Infrastructure management method and system based on data center

Abstract

The invention relates to the technical field of building data center infrastructure management, and discloses an infrastructure management method and system based on a data center. The method comprises the steps of collecting data of a building cooling and lighting system through a sensor, performing edge equipment anomaly detection, performing hybrid model-based energy consumption prediction, performing system optimization and performing feedback monitoring. Compared with the technical problems in the prior art that dynamic optimization cannot be realized in a complex building environment, and especially energy efficiency cannot be ensured difficultly under the condition of large demand fluctuation or high load, the energy efficiency of the cooling and lighting system is improved through edge calculation and a multi-objective optimization algorithm, energy consumption waste is avoided, and the energy efficiency of the cooling and lighting system is improved. And the stability and the response capability of the system are obviously improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of building data center infrastructure management, and in particular, relates to a data center-based infrastructure management method and system. Background Art

[0002] At present, there are significant deficiencies in the energy efficiency optimization of building data center infrastructure management under dynamic loads and complex environments. For example, in the case of large load fluctuations or multiple systems running in parallel, traditional management methods are difficult to dynamically adapt to environmental changes, resulting in increased energy consumption of cooling and lighting systems, and the inability to achieve real-time and efficient energy efficiency management. Existing technologies fail to meet the needs of building data centers for high energy efficiency, stability and flexible adaptation. Therefore, there is an urgent need for a method that can achieve real-time monitoring, adaptive adjustment and multi-objective optimization of cooling and lighting systems under complex conditions to improve the intelligence, response speed and energy saving effect of data center infrastructure management. Summary of the invention

[0003] In view of the above-mentioned technical deficiencies, the purpose of the present invention is to propose an infrastructure management method and system based on a data center, aiming to solve the technical problem that the existing technology cannot achieve dynamic optimization in a complex building environment, especially it is difficult to ensure energy efficiency under conditions of large demand fluctuations or high load.

[0004] In order to solve the above technical problems, the present invention adopts the following technical solution: The present invention provides an infrastructure management method based on a data center, comprising: Step S10: Install a temperature sensor and a humidity sensor on the building cooling system, and install a current sensor and an ambient light sensor on the lighting system, collect temperature information T and humidity information H of the cooling system, and current information I and ambient light information L of the lighting system in real time, and transmit the temperature information T, humidity information H, current information I and ambient light information L to an edge device including an embedded system for processing; Step S20: Introduce a clustering algorithm in the edge device containing the embedded system to perform anomaly detection and automatically identify device failures. The anomaly detection formula is:

[0005] in, is the current input data, including temperature information T, humidity information H, current information I and ambient light information L; is the mathematical expectation, is the standard deviation corresponding to the current input data; When the anomaly score is higher than the preset threshold, an alarm is triggered and the sampling frequency is increased; The current input data with anomaly scores not higher than the preset threshold After pre-processing, it is transmitted to the industrial gateway for centralized processing; Step S30: After receiving the pre-processed temperature information T, humidity information H, current information I, and ambient light information L, the industrial gateway inputs them into a hybrid model of a convolutional neural network CNN and a long short-term memory network LSTM. CNN extracts temperature distribution feature data, humidity gradient feature data, current feature data, and ambient light feature data, and combines the time series characteristics of the LSTM analysis data to predict the energy consumption demand in the next hour. Combined with the temperature information T and humidity information H, the dynamic energy efficiency parameters of the cooling system are generated. , combining current information I and ambient light information L to generate dynamic energy efficiency parameters of the lighting system , where the loss function selected by the hybrid model is:

[0006] in, is the loss function of the hybrid model, which is used to measure the error between the prediction and the true value. The weight parameter is used to balance short-term and long-term accuracy. and is the mean square error of short-term and long-term forecasts; Step S40: Using the dynamic energy efficiency parameters of the cooling system generated in step S30 , combined with the Pareto optimal algorithm, the cooling system is optimized for multiple objectives to meet the temperature requirements of each area of ​​the building under the premise of minimizing energy consumption. The optimization objective function is defined as:

[0007] in, is the objective function, represents the energy consumption of the cooling system, is the balance coefficient between energy efficiency and temperature control, Indicates the deviation between the actual temperature and the set temperature. is the deviation weight factor of the dynamic energy efficiency parameter; Step S50: The lighting system calculates the energy efficiency of the lighting system according to the ambient light information L and the dynamic energy efficiency parameters of the lighting system. The traffic data U collected in real time by the pedestrian flow sensor is used to dynamically adjust the lighting brightness. The adaptive dimming model is used. The formula is as follows:

