Remote analysis based cooling device energy saving control method and system

Abstract

The application relates to the technical field of data processing, and discloses a cooling device energy-saving control method and system based on remote analysis. The method comprises the following steps: installing a sensor to obtain operation parameters and preprocessing to obtain system state data; load prediction is performed in combination with historical records; water pump parameters are optimized based on the predicted value to generate variable frequency control parameters; a temperature difference control standard is established to obtain differentiated target values; a water pump cooperative operation scheme is calculated to form group control instructions; and real-time evaluation and adjustment parameters are calculated to optimize the energy-saving control strategy. The application solves the technical problems of lacking load prediction capability based on historical data analysis, lacking differentiated temperature difference control strategies and lacking a multi-water pump cooperative optimization mechanism in the existing cooling device control method. Through the establishment of an intelligent control system based on remote analysis, efficient and energy-saving operation of the cooling device is realized, and energy consumption is significantly reduced.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to an energy-saving control method and system for a cooling device based on remote analysis. Background Technology

[0002] In modern buildings, cooling water pumps and chilled water pump systems are core components of central air conditioning systems, directly impacting energy consumption and operational efficiency. Traditional cooling unit control methods typically employ a fixed-parameter control strategy, selecting pump capacity based on the maximum design heat load (usually the load at the highest temperature) and maintaining a constant flow rate throughout the year. Under this control method, the pump system operates at a fixed maximum flow rate for extended periods, with the design temperature difference between chilled water and cooling water typically being 6-8℃ and 5-7℃, respectively. With the application of variable frequency technology, some systems have begun to adopt simple variable frequency control, adjusting pump speed based on pressure or temperature difference signals. However, the control strategy remains relatively simple, primarily relying on feedback from local sensors and preset control curves.

[0003] However, these traditional control methods have significant shortcomings. Due to seasonal and diurnal temperature variations, as well as changes in user load, the actual heat load of equipment is far lower than the design load for most of the time. Actual statistics show that the number of operating hours with a load rate below 50% accounts for more than 50% of the total operating time throughout the year. In this situation, the actual temperature difference between chilled water and cooling water is often only 2-4℃, and the water pump system operates in an inefficient state of low temperature difference and high flow rate for a long time, resulting in energy loss in the pipeline system and waste of water pump operating energy. In addition, traditional control methods lack the ability to deeply analyze and predict system operating data, cannot adjust control strategies in advance based on historical operating patterns and changes in the external environment, and also lack coordinated optimization control between multiple water pumps, resulting in low overall system operating efficiency and serious energy waste. Summary of the Invention

[0004] This application provides a method and system for energy-saving control of cooling devices based on remote analysis, which solves the technical problems of existing cooling device control methods, such as lack of load prediction capability based on historical data analysis, lack of differentiated temperature difference control strategy, and lack of multi-pump collaborative optimization mechanism. By establishing an intelligent control system based on remote analysis, the method achieves efficient and energy-saving operation of the cooling device and significantly reduces energy consumption.

[0005] In a first aspect, this application provides an energy-saving control method for a cooling device based on remote analysis. The method includes: acquiring real-time operating parameters by installing temperature sensors, flow sensors, and power monitoring devices on the chilled water pump and cooling water pump system; filtering and preprocessing the operating parameters to obtain cooling system operating status data; performing time-series analysis and trend prediction on the cooling load based on the cooling system operating status data combined with historical load records and ambient temperature changes to obtain a predicted cooling system load value; and based on the predicted cooling system load value and the current pump operating efficiency... The pump speed and flow parameters are calculated and optimized to obtain variable frequency control parameters. Based on the predicted load value of the cooling system and the variable frequency control parameters, temperature difference control standards for different load ranges are established, and the temperature difference between chilled water and cooling water is segmented to obtain differentiated temperature difference target values. According to the differentiated temperature difference target values ​​and the predicted load value of the cooling system, the start-stop status and load distribution of multiple pumps are collaboratively calculated to obtain pump group control commands. Based on the pump group control commands and the operating status data of the cooling system, the control effect is evaluated and parameters are corrected in real time to obtain an optimized energy-saving control strategy.

[0006] Secondly, this application provides a remote analysis-based energy-saving control system for a cooling device, the remote analysis-based energy-saving control system for a cooling device comprising:

[0007] The acquisition module is used to acquire real-time operating parameters by installing temperature sensors, flow sensors and power monitoring devices in the chilled water pump and cooling water pump system, and to perform data filtering and preprocessing on the operating parameters to obtain cooling system operating status data.

[0008] The prediction module is used to perform time-series analysis and trend prediction of the cooling load based on the cooling system operating status data, combined with historical load records and ambient temperature changes, to obtain the predicted value of the cooling system load.

[0009] The calculation module is used to calculate and optimize the pump speed and flow parameters based on the predicted load value of the cooling system and the current pump operating efficiency, so as to obtain the frequency conversion regulation control parameters.

[0010] The setting module is used to set the temperature difference between chilled water and cooling water in segments based on the predicted load value of the cooling system and the frequency conversion regulation control parameters, by establishing temperature difference control standards for different load ranges, and to obtain differentiated temperature difference target values.

[0011] The coordination module is used to perform coordinated calculations on the start-stop status and load distribution of multiple water pumps based on the differentiated temperature difference target value and the cooling system load prediction value, so as to obtain water pump group control commands.

[0012] The calibration module is used to evaluate the control effect and correct the parameters in real time based on the water pump group control command and the cooling system operating status data, so as to obtain an optimized energy-saving control strategy.

[0013] Thirdly, a computer device is provided, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor invokes the instructions in the memory to cause the computer device to execute the above-described energy-saving control method for a cooling device based on remote analysis.

[0014] Fourthly, a computer-readable storage medium is provided, wherein instructions are stored therein, which, when executed on a computer, cause the computer to perform the aforementioned energy-saving control method for a cooling device based on remote analysis.

[0015] The technical solution provided in this application acquires real-time operating parameters by installing a sensor network in the chilled water pump and cooling water pump system. These data are then filtered and preprocessed to obtain high-quality cooling system operating status data, providing a reliable data foundation for subsequent intelligent control. This effectively solves the problems of incomplete data acquisition and unstable data quality in traditional control methods. By combining historical load records and ambient temperature changes to perform time-series analysis and trend prediction of the cooling load, an adaptive prediction model is constructed. This enables the system to perceive load change trends in advance, transforming from passive response to proactive prediction, significantly improving the foresight and accuracy of control. Based on the predicted load value and pump operating efficiency, the pump speed and flow parameters are optimized and calculated to generate variable frequency control parameters, achieving... Precise matching of pump operating conditions avoids energy waste caused by traditional fixed-frequency operation; temperature difference control standards for different load ranges are established based on load prediction values ​​and frequency conversion adjustment parameters, enabling segmented setting of temperature differences between chilled water and cooling water, allowing the system to maintain optimal temperature differences under different load conditions, overcoming the defect of excessively small temperature differences during low-load operation in traditional control; by coordinating the start-stop status and load distribution of multiple pumps, pump group control commands are obtained, achieving overall optimization of the multi-pump system and avoiding overload or inefficient operation of a single pump; real-time evaluation and parameter correction of control effects are performed based on pump group control commands and system operating status data, forming a closed-loop feedback mechanism, enabling the control strategy to continuously self-optimize and adapt to changes in system performance and external environment. Of particular note is the exceptional value of the time-series analysis and trend prediction algorithms employed in this solution, particularly in the field of cooling device control. These algorithms can identify periodic patterns in load changes and temperature-sensitive characteristics. Through in-depth analysis of historical data, they establish accurate load prediction models, providing a basis for decision-making in temperature difference control and water pump regulation. Simultaneously, the water pump collaborative optimization algorithm considers multiple objectives such as balanced equipment operating time and optimal energy efficiency, minimizing energy consumption while ensuring cooling performance. The organic combination of these technical features enables this invention to intelligently adjust system operating parameters according to actual load demands, significantly reducing energy consumption while maintaining cooling performance. Compared to traditional control methods, it can reduce water pump system energy consumption by 15%-30%, with particularly significant energy savings under partial load conditions. Furthermore, it extends equipment lifespan, reduces maintenance costs, and yields significant economic and environmental benefits. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of one embodiment of the energy-saving control method for a cooling device based on remote analysis in this application.

[0018] Figure 2 This is a schematic diagram of one embodiment of the energy-saving control system for a cooling device based on remote analysis in this application.

