Distributed integrated power supply management system and method

Abstract

The application relates to a kind of integrated power supply management system and method based on distribution, to power supply management field, through uninterruptible power supply platform, uninterruptible power supply array power state parameter array is periodically distributed and collected, power failure prediction is carried out;According to the power state parameter array, uninterruptible power supply time is obtained by calculation, when less than rated uninterrupted power supply time, the number of uninterrupted equipment is obtained by combining the power failure time processing;The equipment characteristic parameter array of equipment array is collected, and according to the number of uninterrupted equipment, equipment selection optimization is carried out, and the optimized selected equipment set is obtained;The optimized selected equipment set is used as an integrated power supply scheme, when power failure occurs within a preset time window, power supply management is carried out according to the integrated power supply scheme.The application achieves the technical effect of improving the collaborative management effect of uninterruptible power supply management system and reasonably allocating power supply capacity.

Description

Technical Field

[0001] This invention relates to the field of power supply management, and in particular to an integrated power supply management system and method based on distributed systems. Background Technology

[0002] UPS power supplies are primarily used to provide a stable and uninterrupted power supply to individual computers, computer network systems, or other power electronic equipment such as solenoid valves and pressure transmitters, playing a crucial role in the stable operation of electronic equipment. However, traditional UPS power supply systems are mostly deployed in an islanded manner, lacking coordinated management of distributed power sources. This leads to unreasonable allocation of power supply capacity during power outages, easily resulting in power loss for critical equipment. Traditional power supply management systems suffer from poor coordinated management and unreasonable power supply capacity allocation. Summary of the Invention

[0003] This invention addresses the technical problems of poor collaborative management and unreasonable power supply capacity allocation in existing power supply management systems. It provides a distributed, integrated power supply management system and method to solve these problems.

[0004] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0005] In a first aspect, the present invention provides a distributed integrated power supply management system, comprising:

[0006] The power outage time prediction module periodically collects the power status parameter array of the uninterruptible power supply array through the uninterruptible power supply platform in a distributed manner, and performs power outage prediction based on the operation logs within a preset historical time range to obtain the power outage time.

[0007] The power supply time calculation module calculates the uninterrupted power supply time based on the power status parameter array. When the uninterrupted power supply time is less than the rated uninterrupted power supply time, the number of uninterrupted devices is obtained by combining the power outage time processing.

[0008] The power supply equipment selection module collects the equipment characteristic parameter array of the equipment array, performs equipment selection optimization based on the number of uninterrupted equipment, and obtains an optimized selection equipment set. Among them, the uninterrupted power supply switching time is optimized based on the minimum allowable interruption time of the equipment array.

[0009] The power supply scheme execution module uses the optimized selection of equipment set as an integrated power supply scheme. When a power outage occurs within a preset time window, the module manages the power supply according to the integrated power supply scheme.

[0010] Optionally, the power status parameters of the uninterruptible power supply (UPS) array are periodically and distributedly collected through the UPS platform, and power outage prediction is performed based on the operation logs within a preset historical time range to obtain the power outage time, including:

[0011] The power status parameter array of the uninterruptible power supply array is periodically and distributedly collected through the uninterruptible power supply platform. The uninterruptible power supply array includes multiple uninterruptible power supplies, and the power status parameters include the power level.

[0012] Obtain the operation log of the uninterruptible power supply array within a preset historical time range. The operation log includes records of unexpected power outages and maintenance power outages within the historical time period.

[0013] Based on the operation log, the power outage time is predicted and the outage duration is obtained.

[0014] The process of predicting the power outage time based on the operation log includes:

[0015] Based on the power supply history data of the uninterruptible power supply, a sample operation log set is collected, and the power outage time of the next power outage after collecting the operation logs of different samples is marked to obtain a sample power outage time set.

[0016] Construct a power outage time prediction network;

[0017] The power outage time prediction network is trained under supervision using the sample running log set and the sample power outage time set, and training is completed after convergence.

[0018] The operation log is input into the power outage time prediction network, which predicts and outputs the power outage time.

[0019] Optionally, based on the power state parameter array, the uninterrupted power supply time is calculated. When it is less than the rated uninterrupted power supply time, the number of uninterrupted devices is obtained by combining the power outage time, including:

[0020] Based on the power state parameter array, the total power state parameters are calculated, and combined with the preset power of a single device, the uninterrupted power supply time is calculated.

[0021] The rated uninterrupted power supply time is calculated based on the power outage time and the number of devices in the equipment array;

[0022] Determine whether the uninterrupted power supply time is less than the rated uninterrupted power supply time. If so, calculate and round down the uninterrupted power supply time and power outage time to obtain the number of uninterrupted devices. If not, use the number of devices in the device array as the number of uninterrupted devices.

