stabilizing dynamic loads in a data center, characterized in that: The following steps are involved: S1: Predict and analyze the load fluctuations of the data center, collect the operating data of the flywheel through sensors, and monitor and model the flywheel UPS status; S2: Based on the predicted load fluctuations, simulate the collaborative work between different energy storage systems to create a load adjustment plan and provide feedback to the staff for review and adjustment; S3: Dynamically optimize the charge and discharge cycle and power adjustment strategy of the flywheel UPS based on the monitoring modeling results and the actual load conditions collected by the sensors; S4: Monitor the operating status of the flywheel and external load changes through sensors and data acquisition systems, and adaptively adjust the flywheel UPS based on real-time monitoring data; S5: Establish a digital twin model of the flywheel UPS, synchronize various data and control schemes of the flywheel UPS in real time, evaluate the energy conversion efficiency of the flywheel UPS, and adjust the control parameters; S6: Collect and analyze flywheel UPS performance data in real time, evaluate the effect of the current control strategy, and readjust the flywheel UPS control strategy based on the evaluation results.

2. The AC access flywheel UPS control method for balancing dynamic loads in a data center according to claim 1 is characterized in that: The specific steps of predicting and analyzing the load fluctuation of the data center described in S1 are as follows: P1.1: The flywheel UPS control platform collects multiple groups of load data collected by the data center, removes noise and abnormal data from each group of load data collected, and fills in missing data through interpolation. Then, it uses Z-score standardization to unify each group of load data into a specified range, and then uses variance screening method to extract required feature information from each group of load data, extracts historical load data from the data center, and divides it into training set, test set and validation set, and divides the training set and validation set into multiple training subsets and validation subsets respectively; P1.2: Create a predictive analysis model based on the quantum support vector machine architecture. At the beginning of each round of training, the training subset data is input into the predictive analysis model in turn. The model maps each training set data to the quantum space through the angle encoding method and generates corresponding quantum bits. Then, through the quantum superposition effect, multiple sets of quantum bits of feature information exist simultaneously in one quantum state. After that, the predictive analysis model calculates the inner product between each quantum state in the quantum space through the quantum kernel function, and evaluates the similarity between each quantum state based on the calculated inner product. The specific calculation formula of the angle encoding method is as follows: In the formula, Representative Input data ; Represents around Quantum manipulation of axis rotation; represents the angular parameter of the quantum bit mapping, where , Represents a constant used to adjust the amplitude of the data; represents the initial state of the quantum bit; The specific calculation formula of the quantum kernel function is as follows: In the formula, represents the value of the quantum kernel function, i.e. Input data The quantum state is Input data The similarity between the quantum states; and Respectively represent Input data With Input data The quantum state after mapping; represents the inner product between two sets of quantum states; P1.3: Based on the calculated quantum kernel function values ​​between each set of training subset data, the corresponding kernel matrix is ​​constructed, and the corresponding objective function is constructed with the purpose of maximizing the interval between each input data. The constraints of the optimization target are set, and then the objective function is solved by the gradient descent algorithm based on the constraints to obtain the optimized prediction analysis model parameters; P1.4: After each round of model parameter optimization is completed, the validation subset is input into the predictive analysis model in turn, and the cross-validation error of the model is calculated to evaluate the average performance of the predictive analysis model. If the average performance of the model does not reach the preset threshold, the regularization parameter, number of quantum bits, and inner product calculation method of the predictive analysis model are adjusted, and the predictive analysis model is repeatedly trained and verified until the average performance of the model converges to the preset threshold. The performance of the trained predictive analysis model on unknown data is then tested using the test set data. If the test result reaches the preset expected value, the predictive analysis model is deployed on the flywheel UPS control platform of the data center. Otherwise, the predictive analysis model is retrained. P1.5: The new load data after preprocessing is input into the prediction analysis model. The prediction analysis model performs forward propagation on the new load data. The prediction analysis model converts the load data into quantum state through quantum feature mapping, calculates the similarity between the new data and the training data, and predicts the load data through the decision function. Then, the classification decision is made according to the decision function value to generate the corresponding category label, where the category label value is 0 and 1, 0 indicates that the load fluctuation exceeds the preset threshold, and 1 indicates that the load fluctuation is within the normal range. The specific calculation formula of the decision function is as follows: In the formula, Representative New load data; Represents the total number of training data; Representative Lagrange multipliers; Representative Training data Labels; Represents new load data and training data The kernel function value between ; Represents the bias term.

