An energy-saving optimization control method and system for central air conditioning chiller units

Abstract

This invention provides an energy-saving optimization control method and system for central air conditioning chiller units. The method includes: after collecting at least one year's worth of data by the data acquisition module, optimizing the operating strategy of the chiller units using the following methods: S11, plotting the load rate-operating energy efficiency curves of each chiller unit, and sorting and marking them; S12, establishing and training a multi-step building load prediction model to predict the hourly load of the building for the next 24 hours; S13, referring to the historical operating data of the chiller units, formulating a preliminary operating strategy for the chiller units in each time period of the next day; S14, with the goal of minimizing the total energy consumption of the activated chiller units, formulating an optimized operating strategy for the chiller units in each time period of the next day; S15, according to the optimized operating strategy formulated in S14, activating the corresponding chiller units in each time period of the next day. This invention achieves energy-saving optimization control of central air conditioning systems and has high engineering application value.

Description

Technical Field

[0001] This invention relates to the field of energy conservation, and specifically to an energy-saving optimization control method and system for central air conditioning chiller units. Background Technology

[0002] In recent years, with the rapid development of urbanization, building energy consumption has increased significantly, making building energy conservation research an important direction in energy conservation studies. Among these, air conditioning systems have the highest energy consumption, accounting for over 60% of total building energy consumption, making them the primary energy consumer in building operation. Chillers, as a crucial component of central air conditioning systems, account for 30%-40% of the total energy consumption in actual operation. Optimized operation of chillers significantly impacts the overall energy consumption of the central air conditioning system. Therefore, efficient and energy-saving operation of chillers is vital for central air conditioning systems, and the overall energy efficiency of chillers has a significant impact on building energy consumption, making it a powerful entry point for building energy conservation optimization.

[0003] Therefore, how to adjust the operating strategy of chiller units to achieve overall energy-saving optimization of central air conditioning is an important issue in the field of energy conservation. Summary of the Invention

[0004] The purpose of this invention is to address the aforementioned problems in the prior art by providing an energy-saving optimization control method and system for central air conditioning chiller units, thereby reducing the total energy consumption of the chiller units during operation and achieving energy-saving optimization of the central air conditioning system.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] An energy-saving optimization control method for central air conditioning chiller units includes:

[0007] The data acquisition module collects and stores historical operating data of the chiller units in the central air conditioning system, outdoor meteorological data, and building work schedules. The outdoor meteorological data includes outdoor temperature and outdoor relative humidity. The historical operating data of the chiller units includes the start-up time, shutdown time, maintenance records, load rate and operating energy efficiency of each chiller unit during operation.

[0008] After the data acquisition module has collected at least one year's worth of data, the following methods are used to optimize the operating strategy of the chiller unit:

[0009] S11. Based on the historical operating data of the chiller units, plot the load rate-operating energy efficiency curves of each chiller unit, and sort and mark the operating energy efficiency of all chiller units within different load rate distribution intervals.

[0010] S12. Based on the historical operating data of the chiller unit and outdoor meteorological data, establish and train a multi-step prediction model for building load to predict the hourly load of the building for the next 24 hours.

[0011] S13. Divide the 24 hours of a day into multiple time periods. Refer to the historical operating data of the chiller units, including the number of chiller units in operation and the rated cooling capacity of each chiller unit in each time period, to formulate a preliminary working strategy for the chiller units in each time period of the following day. The preliminary working strategy refers to the number of chiller units that should be turned on in each time period of the following day, and the rated cooling capacity that each chiller unit needs to achieve.

[0012] S14. Based on the primary working strategy determined in S13, the hourly load predicted for the next 24 hours in S12, and the ranking and labeling results of the chiller unit operating efficiency obtained in S11, with the goal of minimizing the total energy consumption of the chiller units in operation, formulate optimized working strategies for the chiller units in each time period of the next day; the optimized working strategy refers to which chiller units should be turned on in each time period of the next day.

[0013] S15. In accordance with the optimized work strategy established in S14, start the corresponding chiller units at different times on the following day.

[0014] Further, S11 includes:

[0015] S111. For each chiller unit, based on the collected load factor (PLR) and operating energy efficiency (COP)... i The following operational energy efficiency model was established, and the model parameters were solved using the least squares method:

[0016] COP i =β0 + β1PLR + β2PLR 2 ;

[0017] In the formula: β0, β1, and β2 are model parameters;

[0018] S112. Based on the operating energy efficiency model of each chiller unit, plot the load rate-operating energy efficiency curve of each chiller unit respectively;

[0019] S113. Based on the load rate-operating energy efficiency curves of each chiller unit, the load rate distribution range of all chiller units is obtained, and the load rate distribution range is further divided into multiple mutually exclusive sub-ranges; within each sub-range, the operating energy efficiency of all chiller units is sorted and marked.