[0019] Figure 3 This is a schematic block diagram of the structure of the computer device in an embodiment of the present invention. Detailed Implementation

[0020] This application provides an energy-saving control method and system for a cooling device based on remote analysis. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data used can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0021] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the energy-saving control method for cooling devices based on remote analysis in this application includes:

[0022] Step S101: Obtain real-time operating parameters by installing temperature sensors, flow sensors and power monitoring devices in the chilled water pump and cooling water pump system, and perform data filtering and preprocessing on the operating parameters to obtain cooling system operating status data;

[0023] Step S102: Based on the cooling system operating status data, combined with historical load records and ambient temperature changes, perform time-series analysis and trend prediction on the cooling load to obtain the predicted value of the cooling system load.

[0024] Step S103: Based on the predicted load value of the cooling system and the current operating efficiency of the water pump, calculate and optimize the water pump speed and flow parameters to obtain the frequency conversion regulation control parameters;

[0025] Step S104: Based on the predicted load value of the cooling system and the frequency conversion regulation control parameters, by establishing temperature difference control standards for different load ranges, the temperature difference between chilled water and cooling water is set in segments to obtain differentiated temperature difference target values.

[0026] Step S105: Based on the target value of the differentiated temperature difference and the predicted value of the cooling system load, perform collaborative calculations on the start-stop status and load distribution of multiple water pumps to obtain the water pump group control command.

[0027] Step S106: Based on the water pump group control command and the cooling system operating status data, the control effect is evaluated and parameters are corrected in real time to obtain the optimized energy-saving control strategy.

[0028] It is understood that the executing entity of this application can be a remote analysis-based energy-saving control system for cooling devices, or it can be a terminal or a server; no specific limitation is made here. This application's embodiments use a server as an example for illustration.

[0029] Specifically, temperature sensors, flow sensors, and power monitoring devices are installed at key locations in the chilled water pump and cooling water pump systems. These sensors are positioned at the inlet and outlet of the chilled water and cooling water systems, as well as on the main pipelines. The power monitoring devices are connected to the pump power supply lines. The raw operating parameters collected by the sensors are subjected to amplitude limiting filtering to remove outliers and then divided into time series according to a sampling frequency of one minute, forming an effective operating parameter matrix. Temperature difference, flow coefficient, and energy efficiency ratio parameters are extracted from this matrix to calculate instantaneous operating load indicators. These are then compared with the design load to determine the real-time load rate and temperature difference utilization rate. Finally, a standardized parameter set is generated, and a multi-dimensional operating state vector is constructed to form the cooling system operating state data. Load prediction is performed based on the cooling system operating state data combined with historical load records and ambient temperature changes. Specifically, nearly 90 days of system operating state data are extracted from the data storage unit and seasonally decomposed according to weekday and non-weekday types to form a load feature library. The system operating state data and ambient temperature data are time-aligned and fused to construct a temperature-load correlation matrix. The load feature database is periodically analyzed using the autoregressive moving average method to extract the daily, weekly, and monthly cooling load variation patterns, generating time-series pattern features. A temperature sensitivity coefficient is calculated based on the temperature-load correlation matrix, and a temperature correction term is established using 24-hour temperature forecast data. This temperature correction term is applied to the time-series pattern features, and the predicted load values ​​for each future period are obtained through weighted recursive calculation. After smoothing and anomaly constraints, the predicted cooling system load values ​​are formed.

[0030] Pump efficiency distribution maps under current operating conditions are extracted from a pump characteristic curve database and compared with measured power data to calculate pump efficiency deviation. The theoretical water flow rate required by the system is determined based on load prediction, and the corresponding head requirement is calculated using the pipeline resistance characteristic curve. The efficiency deviation value is used to correct the pump characteristic curve, constructing a real-time pump performance evaluation index. For the theoretical water flow rate, the required pump speed range is calculated using the flow-frequency correspondence formula, forming an initial frequency selection set. The energy consumption value at each speed point within the initial frequency selection set is quantitatively evaluated, and the speed value corresponding to the lowest energy consumption is selected. This value is converted into inverter control signal parameters, and combined with the start-up curve and buffer time setting, inverter regulation control parameters are generated.

[0031] The predicted load of the cooling system is calculated and divided into three ranges: low load (below 30%), medium load (30%-70%), and high load (above 70%). The temperature difference distribution in each load range is analyzed from historical operating data, and the optimized temperature difference range under typical operating conditions is extracted to form a temperature difference-load correlation dataset. For the low load range, the baseline values ​​for chilled water temperature difference are determined to be 4-5℃ and cooling water temperature difference to be 3-4℃; for the medium load range, the baseline values ​​are determined to be 5-7℃ and 4-6℃; and for the high load range, the baseline values ​​are determined to be 7-8℃ and 6-7℃. The temperature difference control reference values ​​for each load range are corrected using variable frequency drive (VFD) control parameters, and a transition smoothing coefficient is set based on the load change trend to obtain differentiated temperature difference target values.

[0032] By dividing the predicted load of the cooling system into time periods, and classifying the load demand according to off-peak, off-peak, and peak periods, a load period distribution map is generated. Based on the cumulative operating time and maintenance records of each water pump, pump availability is assessed, and a pump priority sequence table is established. The total flow demand under different load conditions is calculated using differentiated temperature difference target values, and the minimum number of pumps to start is determined by combining the flow characteristics of individual pumps. For off-peak load periods, 1-2 pumps ranked high in the pump priority sequence table are selected to operate, while the remaining pumps are set to standby. For off-peak and peak load periods, the load sharing ratio of each pump is calculated according to the principle of balanced distribution. The off-peak pump configuration scheme and the off-peak pump configuration scheme are integrated into a 24-hour pump operation plan table, with start / stop sequence and speed control information added to obtain pump group control commands.

[0033] Continuous optimization is achieved through a closed-loop feedback mechanism. The system's actual operating data after the execution of pump group control commands is collected by a remote monitoring unit and compared with the expected control targets to calculate temperature difference deviation and energy consumption deviation. The control effect is quantitatively scored, establishing a control quality evaluation index system. Trend analysis of the parameters in the control quality evaluation index system identifies weak links in the control, forming an optimization direction list. For weak links, corresponding control parameters are extracted, correction coefficients are calculated, and a parameter correction table is obtained. Based on the parameter correction table, the load prediction model, the pump frequency conversion control curve, and the temperature difference setpoint are fine-tuned to form an improved control scheme. The improved control scheme is then distributed to the field controllers via a remote management platform, updating the control logic and setpoints to obtain the optimized energy-saving control strategy.

[0034] For example, this method was used for energy-saving control of the central air conditioning system in a commercial building. The system includes four chilled water pumps and four cooling water pumps. Data collected by temperature sensors showed that under traditional control, the chilled water temperature difference was only 2.8℃ and the cooling water temperature difference was 3.1℃, far below the design temperature difference. After applying this method, the system dynamically adjusted the temperature difference target based on load prediction results, increasing the chilled water temperature difference to 5.8℃ and the cooling water temperature difference to 4.9℃ under medium load conditions. Simultaneously, the pump group control strategy only operates two pumps during low-load periods (such as at night) and reduces the operating frequency to 32Hz, while using a balanced load operation mode with three pumps during high-load periods. The control effect is continuously evaluated through a remote analysis platform, and control parameters are dynamically adjusted for different seasons and usage patterns, thereby achieving continuous optimization of energy-saving control.

[0035] In this embodiment, a sensor network is installed in the chilled water pump and cooling water pump system to acquire real-time operating parameters. These data are then filtered and preprocessed to obtain high-quality cooling system operating status data, providing a reliable data foundation for subsequent intelligent control. This effectively solves the problems of incomplete data acquisition and unstable data quality in traditional control methods. By combining historical load records and ambient temperature changes to perform time-series analysis and trend prediction of the cooling load, an adaptive prediction model is constructed. This enables the system to perceive load change trends in advance, shifting from passive response to proactive prediction, significantly improving the foresight and accuracy of control. Based on the predicted load value and pump operating efficiency, the pump speed and flow parameters are optimized and calculated to generate variable frequency control parameters, realizing the pump... Precise matching of operating conditions avoids energy waste caused by traditional fixed-frequency operation; temperature difference control standards for different load ranges are established based on load prediction values ​​and frequency conversion adjustment parameters, enabling segmented setting of the temperature difference between chilled water and cooling water, allowing the system to maintain the optimal temperature difference under different load conditions, overcoming the defect of excessively small temperature difference during low-load operation in traditional control; by coordinating the start-stop status and load distribution of multiple water pumps, water pump group control commands are obtained, realizing the overall optimization of the multi-pump system and avoiding overload or inefficient operation of a single pump; the control effect is evaluated and parameters are corrected in real time based on the water pump group control commands and system operating status data, forming a closed-loop feedback mechanism, enabling the control strategy to continuously self-optimize and adapt to changes in system performance and external environment. Of particular note is the exceptional value of the time-series analysis and trend prediction algorithms employed in this solution, particularly in the field of cooling device control. These algorithms can identify periodic patterns in load changes and temperature-sensitive characteristics. Through in-depth analysis of historical data, they establish accurate load prediction models, providing a basis for decision-making in temperature difference control and water pump regulation. Simultaneously, the water pump collaborative optimization algorithm considers multiple objectives such as balanced equipment operating time and optimal energy efficiency, minimizing energy consumption while ensuring cooling performance. The organic combination of these technical features enables this invention to intelligently adjust system operating parameters according to actual load demands, significantly reducing energy consumption while maintaining cooling performance. Compared to traditional control methods, it can reduce water pump system energy consumption by 15%-30%, with particularly significant energy savings under partial load conditions. Furthermore, it extends equipment lifespan, reduces maintenance costs, and yields significant economic and environmental benefits.