[0023] Optionally, an array of device characteristic parameters of the collected device array is used to optimize device selection based on the number of uninterrupted devices, resulting in an optimized set of selected devices, including:

[0024] The device characteristic parameters of each device in the device array are collected to obtain a device characteristic parameter array, wherein the device array is powered by the uninterruptible power supply array, and the device characteristic parameters include data flow.

[0025] Based on the number of uninterrupted devices, randomly select a number of uninterrupted devices from the device array as the first selected device set;

[0026] Based on the first set of device characteristic parameters of the first selected device set, perform uninterrupted power supply selection fitness analysis and calculation to obtain the first selection fitness;

[0027] Obtain the minimum allowable interruption time of the devices in the first selected device set, perform uninterrupted power supply switching time fitness analysis and calculation, and obtain the first time fitness.

[0028] The first power supply management fitness is calculated based on the first selection fitness and the first time fitness.

[0029] The device selection optimization is carried out iteratively until convergence, and the optimized device set with the highest power management fitness is output.

[0030] Specifically, based on the first set of device characteristic parameters of the first selected device set, an uninterrupted power supply selection fitness analysis is performed to obtain the first selection fitness, including:

[0031] The first total equipment characteristic parameters are calculated based on the first set of equipment characteristic parameters of the first selected equipment set.

[0032] Calculate the ratio of the first total equipment characteristic parameters to the preset equipment characteristic parameters to obtain the first selection fitness.

[0033] Specifically, the minimum allowable interruption time of devices within the first selected device set is obtained, and an uninterrupted power supply switching time fitness analysis is performed to obtain a first time fitness, including:

[0034] Obtain the allowed interrupt time for all devices in the first selected device set, filter for the minimum value, and obtain the minimum allowed interrupt time.

[0035] The test obtains the first switching time when the uninterruptible power supply array switches to powering the first selected device set;

[0036] Calculate the ratio of the minimum allowed interrupt time to the first switching time to obtain the first time fitness.

[0037] Secondly, the present invention provides a distributed integrated power supply management method, comprising:

[0038] The power status parameter array is used to calculate the uninterrupted power supply time. When the uninterrupted power supply time is less than the rated uninterrupted power supply time, the number of uninterrupted devices is obtained by combining the power outage time, including:

[0039] Based on the power state parameter array, the total power state parameters are calculated, and combined with the preset power of a single device, the uninterrupted power supply time is calculated.

[0040] The rated uninterrupted power supply time is calculated based on the power outage time and the number of devices in the equipment array;

[0041] Determine whether the uninterrupted power supply time is less than the rated uninterrupted power supply time. If so, calculate and round down the uninterrupted power supply time and power outage time to obtain the number of uninterrupted devices. If not, use the number of devices in the device array as the number of uninterrupted devices.

[0042] In this embodiment of the invention, the power outage time prediction module periodically and distributedly collects the power status parameter array of the uninterruptible power supply array through the uninterruptible power supply platform, and performs power outage prediction based on the operation log within a preset historical time range to obtain the power outage time. This is significantly better than the traditional threshold alarm mechanism and provides accurate input for subsequent power supply strategies.

[0043] In this embodiment of the invention, the power supply time calculation module calculates the uninterrupted power supply time based on the power status parameter array. When the uninterrupted power supply time is less than the rated uninterrupted power supply time, the number of uninterrupted devices is obtained by combining the power outage time processing. This realizes the flexible conversion of power supply strategy based on objective energy storage conditions, which is conducive to the rational use of power supply capacity.

[0044] In this embodiment of the invention, the power supply equipment selection module collects the equipment characteristic parameter array of the equipment array, performs equipment selection optimization based on the number of uninterrupted equipment, obtains an optimized selection equipment set, and quickly filters out the optimal equipment set, ensuring seamless power supply switching and avoiding the risk of equipment downtime.

[0045] In this embodiment of the invention, the power supply scheme execution module uses the optimized selection of equipment set as an integrated power supply scheme. When a power outage occurs within a preset time window, the module manages the power supply according to the integrated power supply scheme, which significantly improves the collaborative management effect of the power supply management system.

[0046] In summary, by implementing this invention, the technical effects of improving the collaborative management effect of the power supply management system and rationally allocating power supply capacity can be achieved. Attached Figure Description

[0047] Figure 1 A schematic diagram of the structure of the distributed integrated power supply management system provided by the present invention;

[0048] Figure 2 This is a flowchart illustrating the distributed integrated power supply management method provided by the present invention.

[0049] In the attached diagram, the components represented by each number are as follows:

[0050] Power outage time prediction module 11, power supply time calculation module 12, power supply equipment selection module 13, power supply scheme execution module 14. Detailed Implementation

[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0053] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0054] Example 1, as Figure 1 As shown, this embodiment of the invention provides a distributed integrated power supply management system, including the following modules:

[0055] Power outage time prediction module 11: It is used to periodically and distributedly collect the power status parameter array of the uninterruptible power supply array through the uninterruptible power supply platform, and perform power outage prediction based on the operation log within a preset historical time range to obtain the power outage time.