[0020] Further, S12 includes:

[0021] S121. Based on the LSTM prediction model, outdoor temperature, outdoor relative humidity and building work schedule are selected as feature variables, and the total load of the chillers running in the building is selected as the output result to establish a multi-step prediction model for building load; the collected historical operating data of the chillers and outdoor meteorological data are used to train the multi-step prediction model for building load.

[0022] S122. Use the trained building load multi-step prediction model to predict the 24-hour hourly load of the building for the next day.

[0023] Furthermore, S14 specifically includes:

[0024] When only one chiller unit needs to be started during a certain period of the next day, the load rate of that period is first calculated based on the prediction results of S12, and then the chiller unit with the highest energy efficiency under that load rate is selected as the target to be started based on the sorting and marking results of S11.

[0025] When n (n≥2) chiller units need to be started during a certain time period the following day, first calculate the total load Q for that time period based on the prediction results of S12. n Then, with the goal of minimizing the total energy consumption of n chillers, the PSO algorithm is used to solve for the load rate of each chiller when the total energy consumption of n chillers is minimized:

[0026] PLR = (PLR1, PLR2, ..., PLR) n );

[0027] The total energy consumption of n chiller units is:

[0028]

[0029] In the formula, PLR i Let COP be the load factor of the i-th chiller unit. i For the operating energy efficiency of the i-th chiller unit, Q i Let i be the rated cooling capacity of the i-th chiller unit, where i = 1, 2, ..., n;

[0030] The optimization constraints of the PSO algorithm are as follows:

[0031] 0 <PLR i <1 (i = 1, 2, ..., n);

[0032]

[0033] After determining the load rate of each chiller unit based on the optimization results of the PSO algorithm, the chiller unit with the highest energy efficiency under each load rate is selected as the start-up target based on the sorting and marking results of S11.

[0034] Furthermore, it also includes: when the data acquisition module has not yet collected a full year's worth of data, a rotation work strategy is adopted to optimize the working strategy of the chiller units; the rotation work strategy is: based on the operating time of each chiller unit within the maintenance cycle, the chiller units with shorter operating times are given priority to be turned on.

[0035] Furthermore, the rotation strategy specifically includes:

[0036] S21. Based on the collected historical operating data of the chiller units, calculate the cumulative operating time t of each chiller unit after the most recent maintenance.

[0037] S22. Sort the chiller units in ascending order of cumulative running time t;

[0038] S23. When it is necessary to start the chiller unit, according to the order in S22, the unit with the shorter cumulative running time t shall be started first.

[0039] S24. After the chiller units have been running for a period of time, the cumulative running time t of each chiller unit is re-sorted. When the chiller units are started up next time, the chiller unit with the shorter cumulative running time t is selected to start according to the new sorting.

[0040] S25. If a chiller unit has undergone maintenance, the cumulative running time t of that chiller unit will be reset to zero and reordered.

[0041] An energy-saving optimization control system for central air conditioning chiller units, used in the methods described above, comprising:

[0042] The data acquisition module is used to collect and store historical operating data of the chiller units in the central air conditioning system, outdoor meteorological data, and building work schedules; the outdoor meteorological data includes outdoor temperature and outdoor relative humidity; the historical operating data of the chiller units includes the start-up time, shutdown time, maintenance records, load rate and operating energy efficiency of each chiller unit during operation.

[0043] The chiller unit load distribution optimization module is used to provide optimized working strategies for the chiller unit after the data acquisition module has collected at least one year's worth of data, so as to reduce the overall energy consumption of the chiller unit.

[0044] The chiller control module is used to control each chiller to start or stop at different times according to the optimized working strategy provided by the chiller load distribution optimization module, so as to distribute the total load to the chillers that are turned on.

[0045] In the chiller unit load distribution optimization module, the method for formulating the optimization strategy is as follows:

[0046] S11. Based on the historical operating data of the chiller units, plot the load rate-operating energy efficiency curves of each chiller unit, and sort and mark the operating energy efficiency of all chiller units within different load rate distribution intervals.