[0036] In one specific embodiment, the process of performing step S101 may specifically include the following steps:

[0037] Temperature data is collected at the inlet and outlet of chilled water and cooling water by temperature sensors, water flow data is measured on the main pipeline by flow sensors, and water pump current and power consumption data are recorded by power monitoring devices to obtain the original operating parameters of the cooling system.

[0038] The original operating parameters are divided into time series according to the sampling frequency per minute, and outliers are removed by amplitude limiting filtering to obtain the effective operating parameter matrix;

[0039] Extract temperature difference, flow coefficient and energy efficiency ratio parameters from the effective operating parameter matrix, and calculate instantaneous operating load indicators;

[0040] By comparing the instantaneous operating load index with the design load, the real-time load rate and temperature difference utilization rate are determined, and a standardized parameter set is generated.

[0041] A multidimensional operating state vector is constructed using a standardized parameter set, and statistical features of the operating state are calculated through time-domain feature extraction to form operating state data of the cooling system.

[0042] Specifically, data acquisition is achieved by installing multiple sensors at key locations in the chilled water pump and cooling water pump systems. Temperature sensors are installed at the inlet and outlet of the chilled water and cooling water systems, including the supply and return pipes, to collect water temperature data. Flow sensors are installed on the main pipelines, including the chilled water main and cooling water main, to measure water flow data. Power monitoring devices are connected to the power supply lines of the pump motors to record current and power consumption data. All sensor data is transmitted in real-time to the data acquisition controller via industrial communication networks (such as Modbus and BACnet protocols) to form the raw operating parameters of the cooling system. Preprocessing the acquired raw operating parameters is a crucial step in ensuring data quality. The data acquisition controller collects data at a sampling frequency of once per minute, forming a time series from the continuous data. For these time series data, an amplitude-limiting filter method is used to remove outliers. This method sets upper and lower limits for the normal range; data points exceeding the range are marked as outliers and replaced with the average of the nearest valid values. For example, when a temperature sensor reading suddenly jumps from 12°C to 35°C and then returns to 13°C, the 35°C reading is identified as an outlier and corrected. The data after outlier processing is organized into a multidimensional matrix, with rows representing different time points and columns representing different parameter indicators, forming an effective operating parameter matrix.

[0043] Extracting key performance indicators from the effective operating parameter matrix is ​​an important means of understanding the system's operating status. Temperature difference calculation includes chilled water temperature difference and cooling water temperature difference, which are the differences between the inlet and outlet temperatures, respectively. The flow coefficient represents the ratio of actual flow rate to design flow rate, reflecting the degree to which the system flow rate deviates from the design value. The energy efficiency ratio parameter represents the cooling effect per unit power input. The formula for calculating the instantaneous operating load index based on the above parameters is:

[0044]

[0045] in, Indicates instantaneous operating load metrics. This indicates the specific heat capacity of water. This indicates the density of water. Indicates the actual measured flow rate. This indicates the temperature difference between the inlet and outlet of the chilled water. This indicates the load value under the design operating conditions.

[0046] By comparing the instantaneous operating load index with the design load, the real-time load rate and temperature difference utilization rate are calculated. These two parameters reflect the current load level and temperature difference utilization efficiency of the system, respectively. By standardizing these calculation results, a standardized parameter set is obtained, allowing parameters with different dimensions to be compared and analyzed on the same scale.

[0047] Finally, a multi-dimensional operating status vector is constructed using a standardized parameter set, including standardized parameters such as temperature difference, flow rate, and load rate. Time-domain features are extracted from this vector, and statistical characteristics such as mean, standard deviation, and peak value are calculated to form cooling system operating status data. This data includes both real-time operating status and trend information over a period of time.

[0048] Taking a central air conditioning system in a commercial building as an example, the system is designed with a chilled water supply and return temperature difference of 7℃ and a cooling water supply and return temperature difference of 5℃. Real-time data collection via temperature sensors installed on the pipes revealed that the actual chilled water supply temperature was 7℃, the return temperature was 9.5℃, and the temperature difference was 2.5℃; the cooling water inlet temperature was 30℃, the outlet temperature was 33℃, and the temperature difference was 3℃. Simultaneously, the flow sensor measured a chilled water flow rate of 500 m³ / h, while the designed flow rate was 300 m³ / h. The power monitoring device recorded a pump power of 75 kW. After amplitude limiting and filtering, abnormal temperature fluctuations (such as instantaneous sensor reading jumps) were smoothly corrected. Calculations showed a flow coefficient of 1.67, indicating that the system flow rate is far higher than the design value; the temperature difference utilization rate was only 35.7%, indicating that the system was operating under low temperature difference conditions; and the real-time load rate was 60%, indicating a moderate system load. These data are standardized to form operating status vectors. Statistical features are extracted through time-domain feature extraction, ultimately generating cooling system operating status data, providing a data foundation for subsequent load prediction and control optimization.

[0049] In one specific embodiment, the process of performing step S102 may specifically include the following steps:

[0050] Nearly 90 days of cooling system operation status data were extracted from the data storage unit and seasonally decomposed according to weekday and non-weekday types to form a load characteristic library.

[0051] The cooling system operating status data is time-aligned and data fused with the ambient temperature data from the meteorological data center to construct a temperature load correlation matrix.

[0052] The load feature library is periodically analyzed using the autoregressive moving average method to extract the daily, weekly, and monthly cooling load variation patterns and generate time-series pattern features.

[0053] The temperature sensitivity coefficient is calculated based on the temperature load correlation matrix, and a temperature correction term is established by combining the temperature forecast data for the next 24 hours.

[0054] A temperature correction term is applied to the time series pattern characteristics, and the predicted load values ​​for each future time period are obtained through weighted recursive calculation.

[0055] The predicted load values ​​for each future time period are smoothed and anomaly constraints are applied to obtain the predicted load values ​​for the cooling system.

[0056] Specifically, the load forecasting process extracts nearly 90 days of cooling system operating status data from the data storage unit. This data includes information such as temperature difference, flow rate, and energy consumption collected and processed in previous steps. The data storage unit refers to a server or cloud database specifically designed to store historical operating data, possessing large-capacity storage and fast retrieval capabilities. The extracted 90-day data undergoes seasonal decomposition according to weekday and non-weekday types. Seasonal decomposition is a statistical method that separates time-series data into trend, seasonal, and random components. Specifically, the 24-hour data is categorized into weekdays (Monday to Friday) and non-weekdays (Saturday, Sunday, and public holidays), and further subdivided according to different months or seasons to form a load feature library. The load feature library is a multi-dimensional data structure containing load pattern information at different time scales (hourly, daily, weekly, and monthly). Time alignment ensures that two sets of data are consistent in timestamps, typically requiring interpolation or downsampling for data with different sampling frequencies. Data fusion integrates the two sets of data into a unified data structure, constructing a temperature-load correlation matrix. The temperature-load correlation matrix is ​​a two-dimensional data table. Each row represents a point in time, and the columns include the ambient temperature and the corresponding system load value. This matrix can be used to analyze the degree of impact of temperature changes on the load.

[0057] The load characteristic database is analyzed periodically using the Autoregressive Moving Average (ARMA) method to extract load variation patterns. The ARMA method is a time series analysis technique that combines autoregressive (AR) and moving average (MA) models to capture short-term and long-term trends in data. In this method, the ARMA model is applied to analyze cooling load variation patterns within a day (different time periods within 24 hours), a week (each day of the week), and a month (different dates within a month). The periodic patterns obtained through analysis are called time series pattern characteristics, which describe the basic pattern of load variation over time without considering temperature changes.

[0058] The temperature sensitivity coefficient is calculated based on the temperature load correlation matrix. This coefficient represents the change in cooling load for every 1°C change in ambient temperature. The calculation method involves performing regression analysis on the data in the temperature load correlation matrix to obtain the relationship function between temperature and load. Then, the temperature sensitivity coefficient is combined with the temperature forecast data for the next 24 hours to establish a temperature correction term. The temperature correction term adjusts the base load forecast value to reflect the impact of future temperature changes on the load.