[0008] in, Indicates the adjusted lighting brightness. is the basic lighting brightness; is the target energy efficiency parameter, i.e., the dynamic energy efficiency parameter of the lighting system in step S30; is the current lighting energy consumption; and is the weight factor; Step S60: Loop through steps S10 to S50, and continuously monitor the dynamic energy efficiency parameters of the cooling system and the lighting system. When the dynamic energy efficiency parameters of the cooling system exceed the preset range or the dynamic energy efficiency parameters of the lighting system exceed the preset range, the industrial gateway automatically pushes abnormal information to the front end and issues an alarm.

[0009] Preferably, in step S10, the sensor will periodically upload its own status information to the industrial gateway, and the sensor includes a temperature sensor, a humidity sensor, a current sensor and an ambient light sensor.

[0010] Preferably, in step S20, the preset threshold is adjusted according to the ambient temperature.

[0011] Preferably, in step S30, the energy consumption demand in the next hour is predicted, including the energy consumption demand of the cooling system, using the formula:

[0012] in, is the predicted energy consumption demand of the cooling system, n is the number of first data samples acquired in advance, i represents the i-th first data sample, is the comprehensive weight coefficient related to the temperature distribution characteristic data and the temperature distribution characteristic data, which is used to adjust The value of is in a reasonable range. is the temperature distribution characteristic data corresponding to the i-th first data sample, is the humidity gradient characteristic data corresponding to the i-th first data sample; The energy demand forecast for the next hour also includes the energy demand forecast for the lighting system, using the formula:

[0013] in, To predict the energy demand of the lighting system, is the number of pre-acquired second data samples, Representative The second data sample, is the comprehensive weight coefficient related to the ambient light characteristic data and the current characteristic data, which is used to adjust The value of is in a reasonable range. For the Ambient light characteristic data corresponding to the second data sample, For the The current characteristic data corresponding to the second data samples.

[0014] Preferably, in step S40, the multi-objective optimization process is reinitialized every 12 hours.

[0015] Preferably, the frequency of collecting traffic data U is dynamically adjusted according to the usage of the building.

[0016] Preferably, step S60 also includes automatically generating a monitoring report once a day for the administrator to review.

[0017] The present invention also provides an infrastructure management system based on a data center, comprising: A data acquisition module is used to install a temperature sensor and a humidity sensor on the building cooling system, and to install a current sensor and an ambient light sensor on the lighting system, to collect temperature information T and humidity information H of the cooling system, and current information I and ambient light information L of the lighting system in real time, and to transmit the temperature information T, humidity information H, current information I and ambient light information L to an edge device including an embedded system for processing; The anomaly detection module is used to introduce clustering algorithms into edge devices containing embedded systems for anomaly detection and automatic identification of device failures. The anomaly detection formula uses:

[0018] in, is the current input data, including temperature information T, humidity information H, current information I and ambient light information L; is the mathematical expectation, is the standard deviation corresponding to the current input data; When the anomaly score is higher than the preset threshold, an alarm is triggered and the sampling frequency is increased; The current input data with anomaly scores not higher than the preset threshold After pre-processing, it is transmitted to the industrial gateway for centralized processing; The prediction and analysis module is used for the industrial gateway to receive the pre-processed temperature information T, humidity information H, current information I, and ambient light information L, and then input them into the hybrid model of the convolutional neural network CNN and the long short-term memory network LSTM. CNN extracts the temperature distribution characteristic data, humidity gradient characteristic data, current characteristic data, and ambient light characteristic data, and combines the time series characteristics of the LSTM analysis data to predict the energy consumption demand in the next hour, and combines the temperature information T and humidity information H to generate the dynamic energy efficiency parameters of the cooling system , combining current information I and ambient light information L to generate dynamic energy efficiency parameters of the lighting system , where the loss function selected by the hybrid model is:

[0019] in, is the loss function of the hybrid model, which is used to measure the error between the prediction and the true value. The weight parameter is used to balance short-term and long-term accuracy. and is the mean square error of short-term and long-term forecasts; A cooling optimization module is used to utilize the dynamic energy efficiency parameters of the cooling system generated in step S30 , combined with the Pareto optimal algorithm, the cooling system is optimized for multiple objectives to meet the temperature requirements of each area of ​​the building under the premise of minimizing energy consumption. The optimization objective function is defined as:

