Water chilling unit multi-unit load allocation system and method

Abstract

The invention discloses a multi-unit load allocation system and method for a water chilling unit, and relates to the technical field of load allocation. Temperature and flow data are collected in real time, and a valve is dynamically adjusted to maintain the set temperature; the standardized energy efficiency coefficient is calculated based on the time sequence, the running number of the water chiller is intelligently regulated and controlled by analyzing the trend of deviating from the optimal interval, and it is ensured that the system is in an efficient running state all the time. According to the method, the load is accurately allocated through the dynamic energy efficiency coefficient, an adaptive energy efficiency reference is established based on machine learning, and advanced intervention is realized by utilizing trend prediction, so that load switching is smoothed, and system oscillation is reduced; through a double-loop mechanism integrating temperature control and energy efficiency optimization, system-level energy saving is achieved while the environmental comfort is ensured, and the energy efficiency bottleneck of traditional load equalization is effectively broken through.

Description

Technical Field

[0001] This invention relates to the field of load distribution technology, specifically to a multi-unit load distribution system and method for chillers. Background Technology

[0002] In large public buildings and industrial refrigeration scenarios, parallel operation of multiple chiller units is a common practice to deal with cooling load fluctuations. Traditional group control systems usually adopt a simple average distribution mode, that is, the total demand at the terminal is proportionally distributed to each unit, so that all chillers can increase or decrease the load synchronously.

[0003] For example, Chinese Patent Publication No. CN119022413A provides a method and apparatus for load allocation of chiller units. This includes acquiring target operating condition data for multiple chiller units at various operating times associated with the current operating time; calculating the corresponding cooling load based on each target operating condition data, and constructing a total cooling load sequence based on each cooling load according to each operating time; decomposing and reconstructing the total cooling load sequence using a singular spectrum analysis algorithm to determine the reconstructed total cooling load sequence and noise components; predicting the load based on the reconstructed total cooling load sequence using a preset total cooling load prediction model, and outputting the predicted total cooling load in conjunction with the noise components; and outputting the load rate of each chiller unit based on the target operating condition data, cooling load, and predicted total cooling load at the current operating time using a preset load allocation control reinforcement learning model. This process improves the flexibility of load allocation for chiller units.

[0004] However, this passive allocation method ignores the differences in operating characteristics of different units. When the overall demand is low, all chillers will fall into a low-load operation state, which will prevent some chiller units from fully utilizing their design efficiency and result in overall energy waste. Summary of the Invention

[0005] Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a multi-unit load allocation system and method for chillers, which solves the problem of the proportional load allocation strategy used in traditional chiller group control causing the units to deviate from their efficient operating range.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for load allocation of multiple chiller units, comprising the following specific steps: Step 1: Real-time acquisition of terminal temperature values ​​and water flow rates; Step 2: Setting a terminal set temperature value and calculating the temperature difference value with the terminal real-time temperature value, then deflecting the terminal valve according to the temperature difference value to allow chilled water produced by the chiller to flow into the terminal, thereby reducing the surrounding temperature of the terminal and making the temperature difference value zero; Step 3: Under time series conditions, comprehensively calculating the water flow rate value and temperature difference value flowing into the terminal and normalizing them to obtain the actual energy efficiency coefficient; setting an optimal energy efficiency coefficient range, controlling the actual energy efficiency coefficient of the chiller within the optimal energy efficiency coefficient range, analyzing the deviation trend of the actual energy efficiency coefficient within the optimal energy efficiency coefficient range, and adjusting the number of chillers in operation to ensure that the actual energy efficiency coefficient of the chiller is always within the optimal energy efficiency coefficient range.

[0007] Furthermore, the specific method for obtaining the temperature difference value is as follows: calculate the difference between the terminal set temperature value and the terminal real-time temperature value to obtain the temperature difference value.

[0008] Furthermore, the specific method for obtaining the actual energy efficiency coefficient is as follows: a preset detection time period is used, and the water flow rate at the inflow end is calculated to obtain the water flow rate coefficient. The temperature difference value is calculated to obtain the heat transfer coefficient at the inflow end. The water flow rate coefficient and the heat transfer coefficient at the inflow end are calculated together and normalized to obtain the actual energy efficiency coefficient. ;in, Indicates the actual energy efficiency coefficient. Indicates the water quantity coefficient. This represents the heat transfer coefficient at the terminal.