[0056] Power supply time calculation module 12: used to calculate the uninterrupted power supply time based on the power status parameter array, and when it is less than the rated uninterrupted power supply time, to obtain the number of uninterrupted devices by combining the power outage time processing.

[0057] Power supply equipment selection module 13: used to collect the equipment characteristic parameter array of the equipment array, and optimize the equipment selection according to the number of uninterrupted equipment to obtain an optimized selection set of equipment, wherein the uninterrupted power supply switching time is optimized according to the minimum allowable interruption time of the equipment array.

[0058] Power supply scheme execution module 14: used to treat the optimized selection of equipment set as an integrated power supply scheme, and to manage power supply according to the integrated power supply scheme when a power outage occurs within a preset time window.

[0059] In this embodiment, the power outage time prediction module 11 uses an uninterruptible power supply (UPS) platform to periodically and distributedly collect the power status parameter array of the UPS array, and performs power outage prediction based on the operation logs within a preset historical time range to obtain the power outage time, including:

[0060] The power status parameter array of the uninterruptible power supply array is periodically and distributedly collected through the uninterruptible power supply platform. The uninterruptible power supply array includes multiple uninterruptible power supplies, and the power status parameters include the power level.

[0061] Obtain the operation log of the uninterruptible power supply array within a preset historical time range. The operation log includes records of unexpected power outages and maintenance power outages within the historical time period.

[0062] Based on the operation log, the power outage time is predicted and the outage duration is obtained.

[0063] This embodiment introduces a distributed integrated power supply management system, primarily applied to the power management of an uninterruptible power supply (UPS) platform. An UPS is a device that continuously supplies power to load equipment when the mains power is abnormal or interrupted. Through built-in batteries and power conversion systems, it ensures the normal operation of critical equipment under conditions of power outages and voltage fluctuations, preventing data loss or hardware damage. Connecting multiple UPSs via circuits forms an UPS array. The UPS platform is a centralized management platform composed of multiple UPSs. It monitors the status of multiple UPSs in the aforementioned UPS array in real time through communication protocols (such as SNMP and Modbus), and supports remote control and data acquisition. The data contained in the power status parameter array includes power status, such as remaining power, which can be measured in watt-hours (Wh). Collecting data such as the remaining power of all UPSs in the UPS array constitutes the power status parameter array.

[0064] For example, the distributed acquisition described in this application embodiment can be achieved by collecting power parameters in real time through sensors and communication modules (such as current transformers and voltage sensors) distributed on multiple UPS nodes, and the data is transmitted to a central server via a network.

[0065] Next, it is necessary to obtain the operation logs of the uninterruptible power supply (UPS) array within a preset historical time range. Then, a computer model (i.e., a power outage time prediction network) is used to analyze the power outage pattern information in the logs to predict the power outage time. The operation logs include records of unexpected power outages and maintenance outages within a historical time period, specifically the start and end times of unexpected power outages and maintenance outages, all accurate to the second. The aforementioned historical time range is the time range from which to extract the records of unexpected and maintenance outages from the operation logs. It is sufficient to ensure that enough feature data of unexpected and maintenance outages can be extracted within this time range to support the subsequent power outage time prediction steps. For example, this historical time can be defined as 3 months, meaning that the operation logs of the UPS array within the past three months need to be extracted to obtain the records of unexpected and maintenance outages within the corresponding time period.

[0066] In this embodiment of the application, the power outage time prediction module 11 predicts the power outage time based on the operation log to obtain the power outage time, including:

[0067] Based on the power supply history data of the uninterruptible power supply, a sample operation log set is collected, and the power outage time of the next power outage after collecting the operation logs of different samples is marked to obtain a sample power outage time set.

[0068] Construct a power outage time prediction network;

[0069] The power outage time prediction network is trained under supervision using the sample running log set and the sample power outage time set, and training is completed after convergence.

[0070] The operation log is input into the power outage time prediction network, which predicts and outputs the power outage time.

[0071] In this embodiment, predicting power outage time requires the use of a computer model. A suitable computer model needs to be selected, a power outage time prediction network needs to be built, and the power outage time prediction network needs to be trained using data from the operation logs over a historical period. By inputting the operation logs into the trained power outage time prediction network, the power outage time can be automatically predicted and output.

[0072] First, a sample set of operation logs needs to be collected. Each sample operation log contains operation log characteristics (including whether there was a power outage, the type of power outage, the start time of the power outage, etc.) within a historical time window (e.g., 24 hours), as well as the duration of the next power outage after the end of that time window (accurate to the second), i.e., the labeled sample power outage time set. The coverage time of the operation logs used for sampling is the aforementioned preset historical time range (e.g., 3 months).