[0020] in, is the objective function, represents the energy consumption of the cooling system, is the balance coefficient between energy efficiency and temperature control, Indicates the deviation between the actual temperature and the set temperature. is the deviation weight factor of the dynamic energy efficiency parameter; Lighting adjustment module, used for lighting system according to ambient light information L and lighting system dynamic energy efficiency parameters The traffic data U collected in real time by the pedestrian flow sensor is used to dynamically adjust the lighting brightness. The adaptive dimming model is used. The formula is as follows:

[0021] in, Indicates the adjusted lighting brightness. is the basic lighting brightness; is the target energy efficiency parameter, i.e., the dynamic energy efficiency parameter of the lighting system in step S30; is the current lighting energy consumption; and is the weight factor; The monitoring feedback module is used to execute steps S10 to S50 in a loop and continuously monitor the dynamic energy efficiency parameters of the cooling system and the dynamic energy efficiency parameters of the lighting system. When the dynamic energy efficiency parameters of the cooling system exceed the preset range or the dynamic energy efficiency parameters of the lighting system exceed the preset range, the industrial gateway automatically pushes abnormal information to the front end and issues an alarm.

[0022] The beneficial effect of the present invention is that compared with the prior art which cannot achieve dynamic optimization in complex building environments, especially the technical problem of being unable to ensure energy efficiency when demand fluctuates greatly or the load is high, the present application improves the energy efficiency of the cooling and lighting systems through edge computing and multi-objective optimization algorithms, avoids energy waste, and significantly improves the stability and responsiveness of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0024] Figure 1 A schematic diagram of a flow chart of a first embodiment of a data center-based infrastructure management method provided by the present invention.

[0025] Figure 2 A system schematic diagram of a data center-based infrastructure management method provided by the present invention.

[0026] Figure 3 A schematic diagram of a device for a data center-based infrastructure management method provided by the present invention. DETAILED DESCRIPTION

[0027] 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.

[0028] Embodiment 1: Figure 1 , which is a flow chart of a first embodiment of an infrastructure management method based on a data center according to the present invention, and proposes a first embodiment of an infrastructure management method based on a data center according to the present invention.

[0029] In a first embodiment, the data center-based infrastructure management method includes: Step S10: Install a temperature sensor and a humidity sensor on the building cooling system, and install a current sensor and an ambient light sensor on the lighting system, collect temperature information T and humidity information H of the cooling system, and current information I and ambient light information L of the lighting system in real time, and transmit the temperature information T, humidity information H, current information I and ambient light information L to an edge device including an embedded system for processing; Specifically, the collected information can be stored in a matrix format for subsequent batch processing and analysis. Various sensor data such as temperature information T, humidity information H, current information I, and ambient light information L can be organized into a matrix according to timestamps and sensor locations, for example:

[0030] In the formula, n represents the number of samples collected or the length of the time series. , , ..., Represents the temperature information collected at different time points. , , ..., Represents the humidity information collected at different time points. , , ..., Represents the current information of the lighting system at different time points, , , ..., Represents ambient light information at different time points.

[0031] It should be understood that through this matrix format, edge devices can more effectively apply anomaly detection algorithms to achieve dynamic optimization of building cooling and lighting systems, further improving the energy efficiency management level of data centers.

[0032] Step S20: Introduce a clustering algorithm in the edge device containing the embedded system to perform anomaly detection and automatically identify device failures. The anomaly detection formula is:

[0033] in, is the current input data, including temperature information T, humidity information H, current information I and ambient light information L; is the mathematical expectation, is the standard deviation corresponding to the current input data; When the anomaly score is higher than the preset threshold, an alarm is triggered and the sampling frequency is increased; The current input data with anomaly scores not higher than the preset threshold After pre-processing, it is transmitted to the industrial gateway for centralized processing; Specifically, the anomaly detection formula measures whether the data is outside the normal range by calculating the degree of deviation of the current input data from its expected value. This degree of deviation is standardized by the standard deviation, so that different types of data can be compared on the same scale. When the anomaly score exceeds the preset threshold, the system will automatically trigger an alarm and increase the sampling frequency to quickly capture abnormal situations.

[0034] It should be understood that the anomaly detection method is designed to improve the detection accuracy and response speed of the system, thereby more effectively managing the building infrastructure of the data center.