[0059] A temperature correction term is applied to the time-series pattern features, and the predicted load values ​​for each future time period are obtained through weighted recursive calculation. The formula for weighted recursive calculation is:

[0060]

[0061] in, This represents the predicted load value at time t. This represents the temporal pattern feature value at time t. This represents the temperature correction term at time t. This represents the predicted load value at time t-1. These are the weighting coefficients for the time-series pattern features, the temperature correction term, and the predicted value from the previous time step, respectively, and satisfy the following conditions: The weighting coefficients can be dynamically adjusted based on historical prediction accuracy to optimize prediction precision.

[0062] Finally, the predicted load values ​​for each future time period are smoothed and anomaly constraints are applied. Smoothing uses a moving average algorithm to reduce abrupt changes in the prediction curve; anomaly constraints set upper and lower thresholds to ensure the prediction results fall within a reasonable range. After these processes, the final predicted cooling system load values ​​are obtained.

[0063] Taking the central air conditioning system of a commercial complex as an example, the system extracted nearly 90 days of operating data from its data storage unit and found significant differences in load patterns between weekdays and non-weekdays. Weekday load began to rise at 8:00 AM, peaking at 12:00 PM and 3:00 PM, and gradually declining after 6:00 PM. Non-weekday load, on the other hand, slowly rose from 10:00 AM, peaking at 2:00 PM, and then gradually declined. Through autoregressive moving average analysis, this intraday and weekly load variation pattern was extracted as a time-series pattern characteristic. Simultaneously, calculations using the temperature-load correlation matrix revealed that the system's temperature sensitivity coefficient was approximately 3% / ℃, meaning that for every 1℃ increase in ambient temperature, the cooling load increased by approximately 3%. In a weekday load forecast, based on historical time-series pattern characteristics, the load was initially predicted to be 65% of the design load. However, the weather forecast indicated that the temperature would be 4℃ higher than the historical average for the same period. Through weighted recursive calculations, a temperature correction term was added and combined with the previous forecast value, ultimately predicting that the load for the day would be 75% of the design load. This forecasting method, which integrates time-series patterns and temperature effects, ensures the accuracy of cooling system load forecasting.

[0064] In one specific embodiment, the process of executing step S103 may specifically include the following steps:

[0065] The pump efficiency distribution map under the current operating condition is extracted from the pump characteristic curve database, compared and analyzed with the measured power data, and the pump efficiency deviation value is calculated.

[0066] The theoretical value of the required water flow rate of the system is determined based on the predicted load value of the cooling system, and the corresponding head requirement is calculated by combining the resistance characteristic curve of the pipeline network.

[0067] The efficiency deviation value is used to correct the pump characteristic curve, and a real-time pump performance evaluation index is constructed.

[0068] Based on the theoretical value of water flow, the required pump speed range is calculated using the flow-frequency correspondence formula to form an initial frequency selection set;

[0069] The energy consumption values ​​of each speed point in the initial frequency selection set are quantitatively evaluated, and the speed value corresponding to the lowest energy consumption is selected by the energy consumption-flow optimization algorithm.

[0070] The speed value corresponding to the lowest energy consumption is converted into inverter control signal parameters, and combined with the start-up curve and buffer time setting, inverter regulation control parameters are generated.

[0071] Specifically, the efficiency distribution map of the pump under the current operating conditions is extracted from the pump characteristic curve database. This database is a dedicated database storing the performance parameters of different models and specifications of pumps under various operating conditions. For each pump, the database stores characteristic curves such as flow-head curve, flow-efficiency curve, and flow-power curve. The extracted efficiency distribution map reflects the efficiency changes of the pump under different flow and head conditions. The efficiency distribution map is compared and analyzed with the power data measured by the power monitoring device to calculate the pump efficiency deviation value. The efficiency deviation value refers to the difference between the actual operating efficiency and the theoretical efficiency, usually expressed as a percentage, reflecting the degree of deviation between the actual operating state and the theoretical performance of the pump. This deviation mainly comes from factors such as pump aging, wear, or improper installation and commissioning. Based on the cooling system load prediction value obtained in the previous step, the theoretical value of the required water flow rate of the system is determined. The cooling system load prediction value refers to the expected cooling load of the system in the future. The theoretical water flow rate required to meet this load is calculated using the formula relating heat load and water flow rate. At the same time, the corresponding head requirement is calculated in conjunction with the pipeline resistance characteristic curve. The pipeline resistance characteristic curve describes the relationship between the resistance of the system pipeline under different flow rates, and is usually expressed as a quadratic function of the flow rate. By using the pipeline resistance characteristic curve, the head required by the system at a specific flow rate can be determined, that is, the head that the water pump needs to provide.

[0072] The pump characteristic curve is corrected using the efficiency deviation value calculated earlier, thus constructing a real-time pump performance evaluation index. The correction process involves multiplying the efficiency value on the original pump characteristic curve by a correction coefficient, which is determined by the efficiency deviation value. The corrected characteristic curve more accurately reflects the actual operating performance of the pump, while the real-time pump performance evaluation index comprehensively considers parameters such as efficiency, power, and head to evaluate the pump's operating status under current conditions.

[0073] Based on the theoretical water flow rate required by the system, the required pump speed range is calculated using the flow rate-frequency correspondence formula, forming an initial frequency selection set. The flow rate-frequency correspondence formula, based on the pump similarity law, describes the relationship between pump speed and flow rate. This formula can be expressed as:

[0074]

[0075] in, Indicates the required inverter output frequency. Indicates the rated frequency of the water pump. This represents the theoretical value of the water flow required by the system. Indicates the rated flow rate of the water pump. This represents the system characteristic correction factor, which takes into account the influence of pipeline characteristics on the frequency-flow relationship.

[0076] The frequency value calculated using the above formula is usually not unique, but rather a range, thus forming an initial frequency selection set. The initial frequency selection set refers to a series of possible inverter output frequency values ​​that meet the flow requirements; it is typically a set formed by floating a certain percentage upwards and downwards from the baseline calculated value.

[0077] The energy consumption values ​​at each rotational speed within the initial frequency selection set are quantitatively evaluated, and the rotational speed value corresponding to the lowest energy consumption is selected using an energy consumption-flow optimization algorithm. The core of the energy consumption-flow optimization algorithm is to construct the objective function:

[0078]

[0079] in, It is to optimize the objective function. Indicates the frequency of the inverter. This represents the power consumption of the water pump at frequency f. This represents the actual flow rate of the water pump at frequency f. This represents the theoretical value of the water flow required by the system. It is a balancing factor used to adjust the weight of the flow deviation penalty.

[0080] By iteratively calculating the objective function value for each frequency point in the initial frequency selection set, the frequency point that minimizes the objective function is found, which is the speed value corresponding to the lowest energy consumption. This speed value satisfies the system's flow requirements while minimizing energy consumption. The speed value corresponding to the lowest energy consumption is converted into inverter control signal parameters, and combined with the start-up curve and buffer time setting, inverter regulation control parameters are generated. Inverter control signal parameters include frequency setpoint, acceleration / deceleration time, minimum frequency limit, etc. The start-up curve refers to the process of the inverter adjusting from the current frequency to the target frequency, usually using an S-shaped curve to reduce mechanical shock. The buffer time setting is the minimum interval time set between two speed adjustment operations to prevent system instability caused by frequent speed adjustments.

[0081] Taking the central air conditioning cooling system of an office building as an example, the system uses four chilled water pumps connected in parallel, each with a rated flow rate of 300 m³ / h and a rated frequency of 50 Hz. The efficiency distribution chart of the current pump model is extracted from the pump characteristic curve database, showing that the pump's efficiency under rated conditions is 78%. The actual power of the pump measured by the power monitoring device is 45 kW, while the theoretical calculated power is 40 kW, resulting in an efficiency deviation of -11.1%, indicating that the actual pump efficiency is lower than the theoretical value. The system load prediction is 60% of the design load, corresponding to a theoretical required water flow rate of 720 m³ / h. Based on the pipe network resistance characteristic curve, the required head at this flow rate is calculated to be 28 meters of water column. The efficiency deviation value is used to correct the pump characteristic curve, constructing a performance evaluation index that considers the actual efficiency reduction. Using the flow-frequency correspondence formula, the frequency required to meet the 720 m³ / h flow rate is approximately 36 Hz. Considering a certain adjustment margin, an initial frequency selection set of 34 Hz ​​to 38 Hz is formed. Energy consumption was evaluated at each frequency point within the initial frequency selection set. It was found that the lowest energy consumption was achieved at 36.5Hz, with a power consumption of approximately 41kW, while still meeting the flow requirements. Ultimately, 36.5Hz was set as the inverter output frequency, with an acceleration time of 20 seconds, a deceleration time of 30 seconds, and a buffer time of 10 minutes, forming complete inverter regulation and control parameters, thus achieving efficient and energy-saving operation of the water pump.