[0073] Therefore, the power outage prediction network includes two input features: the operation log feature and the duration of the next power outage corresponding to that operation log feature. Furthermore, a sliding window strategy can be used to slide samples in 1-hour increments to improve data utilization (e.g., 3 months of logs can generate approximately 2160 samples). Prepare at least 2000 sample data points, dividing them chronologically into a training set (80%) and a validation set (20%) for training the power outage prediction network.

[0074] LightGBM (an ensemble learning model based on gradient boosting decision trees) can be used as the core algorithm for building a power outage time prediction network.

[0075] Regarding specific parameter settings, since the power outage duration needs to be predicted, the prediction target type is set to "regression"; the gradient boosting strategy adopts "Gradient One-Sided Sampling" (GOSS), prioritizing the retention of samples with larger gradients for training; the number of leaf nodes is set to 63 to control the complexity of a single decision tree; the initial learning rate is set to 0.05, and can be gradually reduced later through a decay strategy (e.g., decaying by 10% every 100 rounds) to improve the model's convergence stability; the feature random sampling ratio is 80% of the features randomly selected during the training of each tree to enhance model diversity and prevent over-reliance on a few strong features; the maximum tree depth is not limited (set to -1), allowing the model to automatically adjust according to the data and avoid underfitting due to insufficient depth; each leaf node contains at least 20 samples to suppress noise data interference and reduce the risk of overfitting. The L1 regularization strength is set to 0.1 to reduce the interference of non-critical features by sparser feature weights.

[0076] In training the power outage time prediction network, an early stopping method can be introduced. If the validation set error does not decrease for 10 consecutive rounds, training is terminated. When the MAPE (Mean Absolute Percentage Error) of the validation set is ≤5% or the number of training rounds reaches 1000, training is stopped, the model is judged to have converged, the power outage time prediction network is obtained, and the running log (up to the current time) is input to predict the power outage time (e.g., from 8:00 to 11:00 on the same day).

[0077] In this embodiment of the application, the power supply time calculation module 12 calculates the uninterrupted power supply time based on the power state parameter array. When the uninterrupted power supply time is less than the rated uninterrupted power supply time, the number of uninterrupted devices is obtained by combining the power outage time processing, including:

[0078] Based on the power state parameter array, the total power state parameters are calculated, and combined with the preset power of a single device, the uninterrupted power supply time is calculated.

[0079] The rated uninterrupted power supply time is calculated based on the power outage time and the number of devices in the equipment array;

[0080] Determine whether the uninterrupted power supply time is less than the rated uninterrupted power supply time. If so, calculate and round down the uninterrupted power supply time and power outage time to obtain the number of uninterrupted devices. If not, use the number of devices in the device array as the number of uninterrupted devices.

[0081] In this embodiment, when a power outage occurs, it is necessary to activate the uninterruptible power supply (UPS) array to supply power to the equipment array. First, it is necessary to consider whether the electrical energy stored in the UPS array is sufficient to meet the power needs of the equipment array during the predicted outage period, in order to adjust the power supply strategy accordingly. The equipment mainly consists of servers, and the equipment is connected to the UPS array via circuitry.

[0082] First, it is necessary to calculate the reserve power of the uninterruptible power supply array. Specifically, it is necessary to extract the power status parameters mentioned in S100, namely the remaining power (Wh), count the real-time remaining power of all uninterruptible power supplies in the uninterruptible power supply array, and add them together to obtain the total remaining power of the uninterruptible power supply array (e.g., 4000Wh), which is the total power status parameter.

[0083] Next, by using the preset power of each device in the array, the total remaining power of the uninterruptible power supply (UPS) array can be used to calculate the power supply time for that device. For example, if the preset power of a certain model of device is 40W, and the total remaining power of the UPS array is 400Wh, then the uninterrupted power supply time for a single device is 400Wh / 40W = 10 hours. If there are 5 devices in the array, each with a preset power of 40W, and the power outage time is 3 hours, this is equivalent to supplying power at the rated power of 40W for 5 * 3 = 15 hours, meaning the rated uninterruptible power supply time is 15 hours.

[0084] Determine whether the uninterrupted power supply time is less than the rated uninterrupted power supply time. If so, calculate and round down the uninterrupted power supply time and power outage time to obtain the number of uninterrupted devices. For example, in the above example, when the uninterrupted power supply time is 10 hours, the rated uninterrupted power supply time is 15 hours, the total remaining power of the uninterrupted power supply array is 400Wh, the preset power of a certain model of equipment is 40W, there are 5 equipment with a preset power of 40W, and the power outage time is 3 hours, the calculated and rounded number of uninterrupted devices is 10h / 3h≈3.33 rounded down. Under the conditions of the total remaining power of the uninterrupted power supply array and the power outage time, the maximum number of uninterrupted devices can be maintained at 3. If not, it means that under the conditions of the total remaining power of the uninterrupted power supply array and the power outage time, the maximum number of uninterrupted devices can be maintained at 3. The remaining power is sufficient, so the number of devices in the equipment array is taken as the number of uninterrupted devices.