[0035] Step S30: After receiving the pre-processed temperature information T, humidity information H, current information I, and ambient light information L, the industrial gateway inputs them into a hybrid model of a convolutional neural network CNN and a long short-term memory network LSTM. CNN extracts temperature distribution feature data, humidity gradient feature data, current feature data, and ambient light feature data, and combines the time series characteristics of the LSTM analysis data to predict the energy consumption demand in the next hour. Combined with the temperature information T and humidity information H, the dynamic energy efficiency parameters of the cooling system are generated. , combining current information I and ambient light information L to generate dynamic energy efficiency parameters of the lighting system , where the loss function selected by the hybrid model is:

[0036] in, is the loss function of the hybrid model, which is used to measure the error between the prediction and the true value. The weight parameter is used to balance short-term and long-term accuracy. and is the mean square error of short-term and long-term forecasts; It should be understood that the loss function of the hybrid model optimizes the model by combining the errors of short-term and long-term predictions to achieve a balance between short-term response and long-term trend prediction. This design can improve the accuracy of energy demand prediction and adapt to the dynamic changes of cooling and lighting systems. CNN is used to extract spatial features such as temperature distribution and ambient light intensity, while LSTM is used to capture time series features such as the time dependency of energy consumption trends. The combination of the two improves the prediction ability of the model.

[0037] It should be understood that in order to more accurately generate dynamic energy efficiency parameters for cooling and lighting systems and , the specific process of the model in predicting the energy consumption demand in the next hour can be: The collected temperature information T, humidity information H, current information I and ambient light information L are input into the hybrid model, and the convolutional neural network (CNN) is used to extract spatial features, such as the temperature distribution pattern, humidity gradient information, ambient light color distribution and brightness changes. Through this process, the model can obtain feature information related to spatial distribution. Then, the time series data is processed through the long short-term memory network (LSTM) to capture the trend of data changes over time and predict the trend of future energy consumption, thereby providing time dependency for the subsequent generation of dynamic energy efficiency parameters.

[0038] It should be understood that the model integrates the obtained spatial features and temporal features to calculate the energy consumption demand of the cooling system in the next hour and the energy consumption demand of the lighting system in the next hour. The specific calculation method is: The energy demand for the next hour is predicted, including the energy demand for the cooling system, using the formula:

[0039] in, is the predicted energy consumption demand of the cooling system, n is the number of first data samples acquired in advance, i represents the i-th first data sample, is the comprehensive weight coefficient related to the temperature distribution characteristic data and the temperature distribution characteristic data, which is used to adjust The value of is in a reasonable range. is the temperature distribution characteristic data corresponding to the i-th first data sample, is the humidity gradient characteristic data corresponding to the i-th first data sample; The energy demand forecast for the next hour also includes the energy demand forecast for the lighting system, using the formula:

[0040] in, To predict the energy demand of the lighting system, is the number of pre-acquired second data samples, Representative The second data sample, is the comprehensive weight coefficient related to the ambient light characteristic data and the current characteristic data, which is used to adjust The value of is in a reasonable range. For the Ambient light characteristic data corresponding to the second data sample, For the The current characteristic data corresponding to the second data samples.

[0041] Specifically, the first data sample and the second data sample are randomly selected to ensure the accuracy of calculation, and multiple groups of data samples can also be randomly selected to calculate and take the average value.

[0042] It should be understood that in order to better generate the dynamic energy efficiency parameters of the cooling system And lighting system dynamic energy efficiency parameters , the specific formula is:

[0043] in, is the energy demand of the cooling system in the next hour, The energy consumption required for the lighting system in the next hour. is the current temperature of the cooling system, indicating the internal temperature of the building, is the current humidity percentage (unit: %), indicating the humidity level in the air inside the building. is the ambient light illumination, which indicates the brightness level of the ambient light in the building. is the current of the lighting system, indicating the real-time current consumption of the lighting system, is the temperature weight coefficient, which is used to balance the effect of temperature on the energy efficiency of the cooling system. is the humidity weight coefficient, which is used to balance the effect of humidity on the energy efficiency of the cooling system. is the illuminance weight coefficient, which is used to balance the impact of ambient light on the energy efficiency of the lighting system. is the current weight coefficient, which is used to balance the impact of current on the energy efficiency of the lighting system; These weighting factors can be adjusted according to the actual needs and environmental conditions of the building, thereby improving the energy efficiency of the cooling and lighting systems under different environmental conditions.