[0009] Furthermore, the water volume coefficient is obtained by integrating the detection time period with the water flow rate at the inflow end to obtain the water volume coefficient.

[0010] Furthermore, the specific method for obtaining the heat transfer coefficient at the terminal is as follows: the temperature difference value and the water volume coefficient are calculated together and normalized to obtain the heat transfer coefficient at the terminal. ;in, This represents the heat transfer coefficient at the terminal. Indicates the temperature difference value. Indicates the water quantity coefficient. It represents a positive real number.

[0011] Furthermore, the specific method for setting the optimal energy efficiency coefficient range is as follows: the water volume coefficient and the heat transfer coefficient at the end point accumulated in long-term experiments at the same moment are used as input features and input into the clustering algorithm. Then, combined with the actual operation log of the chiller and expert knowledge, the data clusters corresponding to the optimal energy efficiency state are identified. Finally, by statistically analyzing the distribution boundaries of all actual energy efficiency coefficients within this data cluster, a quantified optimal energy efficiency coefficient range is determined.

[0012] Furthermore, the specific method for analyzing the deviation trend of the actual energy efficiency coefficient within the optimal energy efficiency coefficient range in step three is as follows: set a refresh count threshold, determine the midpoint of the optimal energy efficiency coefficient range, and record it as the median value. Within the refresh count threshold, analyze whether the actual energy efficiency coefficient deviates from the median value. During this process, the actual energy efficiency coefficient is within the optimal energy efficiency coefficient range.

[0013] Furthermore, the specific method for analyzing whether the actual energy efficiency coefficient deviates from the median value is as follows: calculate the difference between the median value and the actual energy efficiency coefficient to obtain the deviation value. The deviation value includes positive deviation value and negative deviation value. If the positive deviation value gradually increases, the actual energy efficiency coefficient gradually deviates from the median value; if the negative deviation value gradually increases, the actual energy efficiency coefficient gradually exceeds the median value; if neither the positive nor the negative deviation value gradually increases, no intervention is required.

[0014] Furthermore, the specific method for adjusting the number of operating chillers in step three is as follows: if the positive deviation value gradually increases, the number of chillers connected in parallel is gradually increased until the positive deviation value of each chiller no longer gradually increases; if the negative deviation value gradually increases, the number of chillers connected in parallel is gradually decreased until the negative deviation value of each chiller no longer gradually increases.

[0015] A multi-unit load distribution system for chillers includes the following modules: a data acquisition module, a temperature feedback control module, and an energy efficiency optimization and decision-making module. The data acquisition module is used to collect real-time terminal temperature values ​​and inflow water flow values. The temperature feedback control module sets a terminal set temperature value and a valve deflection rule library. It calculates the temperature difference between the terminal set temperature value and the terminal real-time temperature value, analyzes whether there is a deviation between the terminal set temperature value and the terminal real-time temperature value, and if no deviation is found, executes the energy efficiency optimization and decision-making module; if a deviation is found, it deflects the valve angle according to the valve deflection rule library to make the terminal set temperature value equal to the terminal real-time temperature value, and then returns the result. The data acquisition module returns to the energy efficiency optimization and decision-making module. Under time series conditions, it comprehensively calculates and normalizes the water flow rate and temperature difference at the inflow end to obtain the actual energy efficiency coefficient (EEC). It sets an optimal EEC range, controlling the chiller's EEC within this range. It analyzes the trend of the EEC's deviation within the optimal range. If the analysis shows no deviation, it returns to the data acquisition module or terminates directly. If the analysis shows a deviation, it adjusts the number of chillers in operation to ensure the chiller's EEC remains within the optimal range, and continues to analyze the trend of the EEC's deviation within the optimal range.