[0085] In this embodiment of the application, the power supply equipment selection module 13 collects the equipment characteristic parameter array of the equipment array, and performs equipment selection optimization based on the number of uninterrupted devices to obtain an optimized selection set, including:

[0086] The device characteristic parameters of each device in the device array are collected to obtain a device characteristic parameter array, wherein the device array is powered by the uninterruptible power supply array, and the device characteristic parameters include data flow.

[0087] Based on the number of uninterrupted devices, randomly select a number of uninterrupted devices from the device array as the first selected device set;

[0088] Based on the first set of device characteristic parameters of the first selected device set, perform uninterrupted power supply selection fitness analysis and calculation to obtain the first selection fitness;

[0089] Obtain the minimum allowable interruption time of the devices in the first selected device set, perform uninterrupted power supply switching time fitness analysis and calculation, and obtain the first time fitness.

[0090] The first power supply management fitness is calculated based on the first selection fitness and the first time fitness.

[0091] The device selection optimization is carried out iteratively until convergence, and the optimized device set with the highest power management fitness is output.

[0092] In this embodiment, if the power supply time calculation module 12 calculates that the total remaining power of the uninterruptible power supply array is insufficient to maintain the operation of all devices in the equipment array during the predicted power outage time, it is necessary to make a selection based on the importance of different devices, and only maintain power supply to a portion of the devices (i.e., the number of devices calculated as uninterruptible devices). The specific selection of which devices in the equipment array to maintain power supply requires consideration of the device characteristic parameters of each device in the equipment array. The device characteristic parameters are mainly the data traffic transmitted and received by the device over a historical period. For example, the device characteristic parameter of a device can be the data traffic transmitted and received by the device in the past day, such as 1GB.

[0093] Then, based on the number of uninterrupted devices calculated by the power supply time calculation module 12 (e.g., 3 devices), the number of uninterrupted devices is randomly selected from the device array (e.g., a total of 5 devices) as the first selected device set.

[0094] In the power supply equipment selection module 13 of this application embodiment, an uninterrupted power supply selection fitness analysis calculation is performed based on the first equipment characteristic parameter set of the first selection equipment set to obtain the first selection fitness, including:

[0095] The first total equipment characteristic parameters are calculated based on the first set of equipment characteristic parameters of the first selected equipment set.

[0096] Calculate the ratio of the first total equipment characteristic parameters to the preset equipment characteristic parameters to obtain the first selection fitness.

[0097] In this embodiment, the first selected device set includes multiple randomly selected uninterrupted device numbers (e.g., 3 devices) obtained through the above steps; and the first device characteristic parameter set of the first selected device set is the data traffic (e.g., 1GB, 2GB, 3GB) sent and received by these 3 devices within a historical time period (e.g., within the last day). The first total device characteristic parameter is the sum of the characteristic parameters in the first device characteristic parameter set of the first selected device set. In the above example, the first total device characteristic parameter is the sum of the data traffic sent and received by these 3 devices within the historical time period, which is 1GB + 2GB + 3GB = 6GB. The preset device characteristic parameter can be the average data traffic sent and received by all devices within the historical time period (e.g., 1.5GB) multiplied by the specific number of uninterrupted devices (e.g., 3). In this case, the preset device characteristic parameter = 1.5GB * 3 = 4.5GB.

[0098] The ratio of the first total device characteristic parameter (e.g., 6GB) to the preset device characteristic parameter (e.g., 4.5GB) is 6GB / 4.5GB≈1.33, and this value is the first selection fitness of the first device characteristic parameter set.

[0099] In the power supply equipment selection module 13 of this application embodiment, the minimum allowable interruption time of the equipment in the first selected equipment set is obtained, and an uninterrupted power supply switching time fitness analysis calculation is performed to obtain a first time fitness, including:

[0100] Obtain the allowed interrupt time for all devices in the first selected device set, filter for the minimum value, and obtain the minimum allowed interrupt time.

[0101] The test obtains the first switching time when the uninterruptible power supply array switches to powering the first selected device set;

[0102] Calculate the ratio of the minimum allowed interrupt time to the first switching time to obtain the first time fitness.

[0103] In this embodiment, different devices have different tolerances for power outage durations. For example, some devices have built-in backup batteries that can maintain power supply for a period of time after the external power is cut off, thus allowing for a longer power outage. Conversely, some devices would lose all data being processed if suddenly powered off, requiring a shorter power outage. The allowable interruption time is defined as the longest power outage that does not affect the normal operation of the device and data recovery. Based on this, the allowable interruption time is evaluated for all devices in the first selected device set, and the minimum value is selected to obtain the minimum allowable interruption time, such as 1 second.