[0082] In one specific embodiment, the process of executing step S104 may specifically include the following steps:

[0083] The load rate of the predicted load value of the cooling system is calculated and the load rate is divided into three intervals: low load interval, medium load interval and high load interval.

[0084] By analyzing the temperature difference distribution in each load range in historical operating data, the temperature difference optimization range under typical operating conditions is extracted to form a temperature difference-load correlation dataset.

[0085] For the low load range, the reference value for chilled water temperature difference is determined to be 4-5℃, and the reference value for cooling water temperature difference is 3-4℃, thus generating a reference value for low load temperature difference control.

[0086] For the medium load range, the reference value for chilled water temperature difference is determined to be 5-7℃, and the reference value for cooling water temperature difference is 4-6℃, thus generating reference values ​​for medium load temperature difference control.

[0087] For the high load range, the reference value for chilled water temperature difference is determined to be 7-8℃, and the reference value for cooling water temperature difference is 6-7℃, thus generating reference values ​​for high load temperature difference control.

[0088] The temperature difference control reference value for each load range is corrected and calculated using the variable frequency regulation control parameters. A transition smoothing coefficient is set in combination with the load change trend to obtain the differentiated temperature difference target value.

[0089] Specifically, the predicted load value of the cooling system is used to calculate the load rate, which is the ratio of the current load to the design maximum load. The calculation method is to divide the predicted cooling system load by the system design load value to obtain a percentage value. Based on the calculated load rate, the load state is divided into three intervals: low load interval (load rate less than 30%), medium load interval (load rate between 30% and 70%), and high load interval (load rate greater than 70%). This interval division is not arbitrary but determined based on the operating characteristics and energy efficiency curve of the cooling system. The optimal operating parameters of the system differ significantly within different load intervals.

[0090] By analyzing the temperature difference distribution across load intervals in historical operating data, the optimal temperature difference range under typical operating conditions is extracted. Historical operating data refers to the actual operating data of the system under various load conditions over a past period (such as the past year), including parameters such as chilled water temperature difference, cooling water temperature difference, and energy efficiency ratio. The data analysis process first categorizes the historical data by load rate. Then, within each load interval, the distribution of chilled water and cooling water temperature differences is statistically analyzed, including average, standard deviation, maximum, and minimum values. From these statistical results, the temperature difference range corresponding to the highest energy efficiency ratio is identified as the optimal temperature difference range for that load interval. This data analysis method creates a temperature difference-load correlation dataset, which records the optimal temperature difference values ​​under different load rates, providing data support for subsequent temperature difference control.

[0091] For the low-load range (load rate less than 30%), analysis of the temperature difference-load correlation dataset revealed that the optimal range for chilled water temperature difference is 4-5℃, and for cooling water temperature difference, it is 3-4℃. This is because under low-load conditions, the system operates at partial load, with relatively high water flow and reduced heat exchange efficiency. Forcibly maintaining the design temperature difference (typically 6-8℃) would lead to frequent pump speed adjustments or insufficient water flow, affecting system stability. Therefore, based on these characteristics, a reference value for low-load temperature difference control was determined: under low-load operation, the chilled water temperature difference should be controlled within the range of 4-5℃, and the cooling water temperature difference within the range of 3-4℃. For the medium-load range (load rate between 30% and 70%), the temperature difference-load correlation dataset showed that the optimal range for chilled water temperature difference is 5-7℃, and for cooling water temperature difference, it is 4-6℃. Under medium-load conditions, the system operates more stably, and the balance point between water flow and temperature difference changes. Appropriately increasing the target temperature difference value can reduce pump energy consumption. Based on this, the reference values ​​for temperature difference control under medium load are determined, namely, when operating under medium load, the temperature difference of chilled water is controlled within the range of 5-7℃, and the temperature difference of cooling water is controlled within the range of 4-6℃.

[0092] Analysis of the temperature difference-load correlation dataset for the high-load range (load rate greater than 70%) shows that the optimal chilled water temperature difference range is 7-8℃, and the optimal cooling water temperature difference range is 6-7℃. Under high-load conditions, the system's heat exchange efficiency is high, achieving an operating state close to the design temperature difference. At this point, a higher temperature difference helps reduce water flow and lower pump energy consumption. Therefore, the reference values ​​for high-load temperature difference control are determined: during high-load operation, the chilled water temperature difference should be controlled within the range of 7-8℃, and the cooling water temperature difference within the range of 6-7℃.

[0093] The variable frequency drive (VFD) control parameters obtained in the previous steps are used to correct the reference values ​​for temperature difference control in each load range. These parameters include information such as pump speed and flow rate, which can be used to estimate the temperature difference trend under actual operating conditions. The specific method for correction is to fine-tune the reference value for temperature difference in each load range based on the relationship between water flow rate and temperature difference determined by the VFD control parameters. Simultaneously, to avoid abrupt changes in the temperature difference setpoint at the load range boundaries, a transition smoothing coefficient is set in conjunction with the load change trend. This transition smoothing coefficient is a value between 0 and 1, used to achieve a smooth transition of the target temperature difference value when the load rate approaches the range boundary. Through this correction and smoothing process, a differentiated target temperature difference value applicable to the entire load range is finally obtained.

[0094] Taking the central air conditioning system of a commercial complex as an example, the system is designed with a chilled water temperature difference of 7℃ and a cooling water temperature difference of 5℃, with a maximum designed cooling load of 5000kW. In actual operation, based on historical data analysis, when the load rate is around 25%, the system achieves the highest chilled water temperature difference with the best energy efficiency ratio at approximately 4.5℃; when the load rate is around 50%, the optimal chilled water temperature difference is approximately 6.3℃; and when the load rate is 85%, the optimal chilled water temperature difference approaches 7.5℃. Based on these data, a differentiated temperature difference control strategy was constructed. On a spring day, at 8:00 AM, the system load forecast was 1350 kW, with a calculated load rate of 27%, falling into the low load range. The target temperature difference was set at 4.8°C for chilled water and 3.5°C for cooling water. At 12:00 PM, the load forecast rose to 3200 kW, reaching a load rate of 64%, falling into the medium load range. The target temperature difference was adjusted accordingly to 6.5°C for chilled water and 5.2°C for cooling water. At 3:00 PM, the load further increased to 4100 kW, with a load rate of 82%, entering the high load range. The target temperature difference was set at 7.3°C for chilled water and 6.5°C for cooling water. Through this differentiated temperature difference control strategy, the system can maintain high operating efficiency under different load conditions throughout the day, achieving intelligent energy-saving control.

[0095] In one specific embodiment, the process of executing step S105 may specifically include the following steps:

[0096] By dividing the predicted load values ​​of the cooling system into time periods, and classifying the load demand according to the off-peak, off-peak, and peak periods, a load time distribution map is formed.

[0097] Based on the cumulative operating time and maintenance status records of each water pump, the availability of the water pumps is assessed, and a water pump priority sequence table is established.

[0098] The total flow demand under different load conditions is calculated using the target value of the differential temperature difference, and the minimum number of pumps to be started is determined by combining the flow characteristics of a single pump.

[0099] For off-peak load periods, select 1-2 pumps ranked first in the pump priority sequence to run, and set the remaining pumps to standby mode to generate an off-peak pump configuration scheme.

[0100] For the off-peak and peak load ranges, calculate the load sharing ratio of each water pump according to the principle of balanced distribution, and generate a water pump configuration scheme for off-peak and peak periods.

[0101] The pump configuration schemes for off-peak and peak periods are integrated into a 24-hour pump operation plan, and start-stop sequence and speed control information are added to obtain pump group control commands.

[0102] Specifically, by dividing the predicted load value of the cooling system into time periods, the load demand within 24 hours is classified into low-load, off-peak, and high-load periods according to load levels. The specific method for time period division is to set thresholds based on the magnitude of the predicted load value. Typically, periods with a load rate below 30% are defined as low-load periods, periods with a load rate between 30% and 70% are defined as off-peak periods, and periods with a load rate above 70% are defined as high-load periods. After time period division, the load situation of different time periods is plotted on the horizontal axis with time as the horizontal axis and load rate as the vertical axis, forming a load time period distribution chart. This chart visually displays the trend of load changes throughout the day and the load level of each time period. Based on the cumulative operating time and maintenance status records of each water pump, the availability of the water pumps is assessed, and a water pump priority sequence list is established. The cumulative operating time refers to the total operating time of each water pump since it was put into use or the last major overhaul, obtained through the operating time counter of the water pump control system. The maintenance status records include information such as the water pump's maintenance history, fault records, and vibration monitoring data, reflecting the health status of the water pumps. Pump availability assessment employs a comprehensive scoring method, considering factors such as total uptime, time until the next scheduled maintenance, and operational stability since the most recent maintenance. Assessment results are ranked according to score, forming a pump priority list. Pumps ranked higher in the list are selected first, while lower-ranked pumps are used as backups or activated under high load conditions.