[0044] Step S40: Using the dynamic energy efficiency parameters of the cooling system generated in step S30 , combined with the Pareto optimal algorithm, the cooling system is optimized for multiple objectives to meet the temperature requirements of each area of ​​the building under the premise of minimizing energy consumption. The optimization objective function is defined as:

[0045] in, is the objective function, represents the energy consumption of the cooling system, is the balance coefficient between energy efficiency and temperature control, Indicates the deviation between the actual temperature and the set temperature. is the deviation weight factor of the dynamic energy efficiency parameter; You should understand that the function Indicates the current energy consumption of the cooling system. Adjusting this item can directly affect the energy efficiency of the system. The coefficient is used to balance the priority between energy efficiency and temperature control. It represents the deviation of the actual temperature from the set temperature, which enables the system to respond quickly to temperature fluctuations. The project is used to control the deviation of cooling energy consumption from the target energy efficiency parameters, so that the system is closer to the dynamic energy efficiency target.

[0046] For example, suppose you have the following data:

[0047] At 2 p.m., the system collected a temperature of 27°C, an ambient light of 700 lux, and a low current demand. As time went on, at 3 p.m., the temperature rose to 29°C due to increased equipment load. The system automatically increased cooling efforts and triggered cooling energy efficiency optimization to control the temperature around the target temperature of 26°C. At 4 p.m., due to reduced external light, the lighting system current demand rose to 4A, and the ambient light information dropped to 300 lux. At this time, the system adjusted the lighting energy efficiency parameters according to the lighting energy efficiency parameters. Automatically adjust lighting levels to ensure adequate light in the room and optimize energy consumption.

[0048] It should be understood that in this case, assuming the current energy consumption of the cooling system =100 kWh, target energy efficiency parameter =95 kWh. According to the optimization objective function:

[0049] The system adjusts the cooling system power according to the temperature deviation ΔT and energy efficiency deviation, making the energy consumption closer to the target energy efficiency and improving the energy saving effect of the building data center under complex load changes.

[0050] Step S50: The lighting system calculates the energy efficiency of the lighting system according to the ambient light information L and the dynamic energy efficiency parameters of the lighting system. The traffic data U collected in real time by the pedestrian flow sensor is used to dynamically adjust the lighting brightness. The adaptive dimming model is used. The formula is as follows:

[0051] in, Indicates the adjusted lighting brightness. is the basic lighting brightness; is the target energy efficiency parameter, i.e., the dynamic energy efficiency parameter of the lighting system in step S30; is the current lighting energy consumption; and is the weight factor; You should understand that Coefficient and Control the priority allocation of energy efficiency and traffic flow to ensure that the brightness is appropriately increased when there is high traffic flow. This model is suitable for achieving a balance between energy saving and comfort in scenarios with changing demand.

[0052] For example, when =0.8 kWh, =1.0 kWh, passenger flow U=20, if =300lux, α=0.5 and β=0.3, then the adjusted brightness is calculated as: =300+0.5×(0.8−1.0)+0.3×20=300−0.1+6=305.9lux In this way, the system automatically adjusts the brightness to balance energy efficiency needs with environmental comfort.

[0053] Step S60: Loop through steps S10 to S50, and continuously monitor the dynamic energy efficiency parameters of the cooling system and the lighting system. When the dynamic energy efficiency parameters of the cooling system exceed the preset range or the dynamic energy efficiency parameters of the lighting system exceed the preset range, the industrial gateway automatically pushes abnormal information to the front end and issues an alarm.

[0054] It should be understood that step S60 realizes continuous monitoring and dynamic adjustment of the energy efficiency of the cooling system and the lighting system by cyclically executing steps S10 to S50. When the dynamic energy efficiency parameters of the cooling or lighting system exceed the preset range, the system will automatically push abnormal information to the front end and issue an alarm to remind the operation and maintenance personnel.

[0055] It can be understood that this process ensures the real-time and reliability of energy efficiency parameters, enabling the system to quickly identify energy efficiency deviations and take necessary optimization measures to prevent excessive energy consumption or unstable system operation. Through this monitoring and feedback mechanism, the system achieves the technical effect of adaptive adjustment, enabling the cooling and lighting systems to operate efficiently under different load and environmental conditions, thereby significantly improving the energy management effect and system stability of the building data center. This technical effect not only reduces operating costs, but also improves the green energy-saving level and intelligent management capabilities of the building.