[0104] Next, it is necessary to calculate the first switching time for the uninterruptible power supply array to switch to powering the first selected device set. Since the uninterruptible power supply array may currently be powering other devices (devices other than the first selected device set), it takes time to switch to powering the devices in the first selected device set. This time is the first switching time (e.g., 500ms).

[0105] The first time fitness is the ratio of the minimum allowed interrupt time to the first switching time. For example, by substituting the minimum allowed interrupt time and the first switching time parameters in the example above, we can calculate the exemplary first time fitness = 1s / 500ms = 2.

[0106] Finally, a first power management fitness is calculated based on the first selection fitness (e.g., 1.33) and the first time fitness (e.g., 2). This first power management fitness is used to measure the appropriateness of the first selected device set and serves as the basis for selecting the device set. Optionally, the first power management fitness can be a weighted sum of the first selection fitness and the first time fitness. That is, based on the different emphases on the data flow and switching time during actual operation, different weights are assigned to the first selection fitness and the first time fitness (e.g., first selection fitness weight 0.3 and first time fitness weight 0.7), and then the sums are added. In one embodiment, the first power management fitness = first selection fitness (e.g., 1.33) * first selection fitness weight + first time fitness (e.g., 2) * first time fitness weight = 1.33 * 0.3 + 2 * 0.7 = 1.799. The larger this value, the more reasonable it is to select the first selected device set as the power supply equipment.

[0107] In this embodiment, after clarifying the calculation method of the first power management fitness, it is necessary to continuously and randomly select a first set of selected devices with an uninterrupted number of devices from the device array, and calculate the first time fitness of the first set of selected devices. When all combinations of the first set of selected devices with an uninterrupted number of devices randomly selected from the device array are exhausted, the first power management fitness of all selected first set of selected devices is compared, and the first set of selected devices with the maximum first power management fitness is selected.

[0108] In this embodiment, the power supply scheme execution module 14 is used to treat the optimized selection device set as an integrated power supply scheme, and to manage power supply according to the integrated power supply scheme when a power outage occurs within a preset time window.

[0109] The optimized selection set of devices refers to the first set of devices selected when the first power supply management fitness is maximized. When a power outage event does occur within the predicted outage time (e.g., from 8:00 to 11:00 on the same day) in the power outage time prediction module 11, power is supplied only to the optimized selection set of devices to maximize the power supply effect.

[0110] Example 2, as Figure 2 As shown, based on the same inventive concept as the distributed integrated power supply management system provided in Embodiment 1, this embodiment of the invention also provides a distributed integrated power supply management method, including:

[0111] S100: Through the uninterruptible power supply platform, the power status parameters of the uninterruptible power supply array are collected periodically in a distributed manner, and the power outage prediction is performed based on the operation logs within a preset historical time range to obtain the power outage time.

[0112] S200: Calculate the uninterrupted power supply time based on the power status parameter array. If the uninterrupted power supply time is less than the rated uninterrupted power supply time, combine the power outage time processing to obtain the number of uninterrupted devices.

[0113] S300: Collect the array of device characteristic parameters of the device array, optimize the device selection based on the number of uninterrupted devices, and obtain an optimized selection set of devices, wherein the uninterrupted power supply switching time is optimized based on the minimum allowable interruption time of the device array.

[0114] S400: The optimized selection of equipment set is used as an integrated power supply scheme. When a power outage occurs within a preset time window, power supply management is carried out in accordance with the integrated power supply scheme.

[0115] Furthermore, in step S100 of this embodiment, the power status parameter array of the uninterruptible power supply (UPS) array is periodically and distributedly collected through the UPS platform, and power outage prediction is performed based on the operation logs within a preset historical time range to obtain the power outage time, including:

[0116] The power status parameter array of the uninterruptible power supply array is periodically and distributedly collected through the uninterruptible power supply platform. The uninterruptible power supply array includes multiple uninterruptible power supplies, and the power status parameters include the power level.

[0117] Obtain the operation log of the uninterruptible power supply array within a preset historical time range. The operation log includes records of unexpected power outages and maintenance power outages within the historical time period.

[0118] Based on the operation log, the power outage time is predicted and the outage duration is obtained.

[0119] The process of predicting the power outage time based on the operation log includes:

[0120] Based on the power supply history data of the uninterruptible power supply, a sample operation log set is collected, and the power outage time of the next power outage after collecting the operation logs of different samples is marked to obtain a sample power outage time set.

[0121] Construct a power outage time prediction network;

[0122] The power outage time prediction network is trained under supervision using the sample running log set and the sample power outage time set, and training is completed after convergence.