[0016] Beneficial effects Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: 1. By introducing a dynamic evaluation index of actual energy efficiency coefficient, the operating efficiency of each unit is quantified in real time. Instead of evenly distributing the load, the system aims to ensure that the actual energy efficiency coefficient of all operating units is within the optimal energy efficiency range. This allows the system to accurately start, stop, or allocate the number of chillers based on real-time terminal demand and the performance of the units themselves, ensuring that each operating unit works at its high-efficiency load point. This avoids energy waste caused by unreasonable allocation from the perspective of system architecture.

[0017] 2. By using machine learning clustering algorithms, the system autonomously mines and establishes the optimal energy efficiency coefficient range from long-term operating data. This range is a summary of the best state that the system has actually achieved in its historical operation, and it can be updated through continuous learning. This makes the system's energy efficiency benchmark adaptive, which can fit the actual performance degradation of chillers in a specific computer room, seasonal changes and other realistic factors, and ensure that the optimization target is always the best energy efficiency that can be achieved under the current conditions, rather than a theoretical value that is detached from reality.

[0018] 3. By monitoring whether the deviation of the actual energy efficiency coefficient from the midpoint of the optimal range gradually increases, the system can predict the trend of energy efficiency deterioration in advance. This means that before load changes cause a significant drop in energy efficiency, the system can take measures to add or remove units in advance, predict in advance, smooth the load switching process, reduce system oscillation, and achieve more stable and energy-efficient operation.

[0019] 4. By organically integrating the traditional temperature control loop, namely the temperature feedback control module, with the advanced energy efficiency optimization loop, namely the energy efficiency optimization and decision-making module, the temperature control module ensures that the cooling needs of the terminal are met and the environmental comfort is maintained; on this basis, the energy efficiency optimization module determines the number of chillers to operate with the goal of maximizing the overall energy efficiency of the system. The two modules work together to pursue operational economy while ensuring stable temperature control.

[0020] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0021] Figure 1 This invention provides a flowchart of a method for load allocation among multiple chiller units.

[0022] Figure 2 This invention relates to a structural diagram of a multi-unit load distribution system for chillers. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0025] Example 1: like Figure 1 As shown, this embodiment of the invention provides a method for load allocation among multiple chiller units, including the following specific steps: Step 1: By placing temperature sensors at the terminal, the real-time temperature value of the terminal is collected in real time to achieve on-demand cooling and improve comfort. For example, the terminal is an air conditioning unit that circulates cold water. The water flow rate value flowing into the terminal is collected in real time by ultrasonic flow meters placed in the terminal. The real-time temperature value of the terminal and the water flow rate value flowing into the terminal need to be initialized after the system is started to clear historical residual data and avoid misreading at startup, which may lead to control malfunction.

[0026] Step Two: Set the terminal set temperature value and valve deflection rule library. The valve deflection rule library is used to set the deflection angle of the terminal valve based on the temperature difference between the terminal set temperature value and the terminal real-time temperature value. The deflection range of the terminal valve is 0 degrees to 90 degrees, where 0 degrees represents fully closed and 90 degrees represents fully open. The temperature difference value is calculated from the terminal set temperature value and the terminal real-time temperature value. The temperature difference value is then analyzed to determine if there is a deviation between the terminal set temperature value and the terminal real-time temperature value. This is done by comparing the temperature difference value to zero. If the temperature difference value is equal to zero, it indicates no deviation. If no deviation is found, proceed to Step Three; if the temperature difference value is not equal to zero... If the value is zero, it indicates a deviation. If a deviation is detected, the valve angle is deflected according to the valve deflection rule library. That is, the valve opening is adjusted proportionally to the size of the temperature difference. For example, a 1-degree Celsius temperature difference corresponds to a 5% change in opening, rather than a fixed value, thus achieving a rapid response. In addition, the specific ratio between the temperature difference and the corresponding valve opening can be modified through the valve deflection rule library. The valve opening at the terminal is used to control the flow rate of chilled water from the chiller into the terminal. That is, after the chilled water is cooled in the chiller, it is pumped to each terminal to exchange heat with the indoor air. The return water after releasing the cooling capacity flows back to the chiller for cooling, and so on, so that the terminal set temperature value is equal to the terminal real-time temperature value, and then returns to step one.