[0056] In addition, the present invention also provides an infrastructure management system based on a data center, please refer to Figure 2 , the system comprising: A data acquisition module is used to install a temperature sensor and a humidity sensor on the building cooling system, and to install a current sensor and an ambient light sensor on the lighting system, to collect temperature information T and humidity information H of the cooling system, and current information I and ambient light information L of the lighting system in real time, and to transmit the temperature information T, humidity information H, current information I and ambient light information L to an edge device including an embedded system for processing; The anomaly detection module is used to introduce clustering algorithms into edge devices containing embedded systems for anomaly detection and automatic identification of device failures. The anomaly detection formula uses:

[0057] in, is the current input data, including temperature information T, humidity information H, current information I and ambient light information L; is the mathematical expectation, is the standard deviation corresponding to the current input data; When the anomaly score is higher than the preset threshold, an alarm is triggered and the sampling frequency is increased; The current input data with anomaly scores not higher than the preset threshold After pre-processing, it is transmitted to the industrial gateway for centralized processing; The prediction and analysis module is used for the industrial gateway to receive the pre-processed temperature information T, humidity information H, current information I, and ambient light information L, and then input them into the hybrid model of the convolutional neural network CNN and the long short-term memory network LSTM. CNN extracts the temperature distribution characteristic data, humidity gradient characteristic data, current characteristic data, and ambient light characteristic data, and combines the time series characteristics of the LSTM analysis data to predict the energy consumption demand in the next hour, and combines the temperature information T and humidity information H to generate the dynamic energy efficiency parameters of the cooling system , combining current information I and ambient light information L to generate dynamic energy efficiency parameters of the lighting system , where the loss function selected by the hybrid model is:

[0058] in, is the loss function of the hybrid model, which is used to measure the error between the prediction and the true value. The weight parameter is used to balance short-term and long-term accuracy. and is the mean square error of short-term and long-term forecasts; A cooling optimization module is used to utilize the dynamic energy efficiency parameters of the cooling system generated in step S30 , combined with the Pareto optimal algorithm, the cooling system is optimized for multiple objectives to meet the temperature requirements of each area of ​​the building under the premise of minimizing energy consumption. The optimization objective function is defined as:

[0059] in, is the objective function, represents the energy consumption of the cooling system, is the balance coefficient between energy efficiency and temperature control, Indicates the deviation between the actual temperature and the set temperature. is the deviation weight factor of the dynamic energy efficiency parameter; Lighting adjustment module, used for lighting system according to ambient light information L and lighting system dynamic energy efficiency parameters The traffic data U collected in real time by the pedestrian flow sensor is used to dynamically adjust the lighting brightness. The adaptive dimming model is used. The formula is as follows:

[0060] in, Indicates the adjusted lighting brightness. is the basic lighting brightness; is the target energy efficiency parameter, i.e., the dynamic energy efficiency parameter of the lighting system in step S30; is the current lighting energy consumption; and is the weight factor; The monitoring feedback module is used to execute steps S10 to S50 in a loop and continuously monitor the dynamic energy efficiency parameters of the cooling system and the dynamic energy efficiency parameters of the lighting system. When the dynamic energy efficiency parameters of the cooling system exceed the preset range or the dynamic energy efficiency parameters of the lighting system exceed the preset range, the industrial gateway automatically pushes abnormal information to the front end and issues an alarm.

[0061] The present invention provides an infrastructure management system based on a data center, please refer to Figure 2, using a data center-based infrastructure management method in the above embodiment, can solve a technical problem of data center-based infrastructure management. Compared with the prior art, the beneficial effects of the data center-based infrastructure management system provided by the present invention are the same as the beneficial effects of the data center-based infrastructure management method provided by the above embodiment, and other technical features in the data center-based infrastructure management system are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0062] The present invention provides an infrastructure management device based on a data center, please refer to Figure 3, an infrastructure management device based on a data center includes: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the infrastructure management method based on a data center in the above-mentioned embodiment 1. An infrastructure management device based on a data center in the embodiment of the present invention may include but is not limited to mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle terminals (such as vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. An infrastructure management device based on a data center is only an example and should not bring any limitation to the functions and scope of use of the embodiment of the present invention. An infrastructure management device based on a data center may include a processing device 1001 (such as a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM) 1004. In RAM 1004, various programs and data required for the operation of a data center-based infrastructure management device are also stored. The processing device 1001, ROM 1002, and RAM 1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow a data center-based infrastructure management device to communicate wirelessly or wired with other devices to exchange data. Although a data center-based infrastructure management device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have alternatively.