[0123] The operation log is input into the power outage time prediction network, which predicts and outputs the power outage time.

[0124] Furthermore, in step S200 of this embodiment, the uninterrupted power supply time is calculated based on the power state parameter array. When the uninterrupted power supply time is less than the rated uninterrupted power supply time, the number of uninterrupted devices is obtained by combining the power outage time processing, including:

[0125] Based on the power state parameter array, the total power state parameters are calculated, and combined with the preset power of a single device, the uninterrupted power supply time is calculated.

[0126] The rated uninterrupted power supply time is calculated based on the power outage time and the number of devices in the equipment array;

[0127] Determine whether the uninterrupted power supply time is less than the rated uninterrupted power supply time. If so, calculate and round down the uninterrupted power supply time and power outage time to obtain the number of uninterrupted devices. If not, use the number of devices in the device array as the number of uninterrupted devices.

[0128] Furthermore, in step S300 of this embodiment, the device characteristic parameter array of the acquired device array is collected, and the device selection optimization is performed based on the number of uninterrupted devices to obtain an optimized selection device set, including:

[0129] The device characteristic parameters of each device in the device array are collected to obtain a device characteristic parameter array, wherein the device array is powered by the uninterruptible power supply array, and the device characteristic parameters include data flow.

[0130] Based on the number of uninterrupted devices, randomly select a number of uninterrupted devices from the device array as the first selected device set;

[0131] Based on the first set of device characteristic parameters of the first selected device set, perform uninterrupted power supply selection fitness analysis and calculation to obtain the first selection fitness;

[0132] Obtain the minimum allowable interruption time of the devices in the first selected device set, perform uninterrupted power supply switching time fitness analysis and calculation, and obtain the first time fitness.

[0133] The first power supply management fitness is calculated based on the first selection fitness and the first time fitness.

[0134] The device selection optimization is carried out iteratively until convergence, and the optimized device set with the highest power management fitness is output.

[0135] Specifically, based on the first set of device characteristic parameters of the first selected device set, an uninterrupted power supply selection fitness analysis is performed to obtain the first selection fitness, including:

[0136] The first total equipment characteristic parameters are calculated based on the first set of equipment characteristic parameters of the first selected equipment set.

[0137] Calculate the ratio of the first total equipment characteristic parameters to the preset equipment characteristic parameters to obtain the first selection fitness.

[0138] Specifically, the minimum allowable interruption time of devices within the first selected device set is obtained, and an uninterrupted power supply switching time fitness analysis is performed to obtain a first time fitness, including:

[0139] Obtain the allowed interrupt time for all devices in the first selected device set, filter for the minimum value, and obtain the minimum allowed interrupt time.

[0140] The test obtains the first switching time when the uninterruptible power supply array switches to powering the first selected device set;

[0141] Calculate the ratio of the minimum allowed interrupt time to the first switching time to obtain the first time fitness.

[0142] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0143] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0144] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0145] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0146] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0147] Although preferred embodiments of the invention have been described, those skilled in the art, once they have learned the basic inventive concept, can make other changes and modifications to these embodiments.

[0148] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A distributed integrated power supply management system, characterized by, The system comprises: The power outage time prediction module periodically and distributively collects the power state parameter array of the uninterruptible power supply array through the uninterruptible power supply platform, and performs power outage prediction according to the operation log within a preset historical time range to obtain the power outage time; The power supply time calculation module calculates the uninterruptible power supply time according to the power state parameter array, and obtains the number of uninterruptible devices when the uninterruptible power supply time is less than the rated uninterruptible power supply time, including: calculating the total power state parameter according to the power state parameter array, and calculating the uninterruptible power supply time according to the preset power of a single device; According to the power outage time and the number of devices in the device array, the rated uninterruptible power supply time is calculated and obtained; If the uninterruptible power supply time is less than the rated uninterruptible power supply time, the number of uninterruptible devices is calculated and obtained according to the uninterruptible power supply time and the power outage time, otherwise, the number of devices in the device array is taken as the number of uninterruptible devices; The power supply device selection module collects the device characteristic parameter array of the device array, and performs device selection optimization according to the number of uninterruptible devices to obtain an optimized selection device set, including: Collecting the device characteristic parameters of each device in the device array to obtain a device characteristic parameter array, wherein the device array is powered by the uninterruptible power supply array, and the device characteristic parameters include data traffic; According to the number of uninterruptible devices, a number of devices equal to the number of uninterruptible devices are randomly selected from the device array as a first selection device set; According to the first device characteristic parameter set of the first selection device set, the first selection fitness is obtained by performing uninterruptible power supply selection fitness analysis calculation; The minimum allowed interruption time of the devices in the first selection device set is obtained, and the first time fitness is obtained by performing uninterruptible power supply switching time fitness analysis calculation; According to the first selection fitness and the first time fitness, the first power supply management fitness is calculated and obtained; Iterative device random selection optimization is performed until convergence, and the optimized selection device set with the maximum power supply management fitness is output; wherein the uninterruptible power supply switching time optimization is performed according to the minimum allowed interruption time of the device array; The power supply scheme execution module takes the optimized selection device set as an integrated power supply scheme, and performs power supply management according to the integrated power supply scheme when a power outage occurs within a preset time window.