[0047] S12. Based on the historical operating data of the chiller unit and outdoor meteorological data, establish and train a multi-step prediction model for building load to predict the hourly load of the building for the next 24 hours.

[0048] S13. Divide the 24 hours of a day into multiple time periods. Refer to the historical operating data of the chiller units, including the number of chiller units in operation and the rated cooling capacity of each chiller unit in each time period, to formulate a preliminary working strategy for the chiller units in each time period of the following day. The preliminary working strategy refers to the number of chiller units that should be turned on in each time period of the following day, and the rated cooling capacity that each chiller unit needs to achieve.

[0049] S14. Based on the primary working strategy determined in S13, the hourly load predicted for the next 24 hours in S12, and the ranking and marking results of the chiller unit operating efficiency obtained in S11, with the goal of minimizing the total energy consumption of the chiller units in operation, formulate an optimized working strategy for the chiller units in each time period of the next day; the optimized working strategy refers to which chiller units should be turned on in each time period of the next day.

[0050] Furthermore, the data acquisition module includes: a chiller unit data acquisition device, an outdoor meteorological data acquisition device, and an industrial control computer;

[0051] The chiller unit data acquisition device is connected to each chiller unit and is used to collect the start-up time, shutdown time, maintenance records, load rate and operating energy efficiency of each chiller unit during operation.

[0052] The outdoor meteorological data acquisition device is installed in the outdoor space and is used to collect outdoor temperature and outdoor relative humidity;

[0053] The industrial control computer is connected to the chiller unit data acquisition device and the outdoor meteorological data acquisition device via a network, and is used to classify, summarize and store the data collected by the chiller unit data acquisition device and the outdoor meteorological data acquisition device.

[0054] Furthermore, in the chiller unit load distribution optimization module, S14 specifically includes:

[0055] When only one chiller unit needs to be started during a certain period of the next day, the load rate of that period is first calculated based on the prediction results of S12, and then the chiller unit with the highest energy efficiency under that load rate is selected as the target to be started based on the sorting and marking results of S11.

[0056] When n (n≥2) chiller units need to be started during a certain time period the following day, first calculate the total load Q for that time period based on the prediction results of S12. n Then, with the goal of minimizing the total energy consumption of n chillers, the PSO algorithm is used to solve for the load rate of each chiller when the total energy consumption of n chillers is minimized:

[0057] PLR = (PLR1, PLR2, ..., PLR) n );

[0058] The total energy consumption of n chiller units is:

[0059]

[0060] In the formula, PLR i Let COP be the load factor of the i-th chiller unit. i For the operating energy efficiency of the i-th chiller unit, Q i Let i be the rated cooling capacity of the i-th chiller unit, where i = 1, 2, ..., n;

[0061] The optimization constraints of the PSO algorithm are as follows:

[0062] 0 <PLR i <1 (i = 1, 2, ..., n);

[0063]

[0064] After determining the load rate of each chiller unit based on the optimization results of the PSO algorithm, the chiller unit with the highest energy efficiency under each load rate is selected as the start-up target based on the sorting and marking results of S11.

[0065] Furthermore, the chiller control module is also used to optimize the chiller's operating strategy by adopting a rotational working strategy when the data acquisition module has not yet collected a full year's worth of data.

[0066] The rotation strategy specifically includes:

[0067] S21. Based on the collected historical operating data of the chiller units, calculate the cumulative operating time t of each chiller unit after the most recent maintenance.

[0068] S22. Sort the chiller units in ascending order of cumulative running time t;

[0069] S23. When it is necessary to start the chiller unit, according to the order in S22, the unit with the shorter cumulative running time t shall be started first.

[0070] S24. After the chiller units have been running for a period of time, the cumulative running time t of each chiller unit is re-sorted. When the chiller units are started up next time, the chiller unit with the shorter cumulative running time t is selected to start according to the new sorting.

[0071] S25. If a chiller unit has undergone maintenance, the cumulative running time t of that chiller unit will be reset to zero and reordered.

[0072] This invention is based on accumulated historical data. By establishing an energy efficiency model for chiller unit operation, it analyzes and plots the load rate and operating energy efficiency of each chiller unit to obtain the operating energy efficiency curve of each chiller unit. Then, based on the changes in total load in historical data, it establishes a multi-step prediction model for building load to predict the hourly load of the building for 24 hours the next day. Finally, by referring to historical data for each time period, it determines the number of chiller units to be turned on and the rated cooling capacity of each chiller unit for different time periods of the next day. Based on the operating energy efficiency curve of the chiller units and the optimization results of the PSO algorithm, it determines which chiller units should be turned on for different time periods of the next day to distribute the total load to the turned-on chiller units, thereby reducing the overall energy consumption of the chiller units.