[0063] The present invention also provides a computer program product, including a computer program, wherein when the computer program is executed by a processor, the steps of the above-mentioned method for infrastructure management based on a data center are implemented. The computer program product provided by the present invention can solve a technical problem of infrastructure management based on a data center. Compared with the prior art, the beneficial effects of the computer program product provided by the present invention are the same as the beneficial effects of the method for infrastructure management based on a data center provided by the above-mentioned embodiment, and will not be described in detail here.

[0064] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through a communication device, or installed from the storage device 1003, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present invention are executed.

[0065] It should be understood that the various parts disclosed in the present invention can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0066] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A data center-based infrastructure management method, characterized in that: include: Step S10: Install a temperature sensor and a humidity sensor on the building cooling system, and install a current sensor and an ambient light sensor on the lighting system, collect temperature information T and humidity information H of the cooling system, and current information I and ambient light information L of the lighting system in real time, and transmit the temperature information T, humidity information H, current information I and ambient light information L to an edge device including an embedded system for processing; Step S20: Introduce a clustering algorithm in the edge device containing the embedded system to perform anomaly detection and automatically identify device failures. The anomaly detection formula is: ;in, is the current input data, including temperature information T, humidity information H, current information I and ambient light information L; is the mathematical expectation, is the standard deviation corresponding to the current input data; When the anomaly score is higher than the preset threshold, an alarm is triggered and the sampling frequency is increased; The current input data with anomaly scores not higher than the preset threshold After pre-processing, it is transmitted to the industrial gateway for centralized processing; Step S30: After receiving the pre-processed temperature information T, humidity information H, current information I, and ambient light information L, the industrial gateway inputs them into a hybrid model of a convolutional neural network CNN and a long short-term memory network LSTM. CNN extracts temperature distribution feature data, humidity gradient feature data, current feature data, and ambient light feature data, and combines the time series characteristics of the LSTM analysis data to predict the energy consumption demand in the next hour. Combined with the temperature information T and humidity information H, the dynamic energy efficiency parameters of the cooling system are generated. , combining current information I and ambient light information L to generate dynamic energy efficiency parameters of the lighting system , where the loss function selected by the hybrid model is: ;in, is the loss function of the hybrid model, which is used to measure the error between the prediction and the true value. The weight parameter is used to balance short-term and long-term accuracy. and is the mean square error of short-term and long-term forecasts; Step S40: Using the dynamic energy efficiency parameters of the cooling system generated in step S30 , combined with the Pareto optimal algorithm, the cooling system is optimized for multiple objectives to meet the temperature requirements of each area of ​​the building under the premise of minimizing energy consumption. The optimization objective function is defined as: ;in, is the objective function, represents the energy consumption of the cooling system, is the balance coefficient between energy efficiency and temperature control, Indicates the deviation between the actual temperature and the set temperature. is the deviation weight factor of the dynamic energy efficiency parameter; Step S50: The lighting system calculates the energy efficiency of the lighting system according to the ambient light information L and the dynamic energy efficiency parameters of the lighting system. The traffic data U collected in real time by the pedestrian flow sensor is used to dynamically adjust the lighting brightness. The adaptive dimming model is used. The formula is as follows: ,in, Indicates the adjusted lighting brightness. is the basic lighting brightness; is the target energy efficiency parameter, i.e., the dynamic energy efficiency parameter of the lighting system in step S30; is the current lighting energy consumption; and is the weight factor; Step S60: Loop through steps S10 to S50, and continuously monitor the dynamic energy efficiency parameters of the cooling system and the lighting system. When the dynamic energy efficiency parameters of the cooling system exceed the preset range or the dynamic energy efficiency parameters of the lighting system exceed the preset range, the industrial gateway automatically pushes abnormal information to the front end and issues an alarm.

2. The data center-based infrastructure management method according to claim 1, characterized in that: In step S10, the sensor will periodically upload its own status information to the industrial gateway, and the sensor includes a temperature sensor, a humidity sensor, a current sensor and an ambient light sensor.

3. The data center-based infrastructure management method according to claim 1, characterized in that: In step S20, the preset threshold is adjusted according to the ambient temperature.

