Search

CN-122015237-A - High-efficient computer lab energy-saving control system based on global optimization

CN122015237ACN 122015237 ACN122015237 ACN 122015237ACN-122015237-A

Abstract

The invention discloses a global optimization-based high-efficiency machine room energy-saving control system, which belongs to the technical field of temperature control and energy saving of a central air conditioner machine room and comprises a multi-source data acquisition module, a refrigeration optimization decision module, an external temperature interaction scheduling module, a control module and a feedback module, wherein the multi-source data acquisition module is used for all-weather collection and fusion of multi-source data, and comprises real-time power consumption, running states of power equipment and refrigeration equipment in the machine room, temperatures of all areas of the machine room and external environment temperature and humidity and weather forecast data acquired through an API (application program interface).

Inventors

  • LI HONGJUN
  • ZHANG JIE

Assignees

  • 浙江同丰科技有限公司

Dates

Publication Date
20260512
Application Date
20260213

Claims (9)

  1. 1. The energy-saving control system of the high-efficiency machine room based on global optimization is characterized by comprising a multi-source data acquisition module, a refrigeration optimization decision module, an external temperature interaction scheduling module, a control module and a feedback module; the multi-source data acquisition module is used for all-weather collection and fusion of multi-source data, and comprises real-time power consumption, running states and temperatures of all areas of a machine room, and external environment temperature and humidity and weather forecast data obtained through an API (application program interface); the refrigeration optimization decision module predicts a machine room temperature rise curve based on historical data, establishes a performance model of each refrigeration device, takes the upper limit of machine room temperature safety as constraint and the lowest total refrigeration energy consumption of the system as a target, and solves the optimal cooperative control strategy of the refrigeration device in real time through an optimization algorithm to realize the dynamic balance of low energy consumption counteracting temperature rise; The external temperature interactive scheduling module is connected with and processes external environment temperature and weather forecast data, quantitatively analyzes the influence rate and amplitude of the external environment temperature and weather forecast data on the internal temperature of the machine room, predicts a specific period of time in which continuous cooling outside can effectively influence the internal temperature of the machine room by combining the starting rules of power equipment in different time periods in historical data, and judges the number of adjustable refrigeration equipment in the period of time in advance; The control module is used for converting the generated strategy instruction into a command executable by specific equipment; and the feedback module continuously collects actual operation data after executing the control strategy, and performs effect evaluation feedback to the refrigeration optimization decision module and the external temperature interaction scheduling module.
  2. 2. The energy-saving control system of a high-efficiency machine room based on global optimization according to claim 1, wherein the multi-source data acquisition module is used for all-weather collection and fusion of multi-source data, and comprises real-time power consumption, running state and temperature of each area of the machine room, and the processing procedures of external environment temperature and humidity and weather forecast data obtained through an API, wherein the real-time power consumption, running state and temperature of power equipment and refrigeration equipment in the machine room are as follows: Through various sensor networks deployed in the machine room, all-weather real-time monitoring and collecting operation data of various power equipment and refrigeration equipment in the machine room, continuously acquiring parameters such as real-time power consumption, operation frequency and start-stop state of each water chilling unit, water pump and cooling tower fan; External environment data which are regularly acquired from a weather service platform, wherein the external environment data comprise real-time outdoor temperature and humidity, wind speed and wind direction and weather forecast information in the future 12 hours; And (3) aligning and integrating two types of data streams of the internal sensor and the external API according to a uniform timestamp, automatically executing data cleaning, removing abnormal values, filling reasonable missing values, normalizing data in different formats and dimensions into a uniform format which can be processed in the system, and forming a complete global data view with a time tag.
  3. 3. The energy-saving control system of a high-efficiency machine room based on global optimization according to claim 1, wherein the refrigeration optimization decision module predicts a machine room temperature rise curve based on historical data, establishes a performance model of each refrigeration device, takes the upper limit of machine room temperature safety as a constraint and the minimum total refrigeration energy consumption of the system as a target, solves the optimal cooperative control strategy of the refrigeration device in real time through an optimization algorithm, and realizes the dynamic balance process of low energy consumption counteraction temperature rise as follows: receiving and integrating current data from a multi-source data acquisition module and similar scene data in a historical database in real time, and predicting the temperature rising trend and speed of a machine room in a period of time in the future under the condition of heat generation of the operation of the existing power equipment by using a machine learning algorithm to form a dynamic machine room temperature rising curve; establishing a refrigeration equipment performance model through original performance data of a refrigeration equipment manufacturer and history data of long-term operation; And (3) coupling a machine room temperature rising curve with performance models of all refrigeration equipment, taking the upper limit of temperature safety which is required to be maintained by the machine room as a core constraint condition, setting the lowest total power consumption of the whole refrigeration system as an optimization target, constructing an optimization problem, solving and calculating through an optimization solving algorithm, and solving the combination which is required to be started by the refrigeration equipment and the set parameters of each operation equipment in a future period of time.