[0073] This invention fully correlates the collected historical operating data of chiller units with outdoor meteorological data to predict future (next day) load demand. Furthermore, it combines the historical start-up patterns and energy efficiency characteristics of chiller units to deeply optimize the start-up strategy of chiller units, which can save a lot of power resources and realize energy-saving optimization control of central air conditioning systems, and has high engineering application value. Attached Figure Description

[0074] Figure 1 This is a flowchart of the alternating operation strategy of the chiller unit in Embodiment 1 of the present invention.

[0075] Figure 2 This is a flowchart of the optimized working strategy of the chiller unit in Embodiment 1 of the present invention. Detailed Implementation

[0076] The technical solution of the present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0077] Example 1

[0078] This invention provides an energy-saving optimization control method for central air conditioning chiller units, comprising:

[0079] The data acquisition module collects and stores historical operating data of the chiller units in the central air conditioning system, outdoor meteorological data, and building work schedule (i.e., the opening time of building equipment, including commuting time, holidays, etc.); the outdoor meteorological data includes outdoor temperature and outdoor relative humidity; the historical operating data of the chiller units includes the start-up time, shutdown time, maintenance records, load rate and operating energy efficiency of each chiller unit during operation.

[0080] When the data acquisition module has not yet collected a full year's worth of data, a rotation work strategy is adopted to optimize the working strategy of the chiller units. The rotation work strategy is as follows: based on the running time of each chiller unit within the maintenance cycle, the chiller unit with the shorter running time is given priority to be turned on in order to maintain the mechanical wear balance of similar equipment.

[0081] like Figure 1 As shown, the rotation strategy specifically includes:

[0082] S21. Based on the collected historical operating data of the chiller units, calculate the cumulative operating time t of each chiller unit after the most recent maintenance.

[0083] S22. Sort the chiller units in ascending order of cumulative running time t;

[0084] S23. When it is necessary to start the chiller unit, according to the order in S22, the unit with the shorter cumulative running time t shall be started first.

[0085] S24. After the chiller units have been running for a period of time, the cumulative running time t of each chiller unit is re-sorted. When the chiller units are started up next time, the chiller unit with the shorter cumulative running time t is selected to start according to the new sorting.

[0086] S25. If a chiller unit has undergone maintenance, the cumulative running time t of that chiller unit will be reset to zero and reordered.

[0087] The rotation strategy can balance the mechanical wear and lifespan of each chiller unit, and also ensure that each chiller unit has enough working time, so that the data acquisition module can collect sufficient historical operating data from each chiller unit, providing sufficient data support for subsequent energy-saving optimization work.

[0088] like Figure 2 As shown, after the data acquisition module has collected at least one year's worth of data, the following methods are used to optimize the operating strategy of the chiller unit:

[0089] S11. Based on the historical operating data of the chiller units, plot the load rate-operating energy efficiency curves of each chiller unit, and sort and mark the operating energy efficiency of all chiller units within different load rate distribution intervals.

[0090] S12. Based on the historical operating data of the chiller unit and outdoor meteorological data, establish and train a multi-step prediction model for building load to predict the hourly load of the building for the next 24 hours.

[0091] S13. Divide the 24 hours of a day into multiple time periods. Refer to the historical operating data of the chiller units, including the number of chiller units in operation and the rated cooling capacity of each chiller unit in each time period, to formulate a preliminary working strategy for the chiller units in each time period of the following day. The preliminary working strategy refers to the number of chiller units that should be turned on in each time period of the following day, and the rated cooling capacity that each chiller unit needs to achieve.

[0092] S14. Based on the primary working strategy determined in S13, the hourly load predicted for the next 24 hours in S12, and the ranking and labeling results of the chiller unit operating efficiency obtained in S11, with the goal of minimizing the total energy consumption of the chiller units in operation, formulate optimized working strategies for the chiller units in each time period of the next day; the optimized working strategy refers to which chiller units should be turned on in each time period of the next day.

[0093] S15. In accordance with the optimized work strategy established in S14, start the corresponding chiller units at different times on the following day.