[0103] The total flow rate demand under different load conditions is calculated using the differentiated temperature difference target values ​​obtained in the previous steps. The specific calculation method is based on the heat transfer formula; given the load value and the temperature difference target value, the water flow rate required to meet the heat transfer demand can be calculated. For chilled water systems, the flow rate calculation formula is: Flow rate = Cooling load ÷ (4.18 × Density × Temperature difference), where 4.18 is the specific heat capacity of water, density is the density of water, and temperature difference is the temperature difference between the inlet and outlet of the chilled water. The calculated total flow rate demand is combined with the flow rate characteristics of individual water pumps to determine the minimum number of pumps required to start. The flow rate characteristics of individual water pumps refer to the range of flow rates that a single water pump can provide at different speeds, usually obtained from the pump nameplate data or test data. The principle for determining the minimum number of pumps to start is to ensure that the total flow rate meets the demand, while simultaneously ensuring that each water pump operates in its high-efficiency range as much as possible, avoiding overloading of any single water pump.

[0104] For off-peak load periods, select 1-2 pumps from the priority list to operate, and set the remaining pumps to standby. The selection is based on the flow demand during off-peak periods and the minimum stable flow rate of each pump. During off-peak periods, the system load is low, and typically 1-2 pumps are sufficient to meet the flow demand. Operating higher-priority pumps at this time balances the usage time of each pump, extending equipment lifespan. For pumps in standby mode, the control system sets them to standby mode but keeps them in a preheated state for rapid startup when needed. This configuration information forms the off-peak pump configuration scheme.

[0105] For off-peak and peak load periods, the load sharing ratio of each pump is calculated according to the principle of balanced distribution. The principle of balanced distribution means that, while meeting the total flow demand, the load rate of each operating pump should be as close as possible to avoid situations where some pumps are overloaded while others are underloaded. The calculation of the load sharing ratio considers the pump's rated flow rate, current efficiency, and operating status. An optimization algorithm determines the optimal operating frequency of each pump to minimize overall energy consumption. During off-peak periods, 2-3 pumps typically need to operate simultaneously, while during peak periods, 3-4 pumps may be required, the specific number depending on the flow demand and pump capacity. The calculated operating status and load sharing ratio of each pump constitute the pump configuration scheme for off-peak and peak periods.

[0106] The pump configuration schemes for off-peak and peak periods are integrated into a 24-hour pump operation plan. During this integration, special attention must be paid to the pump start-up and shutdown arrangements during time period transitions to ensure a smooth transition and avoid frequent start-ups and shutdowns. Start-up and shutdown sequence information, including start-up time, start-up order, and preheating time, as well as speed control information, including the target frequency and speed regulation curve for each pump during each time period, are added to the operation plan. The complete operation plan constitutes the pump group control command, which will be sent to the field controller for execution.

[0107] Taking a hotel's central air conditioning system as an example, the system is equipped with four chilled water pumps, each with a rated flow rate of 250 m³ / h. By analyzing load forecasts, the 24-hour period is divided into off-peak (1:00 AM to 7:00 AM), off-peak (7:00 AM to 2:00 PM and 8:00 PM to 1:00 AM), and peak (2:00 PM to 8:00 PM). Pump operation time records show: Pump 1 operated for 3200 hours, Pump 2 for 2800 hours, Pump 3 for 3500 hours, and Pump 4 for 2500 hours. Maintenance records show that Pump 4 underwent a major overhaul a month ago, Pumps 1 and 3 are operating well, and Pump 2 experiences slight vibration. A comprehensive evaluation yields the following priority sequence: Pump 4 > Pump 1 > Pump 3 > Pump 2. During off-peak periods, the load is approximately 25% of the design load. Based on the target differential temperature difference (4.5℃ for chilled water), the required flow rate is calculated to be approximately 220 m³ / h. Only one pump needs to operate to meet this demand; therefore, pump number 4, with the highest priority, is selected to run, while the other pumps remain on standby. During off-peak periods, the load is approximately 50% of the design load, with a target temperature difference of 6℃. The required flow rate is calculated to be approximately 370 m³ / h, requiring two pumps to operate. Pumps number 4 and 1 are selected, handling 55% and 45% of the load respectively. During peak periods, the load reaches 80% of the design load, with a target temperature difference of 7.5℃. The required flow rate is approximately 480 m³ / h, requiring two pumps to operate at full capacity. Pumps number 4 and 1 are selected to run at full speed. Integrating these configurations, a complete 24-hour operation schedule is created, including detailed instructions such as pump number 4 continuing to run while pump number 1 starts at 7:00 AM, both pumps increasing their speed to maximum at 2:00 PM, and both pumps decreasing their speed at 8:00 PM. Finally, pump group control instructions are generated to achieve efficient and energy-saving operation around the clock.

[0108] In one specific embodiment, the process of executing step S106 may specifically include the following steps:

[0109] The actual system operation data after the water pump group control command is executed is collected by the remote monitoring unit, compared with the expected control target, and the temperature difference deviation value and energy consumption deviation value are calculated.

[0110] Based on the temperature difference deviation and energy consumption deviation, the control effect is quantitatively scored, and a control quality evaluation index system is established.

[0111] By conducting trend analysis on various parameters in the control quality assessment index system, weak links in control are identified, and a list of optimization directions is formed.

[0112] For the weak links in the optimization direction list, the corresponding control parameters are extracted, the correction coefficients are calculated, and the parameter correction table is obtained.

[0113] Based on the parameter correction table, the load prediction model, the pump frequency conversion control curve, and the temperature difference setpoint are finely adjusted to form an improved control scheme.

[0114] The improved control scheme is distributed to the field controller through the remote management platform, the control logic and set values ​​are updated, and the optimized energy-saving control strategy is obtained.

[0115] Specifically, remote monitoring units are used to collect actual system operation data after the execution of water pump group control commands. These remote monitoring units are collections of data acquisition devices and transmission modules installed at key points in the cooling system, transmitting field data to a central analysis platform in real time via an industrial communication network. The collected data includes parameters such as chilled water inlet and outlet temperatures, cooling water inlet and outlet temperatures, water pump operating frequency, actual flow rate, and power consumption. The collected actual operation data is compared with the expected control targets, which are the anticipated operating states the system should achieve based on the previously established control strategy, including target temperature difference and expected energy consumption. Temperature difference deviation and energy consumption deviation are calculated through this comparison. The temperature difference deviation is the difference between the actual temperature difference and the target temperature difference, and the energy consumption deviation is the difference between the actual energy consumption and the expected energy consumption. These two deviations are the basic indicators for evaluating control effectiveness. Based on the calculated temperature difference deviation and energy consumption deviation values, the control effectiveness is quantitatively scored, establishing a control quality evaluation index system. The quantitative scoring uses a weighted scoring method, assigning different weights to each deviation indicator according to its importance, and calculating a weighted total score. The system's quality control assessment index system comprises multiple dimensions of evaluation indicators, primarily categorized into three main types: stability indicators, accuracy indicators, and efficiency indicators. Stability indicators assess the smoothness of system operation, such as temperature fluctuation amplitude and pump frequency change rate. Accuracy indicators assess the closeness of actual operating parameters to target parameters, such as temperature difference achievement rate and flow control accuracy. Efficiency indicators assess the system's energy utilization efficiency, such as unit cooling energy consumption and pump energy efficiency ratio. Each indicator is scored on a 5-point scale; higher scores indicate better control performance in that area. The final scores are compiled into a system control quality assessment report.

[0116] By conducting trend analysis on various parameters in the control quality assessment indicator system, weak links in control are identified. Trend analysis involves performing time series analysis on scoring data over a period of time to observe the changing trends of each indicator. Analysis methods include moving averages and exponential smoothing, which can filter out the impact of short-term fluctuations and reveal the long-term trends of the indicators. If the score of a certain indicator continues to decline or is significantly lower than other indicators, it is identified as a weak link in control. The identified weak links are ranked according to their degree of impact, forming an optimization direction list. This list records the control parameters that need to be optimized and their priorities.