[0027] Step 3: Under time series conditions, the water flow rate and temperature difference at the inflow end are calculated and normalized to obtain the actual energy efficiency coefficient (EEC). An optimal EEC range is set, and the actual EEC of the chiller is controlled within this range. The trend of the actual EEC's deviation within the optimal EEC range is analyzed. If the analysis shows no deviation in the actual EEC, return to Step 1 or end the process directly. If the analysis shows a deviation in the actual EEC, the number of chillers in operation is adjusted to ensure that the actual EEC remains within the optimal EEC range. The analysis continues based on the trend of the actual EEC's deviation within the optimal EEC range.

[0028] Example 2 differs from Example 1 in that: The specific method for obtaining the temperature difference value is as follows: The temperature difference is calculated by subtracting the set temperature value from the real-time temperature value at the terminal. ; in, Indicates the temperature difference value. This indicates the real-time temperature value at the terminal. This indicates the terminal set temperature value. When the temperature difference value is greater than zero, that is, the terminal real-time temperature value is greater than the terminal set temperature value, it means that cooling should be performed. When the temperature difference value is less than zero, that is, the terminal real-time temperature value is less than the terminal set temperature value, it means that passive reheating should be performed, which reduces or stops the flow of chilled water from the chiller into the terminal.

[0029] The specific methods for obtaining the actual energy efficiency coefficient are as follows: The preset detection time period, in seconds, includes the start and end times and is assigned a value of 1, meaning the length from the start to the end time is 1 second. This detection time period can be modified, for example, changing 1 second to 3 seconds, representing the actual energy efficiency coefficient (EEC) change of the chiller every 3 seconds. However, subsequent analysis typically uses the actual EEC change every 1 second. The water flow rate is calculated by combining the detection time period with the inflow water flow rate at the end point to obtain the water flow coefficient. The temperature difference is then calculated with the inflow water flow rate at the end point to obtain the heat transfer coefficient at the end point. Finally, the water flow coefficient and the heat transfer coefficient at the end point are combined for further calculations. The system then performs normalization processing, specifically determining the minimum and maximum values ​​of the water flow coefficient and the heat transfer coefficient at the terminal based on historical operating data. Then, using the Min-Max normalization method, it calculates the normalized values ​​of both the water flow coefficient and the heat transfer coefficient. The normalized value of the water flow coefficient is equal to the water flow coefficient minus its historical minimum value, divided by the difference between the historical maximum and minimum values. Similarly, the normalized value of the heat transfer coefficient is equal to the heat transfer coefficient minus its historical minimum value, divided by the difference between the historical maximum and minimum values. Finally, the normalized water flow coefficient and the normalized heat transfer coefficient are substituted into the actual energy efficiency coefficient calculation formula. This eliminates dimensional differences while achieving a comprehensive calculation of the water flow coefficient and the heat transfer coefficient, yielding the actual energy efficiency coefficient. ; in, This represents the actual energy efficiency coefficient, reflecting the real-time energy efficiency of the chiller. Indicates the water quantity coefficient. This represents the heat transfer coefficient at the terminal, and the heat transfer coefficient at the terminal is not zero when the terminal is started.

[0030] The specific method for obtaining the water quantity coefficient is as follows: The water volume coefficient is obtained by integrating the detection time period with the water flow rate at the inflow end, where the detection time period is a continuous time series. ; in, This represents the water flow coefficient, reflecting the water flow rate into the terminal during the monitoring period. Indicates the start time of the detection period. Indicates the end time of the detection period. This represents the water flow rate at each moment during the detection period. If the water flow rate at each moment during the detection period is zero, the water flow coefficient is also zero, which in turn makes the actual energy efficiency coefficient zero. In this case, the actual energy efficiency coefficient is the smallest, meaning the chiller is the least efficient.