2. The distributed-based integrated power supply management system of claim 1, wherein, The uninterruptible power supply platform periodically and distributively collects the power state parameter array of the uninterruptible power supply array, and performs power outage prediction according to the operation log within a preset historical time range to obtain the power outage time, including: The uninterruptible power supply platform periodically and distributively collects the power state parameter array of the uninterruptible power supply array, wherein the uninterruptible power supply array includes a plurality of uninterruptible power supplies, and the power state parameters include power; The operation log of the uninterruptible power supply array within a preset historical time range is obtained, wherein the operation log includes the record data of accidental power outage and maintenance power outage within the historical time; According to the operation log, the power outage time is predicted and obtained.

3. The distributed-based integrated power supply management system of claim 2, wherein, According to the operation log, the power outage time prediction is performed to obtain a power outage time, including: According to the uninterrupted power supply history data, a sample operation log set is collected, and a sample power outage time set is obtained by labeling the power outage time after different sample operation logs; A power outage time prediction network is constructed; The sample operation log set and the sample power outage time set are used to supervise the training of the power outage time prediction network, and the training is completed after convergence; The operation log is input into the power outage time prediction network to predict the output power outage time.

4. The distributed-based integrated power supply management system of claim 1, wherein, According to the first device characteristic parameter set of the first selected device set, uninterrupted power supply selection fitness analysis calculation is performed to obtain a first selection fitness, including: According to the first device characteristic parameter set of the first selected device set, a first total device characteristic parameter is calculated and obtained; The ratio of the first total device characteristic parameter to the preset device characteristic parameter is calculated to obtain the first selection fitness.

5. The distributed-based integrated power supply management system of claim 4, wherein, The minimum allowable interruption time of the devices in the first selected device set is obtained, and uninterrupted power supply switching time fitness analysis calculation is performed to obtain a first time fitness, including: The minimum allowable interruption time is obtained by screening the minimum value of the allowable interruption time of all devices in the first selected device set; The first switching time is tested and obtained when the uninterrupted power supply array switches to supply power to the first selected device set; The ratio of the minimum allowable interruption time to the first switching time is calculated to obtain the first time fitness.

6. A method for managing power supply based on distributed integrated power supply, characterized in that, The method comprises: Through the uninterrupted power supply platform, the power supply state parameter array of the uninterrupted power supply array is periodically and distributedly collected, and the power outage time is obtained by performing power outage prediction according to the operation log in a preset historical time range; According to the power supply state parameter array, the uninterrupted power supply time is calculated and obtained, and when the uninterrupted power supply time is less than the rated uninterrupted power supply time, the uninterrupted device quantity is obtained by combining the power outage time, including: according to the power supply state parameter array, the total power supply state parameter is calculated and obtained, and the uninterrupted power supply time is calculated and obtained by combining the preset power of a single device; According to the power outage time and the device quantity of the device array, the rated uninterrupted power supply time is calculated and obtained; If the uninterrupted power supply time is less than the rated uninterrupted power supply time, the uninterrupted device quantity is calculated and obtained by rounding according to the uninterrupted power supply time and the power outage time, and if not, the device quantity of the device array is taken as the uninterrupted device quantity; The device characteristic parameter array of the device array is collected, and the optimized selected device set is obtained by performing device selection optimization according to the uninterrupted device quantity, including: The device characteristic parameter array of each device in the device array is collected, wherein the device array is powered by the uninterrupted power supply array, and the device characteristic parameter includes data flow; According to the uninterrupted device quantity, a device set of the uninterrupted device quantity is randomly selected from the device array as a first selected device set; According to the first device characteristic parameter set of the first selected device set, uninterrupted power supply selection fitness analysis calculation is performed to obtain a first selection fitness; acquire the minimum allowed interruption time of the devices in the first selected device set, perform uninterrupted power supply switching time fitness analysis calculation to obtain a first time fitness; calculate a first power supply management fitness according to the first selected fitness and the first time fitness; perform iterative device random selection optimization until convergence, and output an optimized selected device set with the largest power supply management fitness; wherein uninterrupted power supply switching time optimization is performed according to the minimum allowed interruption time of the device array; use the optimized selected device set as an integrated power supply scheme, and perform power supply management according to the integrated power supply scheme when a power outage occurs within a preset time window.

Images