3. The AC access flywheel UPS control method for balancing dynamic loads in a data center according to claim 2, characterized in that: The specific steps of monitoring and modeling the flywheel UPS status described in S1 are as follows: P2.1: After the load fluctuation prediction is completed, various operating data of the flywheel UPS system are collected in real time through various sensors, including speed, voltage, power, charging status, torque and temperature, and the time of collection of various operating parameters is recorded. The collected data is denoised and normalized, and then the flywheel dynamic system model is constructed based on the processed flywheel operating data; P2.2: Take the current flywheel state as the root node of the operation tree, randomly execute various operations of the current flywheel according to the flywheel dynamic system model and the current flywheel state, and generate new states based on each group of operations executed, generate the lower-level child nodes of the root node of the current operation tree, and initialize the number of visits and reward values ​​of each node of the current operation tree; P2.3: Starting from the root node of the current operation tree, calculate the UCB value of each group of child nodes, and then select the child node with the highest UCB value layer by layer according to the upper confidence bound selection strategy until an unvisited child node is selected. Then stop selecting, and according to the flywheel state of the current child node and the flywheel dynamic system model, randomly execute various operations of the current flywheel to expand a new flywheel state, and add it as a new child node to the operation tree; P2.4: Randomly select an operation path for simulation. During the simulation, update the flywheel state through the flywheel dynamic system model until the set simulation time is reached, then stop the simulation, calculate the final simulation reward value based on the flywheel power consumption, energy storage and efficiency during the simulation, and then propagate the simulated reward value from the current simulation node back to the root node, and update the reward value and visit count of each child node in the same path; P2.5: Through multiple rounds of selection, expansion, simulation and backtracking, until the preset number of iterations is reached, then starting from the root node of the operation tree, traverse each child node layer by layer, and select the operation path with the maximum reward according to the reward value of each child node. After the operation path is found, make real-time decisions based on the path, control the operation of the flywheel UPS system, and continuously monitor and adjust the operation path according to the real-time flywheel data.

4. The AC access flywheel UPS control method for balancing dynamic loads in a data center according to claim 3 is characterized in that: The specific steps of simulating the collaborative work between different energy storage systems to create a load regulation solution as described in S2 are as follows: P3.1: According to the design requirements of the flywheel UPS system and the initial load fluctuation, the energy state of each energy storage unit of the flywheel UPS is set, and the generated optimal operation path is used as the initial collaborative work strategy. At the same time, with the goal of maximizing energy utilization, minimizing power loss and extending equipment service life, the corresponding fitness function is constructed, and according to the dynamic changes of energy in each storage unit of the flywheel UPS system and the interaction between different energy storage units, the corresponding virtual ecological model is generated; P3.2: Initialize a group of populations according to the number of storage units in the flywheel UPS system, simulate the dynamic energy changes and interactions of each storage unit under the collaborative working strategy based on the virtual ecological model, calculate the fitness value of each storage unit through the fitness function, and then screen out the storage units below the preset selection threshold; P3.3: Copy the remaining storage units, and adjust the collaborative working strategy of the copied storage units by adjusting the charge and discharge power and the energy flow path according to the energy state of the copied storage units. Then, combine the adjusted storage units with the remaining storage units into a new population, and use the newly generated population as the population for the next iteration; P3.4: The population is updated repeatedly until the charging and discharging strategies of each storage unit are no longer adjusted and the fitness value converges to the preset optimization target. After that, the virtual ecological model outputs the final charging and discharging strategy, energy flow path and energy status of each storage unit, and implements it in the actual flywheel UPS system while performing real-time control.

5. The AC access flywheel UPS control method for balancing dynamic loads in a data center according to claim 4, characterized in that: The specific calculation formula of the fitness function described in P3.1 is as follows: In the formula, Representative Energy loss of each storage unit; Representative The charging efficiency of each storage unit; Representative The input energy of each storage unit; Representative The discharge efficiency of each storage unit; Representative Output energy of each storage unit; Representative Energy utilization of each storage unit; Represents the overall energy loss of the flywheel UPS system and; Representative Energy loss of each storage unit; Represents the overall energy utilization of the flywheel UPS system; Representative Energy utilization of each storage unit; Representative The service life of each storage unit; Representative The number of charge and discharge cycles of a storage unit; Representative Energy loss during each charge and discharge process of a storage unit; Represents the overall service life of the flywheel UPS system; Representative The service life of each storage unit; , as well as Both represent the total number of storage units in the flywheel UPS system; Represents the overall fitness value of the flywheel UPS system; represents the weight of energy utilization; represents the weight of the total power loss; The weight representing the useful life of the equipment.

6. The AC access flywheel UPS control method for balancing dynamic loads in a data center according to claim 4, characterized in that: The specific steps of adaptively adjusting the flywheel UPS according to the real-time monitoring data described in S4 are as follows: P4.1: Real-time monitoring of the operating status of the flywheel UPS system and external load changes, extracting the various states of the flywheel UPS system and the corresponding operations under each state from each set of data received, and constructing a state set based on the collected state information and operation information. And action set , then calculate the probability of transferring to the next random state after collecting any action in the action set in each state in the state set; P4.2: All state-action pairs The initial value of is set to 0, where , , collect all state-action pairs and build a corresponding data repository to store each state-action pair and its corresponding value, and then at each moment , according to the current state and the information in the data repository, select Action implement; P4.3: Selecting an action When a random number is generated, if the random number is higher than the preset selection threshold , then from the action set Randomly select an action, otherwise select The action with the largest value, After the selection is completed, the simulated flywheel UPS system executes the action and enters the next state ; P4.4: Calculate the instant reward in the corresponding state after executing any action based on energy efficiency, power loss and equipment life. , use the Q value update formula to update the current state-action pair , and update the data repository, and then repeat the action selection and data repository update until the state-action pair The value changes converge to the preset range; P4.5: After the iteration, traverse the state-action pairs in the data repository value, and select the state-action pair with the highest value as the optimal response strategy, and at the same time, according to the feedback and The value is updated to pre-set the future operation strategy of the flywheel UPS system.

Images