[0117] For the weak points identified in the optimization direction list, the corresponding control parameters are extracted, and correction coefficients are calculated. Control parameters refer to adjustable factors affecting system performance, such as the weighting coefficients of the load prediction model, the slope parameter of the pump frequency converter control curve, and the temperature difference setpoint. The correction coefficients are calculated based on the relationship between the deviation value and the target value, using the proportional-integral method, considering both the magnitude of the current deviation and the cumulative effect of historical deviations. The calculated correction coefficients form a parameter calibration table, which records each control parameter that needs adjustment and its correction coefficient.

[0118] Based on the parameter calibration table, the load prediction model, the pump frequency converter control curve, and the temperature difference setpoint were fine-tuned to form an improved control scheme. The fine-tuning of the load prediction model mainly involved adjusting the weights of the timing mode characteristics and the temperature correction term to make the prediction results more consistent with actual load variations. The fine-tuning of the pump frequency converter control curve involved adjusting the parameters of the frequency-flow correspondence to optimize the pump's operating efficiency under different loads. The fine-tuning of the temperature difference setpoint involved adjusting the target temperature difference value for different load ranges based on actual operating results, maximizing energy savings while ensuring heat exchange efficiency. These fine-tuning adjustments resulted in an improved control scheme that incorporates entirely new parameter settings and control strategies.

[0119] The improved control scheme is distributed to the field controllers via a remote management platform, updating the control logic and setpoints. The remote management platform is a central software system used to manage and control the cooling system, possessing functions such as remote parameter configuration, command issuance, and status monitoring. The distribution process employs a phased strategy, first testing the effects of the new parameters on a small scale during non-critical periods, and then applying them comprehensively after confirmation. Upon receiving the new control parameters, the field controllers update their internal control logic and setpoints, executing the optimized energy-saving control strategy. The entire process forms a complete closed-loop control system, continuously improving the operating efficiency of the cooling system through continuous monitoring, evaluation, and optimization.

[0120] Taking the central air conditioning system of a commercial center as an example, this system implemented an energy-saving control method for cooling devices based on remote analysis. After one operating cycle, data collected by the remote monitoring unit showed that the actual temperature difference of the chilled water was 5.2℃, while the target temperature difference was 6.0℃, resulting in a temperature difference deviation of -0.8℃; the actual power consumption of the water pump was 65kW, while the expected power consumption was 60kW, resulting in an energy consumption deviation of +5kW. Based on these deviation values, temperature difference control and energy consumption control were scored separately. The temperature difference control scored 3.5 points (out of 5), and the energy consumption control scored 3.2 points. The energy consumption control, with its lower overall score, was identified as a weak link. Trend analysis revealed that the energy consumption deviation increased significantly during periods of large load fluctuations, indicating that the current water pump control algorithm was not sensitive enough to load changes. Based on this analysis, the response sensitivity parameter was extracted from the water pump frequency converter control curve, and a correction coefficient of 1.2 was calculated, indicating a need to improve the response sensitivity. Simultaneously, the temperature difference control parameter also needed fine-tuning, with a correction coefficient of 1.15, indicating a need to slightly increase the target temperature difference value. Based on these correction factors, the control parameters were adjusted: the response time in the pump frequency conversion control algorithm was shortened from 5 minutes to 4 minutes, and the target value for the chilled water temperature difference in the medium load range was adjusted from 6.0℃ to 6.2℃. These adjustments formed a new control scheme, which was distributed to the field controllers through the remote management platform, achieving continuous optimization of the energy-saving control strategy and improving system operating efficiency.

[0121] The above describes the energy-saving control method for a cooling device based on remote analysis in the embodiments of this application. The following describes the energy-saving control system for a cooling device based on remote analysis in the embodiments of this application. Please refer to [link / reference]. Figure 2 One embodiment of the energy-saving control system for a cooling device based on remote analysis in this application includes:

[0122] The acquisition module is used to acquire real-time operating parameters by installing temperature sensors, flow sensors and power monitoring devices in the chilled water pump and cooling water pump system, and to perform data filtering and preprocessing on the operating parameters to obtain cooling system operating status data.

[0123] The prediction module is used to perform time-series analysis and trend prediction of the cooling load based on the cooling system operating status data, combined with historical load records and ambient temperature changes, to obtain the predicted value of the cooling system load.

[0124] The calculation module is used to calculate and optimize the pump speed and flow parameters based on the predicted load value of the cooling system and the current pump operating efficiency, so as to obtain the frequency conversion regulation control parameters.

[0125] The setting module is used to set the temperature difference between chilled water and cooling water in segments based on the predicted load value of the cooling system and the frequency conversion regulation control parameters, by establishing temperature difference control standards for different load ranges, and to obtain differentiated temperature difference target values.

[0126] The coordination module is used to perform coordinated calculations on the start-stop status and load distribution of multiple water pumps based on the differentiated temperature difference target value and the cooling system load prediction value, so as to obtain water pump group control commands.

[0127] The calibration module is used to evaluate the control effect and correct the parameters in real time based on the water pump group control command and the cooling system operating status data, so as to obtain an optimized energy-saving control strategy.

[0128] Through the collaborative efforts of the aforementioned components, real-time operating parameters are acquired by installing a sensor network in the chilled water pump and cooling water pump systems. This data is then filtered and preprocessed to obtain high-quality cooling system operating status data, providing a reliable data foundation for subsequent intelligent control. This effectively solves the problems of incomplete data acquisition and unstable data quality in traditional control methods. By combining historical load records and ambient temperature changes for time-series analysis and trend prediction of the cooling load, an adaptive prediction model is constructed. This enables the system to anticipate load change trends, shifting from passive response to proactive prediction, significantly improving the foresight and accuracy of control. Based on the predicted load value and pump operating efficiency, the pump speed and flow parameters are optimized and calculated to generate variable frequency control parameters. It achieves precise matching of pump operating conditions, avoiding energy waste caused by traditional fixed-frequency operation; it establishes temperature difference control standards for different load ranges based on load prediction values ​​and frequency conversion adjustment parameters, realizing segmented setting of temperature difference between chilled water and cooling water, enabling the system to maintain the optimal temperature difference under different load conditions, overcoming the defect of excessively small temperature difference during low-load operation in traditional control; by coordinating the start-stop status and load distribution of multiple pumps, it obtains pump group control commands, realizing the overall optimization of the multi-pump system and avoiding overload or inefficient operation of a single pump; based on the pump group control commands and system operating status data, it performs real-time evaluation and parameter correction of the control effect, forming a closed-loop feedback mechanism, enabling the control strategy to continuously self-optimize and adapt to changes in system performance and external environment. Of particular note is the exceptional value of the time-series analysis and trend prediction algorithms employed in this solution, particularly in the field of cooling device control. These algorithms can identify periodic patterns in load changes and temperature-sensitive characteristics. Through in-depth analysis of historical data, they establish accurate load prediction models, providing a basis for decision-making in temperature difference control and water pump regulation. Simultaneously, the water pump collaborative optimization algorithm considers multiple objectives such as balanced equipment operating time and optimal energy efficiency, minimizing energy consumption while ensuring cooling performance. The organic combination of these technical features enables this invention to intelligently adjust system operating parameters according to actual load demands, significantly reducing energy consumption while maintaining cooling performance. Compared to traditional control methods, it can reduce water pump system energy consumption by 15%-30%, with particularly significant energy savings under partial load conditions. Furthermore, it extends equipment lifespan, reduces maintenance costs, and yields significant economic and environmental benefits.

[0129] Reference Figure 3 This invention also provides a computer device, which can be a server, and its internal structure can be as follows: Figure 3As shown, the computer device includes a processor, memory, display screen, input device, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores the data corresponding to this embodiment. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements the above-described method.

[0130] Those skilled in the art will understand that Figure 3 The structures shown are merely block diagrams of some structures related to the present invention and do not constitute a limitation on the computer devices on which the present invention is applied.

[0131] An embodiment of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method. It is understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0132] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A remote analysis-based cooling device energy saving control method, characterized by, The cooling device energy-saving control method based on remote analysis comprises: Real-time operation parameters are obtained by installing temperature sensors, flow sensors and power monitoring devices on the chilled water pump and cooling water pump system, data filtering and preprocessing are performed on the operation parameters, and cooling system operation state data is obtained; According to the cooling system operation state data, historical load records and environmental temperature changes, time series analysis and trend prediction are performed on the cooling load, and a cooling system load prediction value is obtained; According to the cooling system load prediction value and the current water pump operation efficiency, the water pump rotating speed and flow parameters are calculated and optimized, and variable frequency regulation control parameters are obtained; Based on the cooling system load prediction value and the variable frequency regulation control parameters, by establishing temperature difference control standards in different load intervals, the chilled water and cooling water temperature difference is set segmentally, and a differentiated temperature difference target value is obtained; According to the differentiated temperature difference target value and the cooling system load prediction value, the start-stop state and load distribution of multiple water pumps are calculated cooperatively, and water pump group control instructions are obtained; According to the water pump group control instructions and the cooling system operation state data, the control effect is evaluated and parameter corrected in real time, and an optimized energy-saving control strategy is obtained.