[0031] The specific method for obtaining the heat transfer coefficient at the terminal is as follows: The temperature difference value and water volume coefficient are calculated together and normalized. Specifically, the minimum and maximum values ​​of the water volume coefficient and temperature difference value are determined based on historical operating data. Then, the Min-Max normalization method is used to calculate the normalized values ​​of the water volume coefficient and the temperature difference value respectively. The normalized value of the water volume coefficient is equal to the water volume coefficient minus its historical minimum value and divided by the difference between the historical maximum and minimum values. The normalized value of the temperature difference is equal to the temperature difference value minus its historical minimum value and divided by the difference between the historical maximum and minimum values. Finally, the normalized water volume coefficient and the normalized temperature difference value are substituted into the formula for calculating the heat transfer coefficient at the terminal. This eliminates dimensional differences and realizes the comprehensive calculation of the water volume coefficient and temperature difference value to obtain the heat transfer coefficient at the terminal. ; in, This represents the heat transfer coefficient at the terminal. Indicates the temperature difference value. Indicates the water quantity coefficient. This represents a positive real number to avoid the situation where the heat transfer coefficient at the terminal is meaningless when the flow rate coefficient is zero. If the flow rate coefficient is zero, the actual energy efficiency coefficient is zero, meaning no cold water flows into the terminal, and therefore the actual energy efficiency coefficient is also zero. If the flow rate coefficient is positive and constant, and the temperature difference is negative, the heat transfer coefficient at the terminal is also negative, meaning the real-time temperature at the terminal is lower than the set temperature, resulting in low heat transfer efficiency. When the heat transfer coefficient at the terminal is less than zero, it means the actual energy efficiency coefficient is low. If the temperature difference is zero, the heat transfer coefficient at the terminal is also zero, meaning the real-time temperature at the terminal is equal to the set temperature, and the heat transfer efficiency is the flow rate coefficient, indicating a moderate actual energy efficiency coefficient. If the temperature difference is positive, then... A value less than one indicates that the actual energy efficiency coefficient is greater than the water flow coefficient, meaning the actual energy efficiency coefficient is large. However, in reality, the water flow coefficient and the temperature difference are constantly changing, so the above assumption does not hold true. Therefore, it is necessary to conduct experiments using the water flow coefficient and the heat transfer coefficient at the terminal to obtain the optimal energy efficiency coefficient range. Then, the efficiency of the chiller can be analyzed by checking whether the actual energy efficiency coefficient is within the optimal energy efficiency coefficient range.

[0032] The specific method for setting the optimal energy efficiency coefficient range is as follows: The water volume coefficient and the heat transfer coefficient at the terminal end, accumulated from long-term experiments, are used as input features and fed into a clustering algorithm, such as K-Means, for calculation. This algorithm automatically divides the data points into several significantly different clusters based on their distribution density and distance in the feature space. Subsequently, by combining the actual operating logs of the chiller with preset objective rules for energy efficiency status, the data clusters corresponding to the optimal energy efficiency status are identified. Finally, by statistically analyzing the distribution boundaries of all actual energy efficiency coefficients within this data cluster, such as by using quantiles, a quantified optimal energy efficiency coefficient range is determined. The objective energy efficiency status judgment rule filters and identifies data clusters generated by clustering based on multiple quantifiable conditions. Specifically, it calculates the actual energy efficiency coefficient (EEC) of each cluster based on the water volume coefficient and the heat transfer coefficient at the terminal, and statistically analyzes the average and standard deviation of the EEC of each cluster. The filtering conditions are set as follows: the first condition is that the average EEC of this cluster is higher than the total average of the actual EECs of all clusters to ensure high energy efficiency; the second condition is that the standard deviation of the actual EEC of this cluster is lower than the total standard deviation of the actual EECs of all clusters to ensure strong stability. Data clusters that meet the first and second conditions are selected, and combined with the actual operation logs of the chiller, data clusters with extreme valve openings at the terminal of the corresponding unit or periods with fault alarms are removed. Finally, by statistically analyzing the distribution boundaries of all actual EECs within the selected data cluster, the smallest actual EEC within this cluster is taken as the lower limit of the interval, and the largest actual EEC within this cluster is taken as the upper limit of the interval, thus determining a quantified optimal energy efficiency coefficient interval.