4. The data center-based infrastructure management method according to claim 1, characterized in that: In step S30, the energy consumption demand for the next hour is predicted, including the energy consumption demand of the cooling system, using the formula: ;in, is the predicted energy consumption demand of the cooling system, n is the number of first data samples acquired in advance, i represents the i-th first data sample, is the comprehensive weight coefficient related to the temperature distribution characteristic data and the temperature distribution characteristic data, which is used to adjust The value of is in a reasonable range. is the temperature distribution characteristic data corresponding to the i-th first data sample, is the humidity gradient characteristic data corresponding to the i-th first data sample; The energy demand forecast for the next hour also includes the energy demand forecast for the lighting system, using the formula: ;in, To predict the energy demand of the lighting system, is the number of pre-acquired second data samples, Representative The second data sample, is the comprehensive weight coefficient related to the ambient light characteristic data and the current characteristic data, which is used to adjust The value of is in a reasonable range. For the Ambient light characteristic data corresponding to the second data sample, For the The current characteristic data corresponding to the second data samples.

5. The data center-based infrastructure management method according to claim 1, characterized in that: In step S40, the multi-objective optimization process is reinitialized every 12 hours.

6. The data center-based infrastructure management method according to claim 1, characterized in that: In step S50, the frequency of collecting the flow data U is dynamically adjusted according to the usage of the building.

7. The data center-based infrastructure management method according to claim 1, characterized in that: Step S60 also includes automatically generating a monitoring report once a day for the administrator to review.

8. An infrastructure management system based on a data center, applied to an infrastructure management method based on a data center as claimed in any one of claims 1 to 7, characterized in that: The system comprises: A data acquisition module is used to install a temperature sensor and a humidity sensor on the building cooling system, and to install a current sensor and an ambient light sensor on the lighting system, to collect temperature information T and humidity information H of the cooling system, and current information I and ambient light information L of the lighting system in real time, and to transmit the temperature information T, humidity information H, current information I and ambient light information L to an edge device including an embedded system for processing; The anomaly detection module is used to introduce clustering algorithms into edge devices containing embedded systems for anomaly detection and automatic identification of device failures. The anomaly detection formula uses: ;in, is the current input data, including temperature information T, humidity information H, current information I and ambient light information L; is the mathematical expectation, is the standard deviation corresponding to the current input data; When the anomaly score is higher than the preset threshold, an alarm is triggered and the sampling frequency is increased; The current input data with anomaly scores not higher than the preset threshold After pre-processing, it is transmitted to the industrial gateway for centralized processing; The prediction and analysis module is used for the industrial gateway to receive the pre-processed temperature information T, humidity information H, current information I, and ambient light information L, and then input them into the hybrid model of the convolutional neural network CNN and the long short-term memory network LSTM. CNN extracts the temperature distribution characteristic data, humidity gradient characteristic data, current characteristic data, and ambient light characteristic data, and combines the time series characteristics of the LSTM analysis data to predict the energy consumption demand in the next hour, and combines the temperature information T and humidity information H to generate the dynamic energy efficiency parameters of the cooling system , combining current information I and ambient light information L to generate dynamic energy efficiency parameters of the lighting system , where the loss function selected by the hybrid model is: ;in, is the loss function of the hybrid model, which is used to measure the error between the prediction and the true value. The weight parameter is used to balance short-term and long-term accuracy. and is the mean square error of short-term and long-term forecasts; A cooling optimization module is used to utilize the dynamic energy efficiency parameters of the cooling system generated in step S30 , combined with the Pareto optimal algorithm, the cooling system is optimized for multiple objectives to meet the temperature requirements of each area of ​​the building under the premise of minimizing energy consumption. The optimization objective function is defined as: ;in, is the objective function, represents the energy consumption of the cooling system, is the balance coefficient between energy efficiency and temperature control, Indicates the deviation between the actual temperature and the set temperature. is the deviation weight factor of the dynamic energy efficiency parameter; Lighting adjustment module, used for lighting system according to ambient light information L and lighting system dynamic energy efficiency parameters The traffic data U collected in real time by the pedestrian flow sensor is used to dynamically adjust the lighting brightness. The adaptive dimming model is used. The formula is as follows: ;in, Indicates the adjusted lighting brightness. is the basic lighting brightness; is the target energy efficiency parameter, i.e., the dynamic energy efficiency parameter of the lighting system in step S30; is the current lighting energy consumption; and is the weight factor; The monitoring feedback module is used to execute steps S10 to S50 in a loop and continuously monitor the dynamic energy efficiency parameters of the cooling system and the dynamic energy efficiency parameters of the lighting system. When the dynamic energy efficiency parameters of the cooling system exceed the preset range or the dynamic energy efficiency parameters of the lighting system exceed the preset range, the industrial gateway automatically pushes abnormal information to the front end and issues an alarm.

Images