  4. 4. The energy-saving control system of a high-efficiency machine room based on global optimization according to claim 3, wherein the machine room temperature rising curve is coupled with performance models of the refrigeration equipment, the upper limit of temperature safety which must be maintained by the machine room is used as a core constraint condition, the lowest total electricity consumption of the whole refrigeration system is set as an optimization target, an optimization problem is constructed, and a solution calculation is performed through an optimization solution algorithm, so that a combination of the refrigeration equipment to be started in a future period of time and how many processing procedures the set parameters of each operation equipment should be adjusted are as follows: The current equipment operation data and the historical data are fused, a prediction model is used for calculating an accurate track of the temperature rise of the machine room along with time under the existing heat load in a future period, namely a temperature rise curve of the machine room; Calling a built-in refrigeration equipment performance model, dynamically associating a predicted heating curve with performance models of all available refrigeration equipment, converting the requirement that the temperature of a machine room is required to be kept below a safety upper limit into an inequality constraint condition of an optimization problem, constructing a mixed integer optimization problem comprising a continuous variable, namely a set parameter and a discrete variable, namely a start-stop state by taking the sum of total power consumption of all possibly started refrigeration equipment in a prediction period as a target, and obtaining an optimal decision sequence in each future control period through a real-time optimization solution algorithm; the model formula of the objective function is as follows: ; Constraints that need to be met include thermal equilibrium dynamic constraints, temperature safety constraints, equipment operating constraints: wherein the thermal equilibrium dynamic constraint: ; ; wherein, temperature safety constraint: ; Wherein the device operation constraints: ; Wherein, the Is the total energy consumption of the predicted time, In order to predict the number of time-domain steps, For the total number of refrigeration devices, Is an apparatus At the moment of time Is used for the power consumption of the (c) a, Is the moment of time Is used for predicting the temperature of the machine room, Is an apparatus At the setting The refrigerating power of the air conditioner is lower, Is a binary decision variable for starting and stopping the equipment, the value is 0, namely the equipment is closed, the value is 1, namely the equipment is opened, It is the control step-size time that, Is the equivalent heat capacity of the machine room, Is the first When predicting the step length, the total heating power generated by all power equipment in the machine room, The upper limit of the maximum temperature allowed for the safe operation of the equipment in the machine room, Separately for each refrigerating equipment Which can adjust the physical lower and upper limits of the set parameters.
  5. 5. The energy-saving control system of a high-efficiency machine room based on global optimization according to claim 1, wherein the external temperature interaction scheduling module is connected with and processes external environment temperature and weather forecast data, quantitatively analyzes the influence rate and amplitude of the external environment temperature and weather forecast data on the internal temperature of the machine room, predicts a specific period of time in which continuous cooling can effectively influence the interior of the machine room in combination with the starting rules of power equipment in different time periods in historical data, judges the number of adjustable refrigeration equipment in the period in advance, and comprises the following processing procedures: Continuously accessing and processing real-time monitoring data and short-term forecast data from a meteorological department through a preset application program interface, wherein the real-time monitoring data and the short-term forecast data comprise outdoor dry bulb temperature, wet bulb temperature, wind speed and a temperature change curve of 12 hours in the future to form a structured external environment data sequence; Training a heat transfer quantization model based on thermal performance and historical data of the machine room enclosure structure, converting the accessed external temperature data into a direct influence function on the internal temperature of the machine room through calculation, and outputting the machine room temperature drop rate and potential amplitude caused by the external environment in unit time; Synchronously searching a historical database, analyzing the opening rules and power consumption curves of power equipment in a machine room in different date types and different time periods, establishing an internal heat load time sequence model of a typical day, and determining the change rule of the heat generated by the machine room along with the time; Coupling calculation is carried out on an external temperature drop influence function and an internal heat load time sequence model, the external natural cooling capacity of which continuous periods in the future are predicted through simulation comparison, the net heat load in a machine room is sufficiently offset or obviously reduced, a specific time window capable of actively utilizing a natural cold source is identified, and a natural cold source utilization time period table is obtained; Based on the predicted available period of the natural cold source, calculating the specific number, type and capacity of the mechanical refrigeration equipment which can be safely shut down or derated to form a pre-adjustment strategy according to the pre-estimated cold output of natural cooling in the period and the performance characteristics of the refrigeration equipment; The generated natural cold source utilization time period table and the corresponding refrigeration equipment adjustment proposal are used as a prospective constraint condition and optimization target and transmitted to a control module in real time; When the global control strategy is prepared, the method can be planned in advance and is not integrated with a natural cooling scheme, so that the total energy consumption of the system is minimized.