[0033] The specific method for analyzing the deviation trend of the actual energy efficiency coefficient from the optimal energy efficiency coefficient range is as follows: Setting a refresh frequency threshold represents an empirical value used in industrial control to balance response speed and anti-interference capabilities. The system refreshes once per detection period to regulate the refresh frequency and determine the midpoint of the optimal energy efficiency coefficient range, denoted as the median value. This median value represents the optimal energy efficiency coefficient in the experiment, signifying the benchmark point in historical data where the theoretical energy efficiency is at its best. It is not a fixed value but an adaptive, dynamic target learned from actual system operating data. While the upper limit of the optimal energy efficiency coefficient range represents the theoretically highest energy efficiency point, any small disturbance increases the probability of the actual energy efficiency coefficient frequently crossing the upper limit boundary, causing oscillations. Using the median value as the target provides a two-way buffer space for the system. It can accommodate normal operating fluctuations, maintain control stability, and achieve predictive intervention through continuous monitoring of deviations in both positive and negative directions. This means adjusting the number of units before the actual energy efficiency deteriorates to the point of exceeding the limit, ensuring that the system balances operational reliability and overall energy consumption while pursuing high efficiency. Within the refresh frequency threshold, the system analyzes whether the actual energy efficiency coefficient deviates from the median value, ensuring that the actual energy efficiency coefficient remains within the optimal energy efficiency coefficient range during this process.

[0034] The specific method for analyzing whether the actual energy efficiency coefficient deviates from the median value is as follows: The difference between the median value and the actual energy efficiency coefficient is calculated to obtain the deviation value. The deviation value includes positive and negative deviation values. A positive deviation value indicates that the actual energy efficiency coefficient is gradually decreasing towards the median value, while a negative deviation value indicates that the actual energy efficiency coefficient is gradually increasing towards the median value. A preset single-increase threshold is set, and the system first analyzes whether the rapid response condition is met. This indicates whether the magnitude of a single increase in the deviation value exceeds the single-increase threshold. This is used to bypass the slow trend analysis process and immediately take emergency intervention measures, such as increasing or decreasing the number of chillers, in the event of sudden and drastic fluctuations in the water flow rate at the terminal due to valve malfunctions. If the condition is met... If the number of chillers connected in parallel is increased or decreased, the deviation value is analyzed until the single increase does not exceed the single increase threshold. If this is not met, the analysis continues to check if the deviation value is gradually increasing. If the positive deviation value is gradually increasing, the actual energy efficiency coefficient is gradually decreasing to less than the median value, indicating that the energy efficiency is slowly deteriorating. For example, due to the continuous increase in terminal load, the operating chiller units are overloaded and their efficiency decreases. If the negative deviation value is gradually increasing, the actual energy efficiency coefficient is gradually increasing to more than the median value, indicating that the energy efficiency is oversaturated, leading to increased energy consumption. If neither the positive nor the negative deviation value is gradually increasing, no intervention is needed.

[0035] The specific method for adjusting the number of chillers in operation is as follows: If the positive deviation value gradually increases, it indicates that the actual energy efficiency coefficient of this chiller unit is decreasing, which in turn indicates that it is more difficult to reduce the real-time temperature at the terminal. For example, an increase in the flow of people around the terminal leads to an increase in the ambient temperature, or the cooling at night makes it easier to reduce the real-time temperature at the terminal. Therefore, the number of chillers connected in parallel is increased until the positive deviation value of each chiller no longer gradually increases. That is, depending on the location of the terminal, the temperature around the terminal is slightly different. Some chillers have a constant actual energy efficiency coefficient, while others have an actual energy efficiency coefficient that gradually approaches the middle value. If the negative deviation value gradually increases, it indicates that the actual energy efficiency of this chiller unit is too saturated, wasting energy. Therefore, the number of chillers connected in parallel should be reduced until the negative deviation value of each chiller no longer gradually increases. In this case, increasing or decreasing the number of chillers connected in parallel should be done gradually.