2. The remote analytics based cooling plant energy saving control method according to claim 1, wherein, The real-time operation parameters are obtained by installing temperature sensors, flow sensors and power monitoring devices on the chilled water pump and cooling water pump system, data filtering and preprocessing are performed on the operation parameters, and cooling system operation state data is obtained, comprising: Water temperature data is collected by temperature sensors at the inlet and outlet of chilled water and cooling water, water flow data is measured by flow sensors on the main pipeline, and water pump current and power consumption data are recorded by power monitoring devices to obtain cooling system original operation parameters; The original operation parameters are divided into time series according to the sampling frequency of each minute, and abnormal values are removed by amplitude limiting filtering method to obtain an effective operation parameter matrix; Temperature difference, flow coefficient and energy efficiency ratio parameters are extracted from the effective operation parameter matrix, and instantaneous operation load index is calculated; The instantaneous operation load index is compared with the design load to determine the real-time load rate and temperature difference utilization rate, and a standardized parameter set is generated; A multi-dimensional operation state vector is constructed using the standardized parameter set, and operation state statistical characteristics are calculated by time domain feature extraction to form cooling system operation state data.

3. The remote analytics based cooling plant energy saving control method according to claim 1, wherein, The cooling system operation state data is combined with historical load records and environmental temperature changes to perform time series analysis and trend prediction on the cooling load, and a cooling system load prediction value is obtained, comprising: Cooling system operation state data of nearly 90 days is extracted from the data storage unit, and seasonal decomposition is performed according to working day and non-working day types to form a load feature library; The cooling system operation state data and environmental temperature data from the meteorological data center are time-aligned and data-fused to construct a temperature load correlation matrix; Periodic analysis is performed on the load feature library by the autoregressive moving average method, the cooling load variation law within a day, a week and a month is extracted, and time series mode features are generated; Temperature sensitivity coefficients are calculated according to the temperature load correlation matrix, and temperature correction terms are established combining with 24-hour temperature forecast data; The temperature correction term is applied to the time sequence pattern feature, and future period prediction load values are calculated through weight recursion; The future period prediction load values are smoothed and subjected to abnormal constraint to obtain cooling system load prediction values.

4. The remote analytics based cooling plant energy saving control method according to claim 1, wherein, According to the cooling system load prediction values and current water pump operation efficiency, water pump rotating speed and flow parameters are calculated and optimized to obtain variable frequency regulation control parameters, including: The water pump efficiency distribution graph under the current working condition is extracted from the water pump characteristic curve database, compared with the measured power data, and the water pump efficiency deviation value is calculated; According to the cooling system load prediction values, the theoretical water flow value required by the system is determined, and the corresponding pressure head requirement is calculated in combination with the pipe network resistance characteristic curve; The efficiency deviation value is used to correct the water pump characteristic curve, and a real-time water pump performance evaluation index is constructed; For the theoretical water flow value, the required water pump rotating speed range is calculated through the flow-frequency corresponding relationship formula to form a frequency preliminary set; The energy consumption values of each rotating speed point in the frequency preliminary set are quantitatively evaluated, and the rotating speed value corresponding to the lowest energy consumption is selected through the energy consumption-flow optimization algorithm; The rotating speed value corresponding to the lowest energy consumption is converted into a variable frequency converter control signal parameter, and the variable frequency regulation control parameter is generated in combination with the start-up curve and buffer time setting.

5. The remote analytics based cooling plant energy saving control method according to claim 1, wherein, Based on the cooling system load prediction values and the variable frequency regulation control parameters, the temperature difference control standards of different load intervals are established, the temperature difference between chilled water and cooling water is set in sections, and differential temperature difference target values are obtained, including: The load rate of the cooling system load prediction value is calculated, and the load rate is divided into three interval segments, i.e., a low load interval, a medium load interval and a high load interval; By analyzing the temperature difference distribution of each load interval in the historical operation data, the temperature difference optimization range under the typical working condition is extracted to form a temperature difference-load correlation data set; For the low load interval, the chilled water temperature difference reference value is determined to be 4-5℃, the cooling water temperature difference reference value is determined to be 3-4℃, and the low load temperature difference control reference value is generated; For the medium load interval, the chilled water temperature difference reference value is determined to be 5-7℃, the cooling water temperature difference reference value is determined to be 4-6℃, and the medium load temperature difference control reference value is generated; For the high load interval, the chilled water temperature difference reference value is determined to be 7-8℃, the cooling water temperature difference reference value is determined to be 6-7℃, and the high load temperature difference control reference value is generated; The variable frequency regulation control parameters are used to correct and calculate the temperature difference control reference values of each load interval, and a transition smoothing coefficient is set in combination with the load change trend to obtain differential temperature difference target values.

6. The remote analytics based cooling plant energy saving control method according to claim 1, wherein, According to the differential temperature difference target values and the cooling system load prediction values, the start-stop state and load distribution of multiple water pumps are calculated cooperatively to obtain water pump group control instructions, including: The cooling system load prediction values are divided into time intervals, the load demand is classified according to the off-peak period, the flat peak period and the peak period, and a load time interval distribution graph is formed; The water pump availability is evaluated according to the running time cumulative value and the maintenance state record of each water pump, and a water pump priority sequence table is established; The total flow demand under different load conditions is calculated by using the differential temperature target value, and the minimum number of pumps to be started is determined in combination with the single pump flow characteristics; For the low-load period, 1-2 pumps ranked at the top of the pump priority list are selected to run, and the remaining pumps are set to standby state, thereby generating a low-load period pump configuration scheme; For the flat peak and peak load period, the load sharing ratio of each pump is calculated according to the principle of balanced distribution, thereby generating a flat peak and peak period pump configuration scheme; The low-load period pump configuration scheme and the flat peak and peak period pump configuration scheme are integrated into a 24-hour pump operation schedule, and start-stop timing and speed control information are added to obtain a pump group control instruction.

7. The remote analytics based cooling plant energy saving control method according to claim 1, wherein, The control effect is evaluated and the parameters are corrected in real time according to the pump group control instruction and the cooling system running state data, and an optimized energy-saving control strategy is obtained, including: The system actual running data after the pump group control instruction is executed is collected by the remote monitoring unit, compared with the expected control target, and the temperature difference deviation value and the energy consumption deviation value are calculated; The control effect is quantitatively scored according to the temperature difference deviation value and the energy consumption deviation value, and a control quality evaluation index system is established; The weak links in the control are identified by trend analysis of each parameter in the control quality evaluation index system, and an optimization direction list is formed; For the weak links in the optimization direction list, the corresponding control parameters are extracted, the correction coefficient is calculated, and a parameter correction table is obtained; The load prediction model, the pump frequency conversion control curve and the temperature difference setting value are fine-tuned according to the parameter correction table, and an improved control scheme is formed; The improved control scheme is issued to the field controller through the remote management platform to update the control logic and the setting value, and an optimized energy-saving control strategy is obtained.

8. A remote analysis-based cooling device energy-saving control system for implementing the remote analysis-based cooling device energy-saving control method according to any one of claims 1 to 7, characterized by, The cooling device energy-saving control system based on remote analysis comprises: An acquisition module is configured to acquire real-time running parameters by installing temperature sensors, flow sensors and power monitoring devices in the chilled water pump and cooling water pump system, perform data filtering and preprocessing on the running parameters, and obtain cooling system running state data; A prediction module is configured to perform time series analysis and trend prediction on the cooling load according to the cooling system running state data in combination with historical load records and environmental temperature changes, and obtain a cooling system load prediction value; A calculation module is configured to calculate and optimize pump speed and flow parameters according to the cooling system load prediction value and current pump running efficiency, and obtain frequency regulation control parameters; A setting module is configured to set chilled water and cooling water temperature difference in different load intervals by establishing temperature difference control standards based on the cooling system load prediction value and frequency regulation control parameters, and obtain differential temperature target values; A coordination module is configured to perform coordinated calculation on the start-stop state and load distribution of multiple pumps according to the differential temperature target values and the cooling system load prediction value, and obtain a pump group control instruction; A correction module is configured to evaluate and correct the control effect in real time according to the pump group control instruction and the cooling system running state data, and obtain an optimized energy-saving control strategy.

9. A computer device, comprising: An apparatus comprising a memory storing a computer program executable on a processor, wherein the processor implements the remote analysis based cooling device energy saving control method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, A computer program product, wherein the computer program product stores a computer program, and the computer program causes a processor to execute the remote analysis based cooling device energy saving control method according to any one of claims 1 to 7 when the computer program is executed by the processor.

Images