  6. 6. The energy-saving control system of a high-efficiency machine room based on global optimization according to claim 5, wherein the heat transfer quantization model is trained based on the thermal performance and the historical data of the enclosure structure of the machine room, the accessed external temperature data is converted into a direct influence function on the internal temperature of the machine room by calculation, and the processing process of the temperature drop rate and the potential amplitude of the machine room caused by the external environment in unit time is output as follows: Based on a building drawing and material parameters of a machine room, a thermal physical model of an enclosure structure comprising a wall body and a roof is established to describe a theoretical relationship between outdoor temperature and indoor temperature; Calling long-term stored historical data, including a contemporaneous outdoor temperature sequence and a corresponding indoor temperature change sequence of a machine room, aligning a physical model with the actual data on a time axis, and reversely pushing and calibrating a comprehensive heat transfer coefficient and a room effective heat capacity in the physical model by using the historical data by adopting a parameter estimation method so as to enable the output of the model and a historical real temperature change curve to achieve best fit; And the function takes outdoor temperature data accessed in real time as input, directly outputs the temperature drop rate of a machine room and the finally possibly reached steady-state temperature difference amplitude which can be caused in unit time under the current and predicted external temperature conditions by solving a heat balance equation.
  7. 7. The energy-saving control system of a high-efficiency machine room based on global optimization according to claim 5, wherein the synchronous retrieval history database analyzes the opening rules and power consumption curves of power equipment in the machine room in different date types and different time periods, establishes an internal heat load time sequence model of a typical day, and defines the regular processing process of the change of the self heat generation of the machine room along with the time as follows: extracting structured running logs in a specified time range from a historical database, wherein the data comprise accurate time stamps, start-stop states and real-time power readings of each power equipment, and performing data cleaning and time sequence alignment to form a continuous and complete equipment load data sequence; According to the date attribute, the data are automatically classified into different date type templates, potential differences of equipment operation modes under different social activity periods are identified, the data are analyzed according to hour time granularity, the average starting probability, typical load rate and power consumption distribution of power equipment in each time period under each date type are calculated, and a load characteristic curve of time periods is drawn; And extracting a representative load mode of each date type through a regression algorithm, constructing an internal heat load time sequence model taking time as an independent variable, and quantitatively describing a typical rule of the change of the heat generating power of the machine room along with the time of day.
  8. 8. The energy-saving control system of a high-efficiency machine room based on global optimization according to claim 5, wherein the coupling calculation is performed on the external temperature drop influence function and the internal heat load time sequence model, the external natural cooling capacity of which continuous periods in the future are predicted by analog comparison, the net heat load in the machine room is sufficiently offset or remarkably reduced, a specific time window capable of actively utilizing a natural cold source is identified, and the processing procedure of a natural cold source utilization time period table is as follows: Synchronously superposing and calculating the natural temperature drop rate of each future period based on weather forecast output by the external temperature drop influence function and the self heat generation rate of the machine room in the corresponding period predicted by the internal heat load time sequence model to obtain the net heat load of each future time point; Setting a judging threshold value, and identifying continuous time periods in which the net heat load is continuously lower than the judging threshold value, wherein the time periods represent that the external natural cooling capacity can reduce the heat load in the machine room; and combining and optimizing the continuous time periods meeting the conditions, removing segments which are insufficient for stably switching the system due to short duration, generating a natural cold source utilization time period table by combining the minimum start-stop time of the refrigeration equipment and the engineering constraint of the thermal inertia of the system, and marking specific start-stop time windows for starting utilization, maximizing utilization and ending utilization of the natural cold source.
  9. 9. The energy-saving control system of a high-efficiency machine room based on global optimization according to claim 1, wherein the feedback module continuously collects actual operation data after executing a control strategy, and performs effect evaluation feedback to the refrigeration optimization decision module and the external temperature interaction scheduling module, and the processing procedure is as follows: The method comprises the steps of continuously and real-timely collecting actual operation data generated after the latest control strategy is executed through a sensor network and an equipment controller which are arranged in a machine room, wherein the actual operation data comprise final stable temperature of each area of the machine room, instantaneous and accumulated energy consumption of each refrigeration equipment and actual operation state and parameters of the equipment; And carrying out fine comparison on the collected actual data and the predicted data made by the previous refrigeration optimization decision module, calculating the deviation of the key index, and feeding back deviation information to the refrigeration optimization decision module and the external temperature interaction scheduling module.