[0036] Example 3: like Figure 2 As shown: A multi-unit load distribution system for chillers includes the following specific modules: Data acquisition module: Real-time acquisition of terminal temperature and inflow water flow. This module is the basic data input part of the system, responsible for continuously monitoring and collecting key operating parameters, supporting temperature control and energy efficiency calculation. Temperature feedback control module: This module sets the terminal setpoint temperature value and a valve deflection rule library. The valve deflection rule library stores rules relating valve opening and temperature difference to ensure control accuracy. The temperature difference is calculated by comparing the terminal setpoint temperature value with the terminal real-time temperature value. A positive temperature difference indicates a need for cooling, while a negative temperature difference indicates a need for heating or reduced cooling. The module analyzes the temperature difference to check for deviations between the terminal setpoint temperature value and the terminal real-time temperature value. If no deviation is found, the temperature control target has been achieved, and valve adjustment is unnecessary. The energy efficiency optimization and decision module then skips temperature control and directly enters the energy efficiency optimization stage to improve system efficiency. If a deviation is found, intervention is required. The valve angle is deflected according to the valve deflection rule library, and the chilled water flow rate is changed to correct the temperature, ensuring the terminal setpoint temperature value equals the terminal real-time temperature value. Once the temperature deviation is eliminated, the module returns to the data acquisition module for re-acquisition, forming a continuous monitoring cycle. Energy Efficiency Optimization and Decision Module: Under time series conditions, the inflow water flow rate and temperature difference at the terminal are comprehensively calculated and normalized to obtain the actual energy efficiency coefficient (EEC), thereby quantifying the real-time energy efficiency of the chiller and providing a basis for optimization. An optimal EEC range is set to control the chiller's actual EEC within this range. The module analyzes the deviation trend of the actual EEC within the optimal range to monitor energy efficiency changes. If the analysis shows no deviation in the actual EEC, the process returns to the data acquisition module or terminates directly. If the analysis shows a deviation in the actual EEC, i.e., an early prediction of energy efficiency deterioration, the module adjusts the number of chillers in operation—a response to energy efficiency deterioration—and redistributes the load to ensure the chiller's actual EEC remains within the optimal range. The module continues to analyze the deviation trend of the actual EEC within the optimal range for real-time monitoring and dynamic adjustment.

[0037] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for load distribution among multiple chiller units, characterized in that: The specific steps include the following: Step 1: Real-time collection of terminal temperature and inflow water flow rate; Step 2: Set the terminal set temperature value and calculate the temperature difference value with the terminal real-time temperature value. Then, deflect the terminal valve according to the temperature difference value to allow the chilled water produced by the chiller to flow into the terminal, thereby reducing the temperature around the terminal and making the temperature difference value zero. Step 3: Under time series conditions, the water flow rate and temperature difference at the inflow end are calculated together and normalized to obtain the actual energy efficiency coefficient; the optimal energy efficiency coefficient range is set, and the actual energy efficiency coefficient of the chiller is controlled within the optimal energy efficiency coefficient range. The deviation trend of the actual energy efficiency coefficient within the optimal energy efficiency coefficient range is analyzed, and the number of chillers in operation is adjusted to ensure that the actual energy efficiency coefficient of the chiller is always within the optimal energy efficiency coefficient range.

2. The method for multi-unit load distribution of chillers according to claim 1, characterized in that: The specific method for obtaining the temperature difference value is as follows: The temperature difference is calculated by subtracting the set temperature value from the real-time temperature value at the terminal.

3. The method for multi-unit load distribution of chillers according to claim 2, characterized in that: The specific method for obtaining the actual energy efficiency coefficient is as follows: The system presets a detection time period, calculates the water flow rate based on the detection time period and the water flow rate at the end of the terminal, obtains the water flow rate coefficient, calculates the temperature difference value based on the water flow rate at the end of the terminal, obtains the heat transfer coefficient at the end of the terminal, calculates the water flow rate coefficient and the heat transfer coefficient at the end of the terminal, and performs normalization processing to obtain the actual energy efficiency coefficient. ； in, Indicates the actual energy efficiency coefficient. Indicates the water quantity coefficient. This represents the heat transfer coefficient at the terminal.

4. The method for multi-unit load distribution of chillers according to claim 3, characterized in that: The specific method for obtaining the water quantity coefficient is as follows: The water volume coefficient is obtained by integrating the detection time period with the water flow rate at the inflow end.

5. A method for load distribution among multiple chiller units according to claim 4, characterized in that: The specific method for obtaining the heat transfer coefficient at the terminal is as follows: The temperature difference value and water volume coefficient are calculated together and normalized to obtain the heat transfer coefficient at the terminal. ； in, This represents the heat transfer coefficient at the terminal. Indicates the temperature difference value. Indicates the water quantity coefficient. It represents a positive real number.

6. The method for multi-unit load distribution of chillers according to claim 5, characterized in that: The specific method for setting the optimal energy efficiency coefficient range is as follows: The water volume coefficient and the heat transfer coefficient at the terminal end, accumulated from long-term experiments, are used as input features and fed into a clustering algorithm. Then, combined with the actual operation logs of the chiller and expert knowledge, the data clusters corresponding to the optimal energy efficiency state are identified. Finally, by statistically analyzing the distribution boundaries of all actual energy efficiency coefficients within this data cluster, a quantitative optimal energy efficiency coefficient range is determined.

7. A method for load distribution among multiple chiller units according to claim 6, characterized in that: The specific method for analyzing the deviation trend of the actual energy efficiency coefficient within the optimal energy efficiency coefficient range in step three is as follows: Set a refresh count threshold to determine the midpoint of the optimal energy efficiency coefficient range, denoted as the median value. Within the refresh count threshold, analyze whether the actual energy efficiency coefficient deviates from the median value. During this process, the actual energy efficiency coefficient remains within the optimal energy efficiency coefficient range.

8. A method for load distribution among multiple chiller units according to claim 7, characterized in that: The specific method for analyzing whether the actual energy efficiency coefficient deviates from the median value is as follows: The difference between the median value and the actual energy efficiency coefficient is calculated to obtain the deviation value. The deviation value includes positive deviation value and negative deviation value. If the positive deviation value gradually increases, the actual energy efficiency coefficient gradually deviates from the median value. If the negative deviation value gradually increases, the actual energy efficiency coefficient gradually exceeds the median value; if neither the positive nor the negative deviation value gradually increases, no intervention is needed.

9. A method for load distribution among multiple chiller units according to claim 8, characterized in that: The specific method for adjusting the number of chillers in step three is as follows: If the positive deviation value gradually increases, the number of chillers connected in parallel is gradually increased until the positive deviation value of each chiller no longer gradually increases; if the negative deviation value gradually increases, the number of chillers connected in parallel is gradually decreased until the negative deviation value of each chiller no longer gradually increases.

10. A multi-unit load distribution system for chillers, used to implement the multi-unit load distribution method for chillers as described in any one of claims 1-9, characterized in that, The chiller unit multi-unit load dispatching system includes: a data acquisition module, a temperature feedback control module, and an energy efficiency optimization and decision-making module; The data acquisition module is used to collect the real-time temperature value at the terminal and the water flow rate at the inflow terminal. The temperature feedback control module sets a terminal set temperature value and a valve deflection rule library. It calculates the temperature difference value by comparing the terminal set temperature value with the terminal real-time temperature value. Based on the temperature difference value, it analyzes whether there is a deviation between the terminal set temperature value and the terminal real-time temperature value. If no deviation is found, it executes the energy efficiency optimization and decision module. If a deviation is found, it deflects the valve angle according to the valve deflection rule library to make the terminal set temperature value equal to the terminal real-time temperature value, and returns the result to the data acquisition module. The energy efficiency optimization and decision-making module: Under time series conditions, it comprehensively calculates and normalizes the water flow rate and temperature difference at the inflow end to obtain the actual energy efficiency coefficient; it sets the optimal energy efficiency coefficient range and controls the actual energy efficiency coefficient of the chiller within the optimal range; it analyzes the deviation trend of the actual energy efficiency coefficient within the optimal range; if the analysis shows that the actual energy efficiency coefficient has not deviated, it returns to the data acquisition module or terminates directly; if the analysis shows that the actual energy efficiency coefficient has deviated, it adjusts the number of chillers in operation to ensure that the actual energy efficiency coefficient of the chiller always remains within the optimal range, and continues to analyze the deviation trend of the actual energy efficiency coefficient within the optimal range.

Images