Description

High-efficient computer lab energy-saving control system based on global optimization Technical Field The invention belongs to the technical field of temperature control and energy conservation of a central air conditioner room, and particularly relates to a global optimization-based energy-saving control system of a high-efficiency machine room. Background The central air conditioning machine room is used as a core supporting facility of a central air conditioning system, heat is continuously generated in the operation process of internal power equipment of the central air conditioning machine room, the temperature is regulated and controlled by means of refrigeration equipment to ensure the stable operation of the system, but the conventional machine room refrigeration control mostly adopts a traditional rough mode, the problem of energy waste is outstanding, the traditional control mode cannot accurately predict a machine room heating curve according to the actual starting quantity and heating characteristics of the power equipment, the starting, stopping and operation parameters of the refrigeration equipment are set to lack scientific basis, the situation that a large number of refrigeration equipment is still started when only part of power equipment is started easily occurs, the invalid consumption of refrigeration energy is caused, meanwhile, the traditional control does not fully utilize the cooling factor of the external natural environment, the energy-saving potential of natural cold sources is ignored, the refrigeration equipment is excessively operated, the whole refrigeration energy consumption of the machine room is high, the comprehensive consideration of the internal and external factors of the machine room is lacked in the existing control mode, the temperature control requirement and the energy-saving aim are difficult to be realized, and therefore the development of a high-efficiency machine room energy-saving control system based on global optimization is urgently needed to solve the problems of the industry. Disclosure of Invention The present invention aims to solve at least one of the technical problems existing in the prior art. Therefore, the invention provides the high-efficiency machine room energy-saving control system based on global optimization, so that related personnel can conveniently solve the problems of high refrigeration energy consumption and inaccurate temperature control strategy of the machine room. In order to achieve the above purpose, the present invention adopts the following technical scheme: an energy-saving control system of a high-efficiency machine room based on global optimization comprises a multi-source data acquisition module, a refrigeration optimization decision module, an external temperature interaction scheduling module, a control module and a feedback module; the multi-source data acquisition module is used for all-weather collection and fusion of multi-source data, and comprises real-time power consumption, running states and temperatures of all areas of a machine room, and external environment temperature and humidity and weather forecast data obtained through an API (application program interface); the refrigeration optimization decision module predicts a machine room temperature rise curve based on historical data, establishes a performance model of each refrigeration device, takes the upper limit of machine room temperature safety as constraint and the lowest total refrigeration energy consumption of the system as a target, and solves the optimal cooperative control strategy of the refrigeration device in real time through an optimization algorithm to realize the dynamic balance of low energy consumption counteracting temperature rise; The external temperature interactive scheduling module is connected with and processes external environment temperature and weather forecast data, quantitatively analyzes the influence rate and amplitude of the external environment temperature and weather forecast data on the internal temperature of the machine room, predicts a specific period of time in which continuous cooling outside can effectively influence the internal temperature of the machine room by combining the starting rules of power equipment in different time periods in historical data, and judges the number of adjustable refrigeration equipment in the period of time in advance; The control module is used for converting the generated strategy instruction into a command executable by specific equipment; and the feedback module continuously collects actual operation data after executing the control strategy, and performs effect evaluation feedback to the refrigeration optimization decision module and the external temperature interaction scheduling module. Further, the multi-source data acquisition module is configured to collect and fuse multi-source data in all weather, and includes real-time power consumption, running state and temperature of ea