[0094] Specifically, S11 includes:

[0095] S111. For each chiller unit, based on the collected load factor (PLR) and operating energy efficiency (COP)... i The following operational energy efficiency model was established, and the model parameters were solved using the least squares method:

[0096] COP i =β0 + β1PLR + β2PLR 2 ;

[0097] In the formula: β0, β1, and β2 are model parameters;

[0098] S112. Based on the operating energy efficiency model of each chiller unit, plot the load rate-operating energy efficiency curve of each chiller unit respectively;

[0099] S113. Based on the load rate-operating energy efficiency curves of each chiller unit, the load rate distribution range of all chiller units is obtained, and the load rate distribution range is further divided into multiple mutually exclusive sub-ranges; within each sub-range, the operating energy efficiency of all chiller units is sorted and marked.

[0100] Further, S12 includes:

[0101] S121. Based on the LSTM prediction model, outdoor temperature, outdoor relative humidity and building work schedule are selected as feature variables, and the total load of the chillers running in the building is selected as the output result to establish a multi-step prediction model for building load; the collected historical operating data of the chillers and outdoor meteorological data are used to train the multi-step prediction model for building load.

[0102] S122. Use the trained building load multi-step prediction model to predict the hourly load of the building for the next 24 hours (i.e., the total load for each hour of the next 24 hours).

[0103] Furthermore, in S13, the number of chiller units operating and their rated cooling capacity within a certain time period each day can be statistically analyzed. The average value of the data for that time period each day is then rounded down to serve as the initial operating strategy for the next day. Alternatively, the outdoor temperature and relative humidity for the next day can be obtained from weather forecasts. Then, a specific day or several days matching the outdoor meteorological data can be found in historical data, and the initial operating strategy for the next day can be determined similarly by referring to the matching historical data. In addition, those skilled in the art can use many other different methods to formulate the initial operating strategy for the next day by referring to the historical operating data of the chiller units; all of these should be included within the scope of the technical solutions claimed in this invention, and will not be elaborated upon here.

[0104] Furthermore, S14 specifically includes:

[0105] When only one chiller unit needs to be started during a certain period of the next day, the load rate of that period is first calculated based on the prediction results of S12, and then the chiller unit with the highest energy efficiency under that load rate is selected as the target to be started based on the sorting and marking results of S11.

[0106] When n (n≥2) chiller units need to be started during a certain time period the following day, first calculate the total load Q for that time period based on the prediction results of S12. n Then, with the goal of minimizing the total energy consumption of n chillers, the PSO algorithm is used to solve for the load rate of each chiller when the total energy consumption of n chillers is minimized:

[0107] PLR = (PLR1, PLR2, ..., PLR) n);

[0108] The total energy consumption of n chiller units is:

[0109]

[0110] In the formula, PLR i Let COP be the load factor of the i-th chiller unit. i For the operating energy efficiency of the i-th chiller unit, Q i Let i be the rated cooling capacity of the i-th chiller unit, where i = 1, 2, ..., n;

[0111] The optimization constraints of the PSO algorithm are as follows:

[0112] 0 <PLR i <1 (i = 1, 2, ..., n);

[0113]

[0114] After determining the load rate of each chiller unit based on the optimization results of the PSO algorithm, the chiller unit with the highest energy efficiency under each load rate is selected as the start-up target based on the sorting and marking results of S11.

[0115] Example 2

[0116] This invention provides an energy-saving optimization control system for central air conditioning chillers, used to implement the method described in Embodiment 1. This embodiment specifically includes:

[0117] The data acquisition module is used to collect and store historical operating data of the chiller units in the central air conditioning system, outdoor meteorological data, and building work schedules; the outdoor meteorological data includes outdoor temperature and outdoor relative humidity; the historical operating data of the chiller units includes the start-up time, shutdown time, maintenance records, load rate and operating energy efficiency of each chiller unit during operation.

[0118] The chiller unit load distribution optimization module is used to provide optimized working strategies for the chiller unit after the data acquisition module has collected at least one year's worth of data, so as to reduce the overall energy consumption of the chiller unit.

[0119] The chiller control module is used to control each chiller unit to start or stop at different times according to the optimized working strategy provided by the chiller unit load distribution optimization module, so as to distribute the total load to the chillers that are turned on.

[0120] Furthermore, the data acquisition module specifically includes: a chiller unit data acquisition device, an outdoor meteorological data acquisition device, and an industrial control computer;

[0121] The chiller unit data acquisition device is connected to each chiller unit and is used to collect the start-up time, shutdown time, maintenance records, load rate and operating energy efficiency of each chiller unit during operation.

[0122] The outdoor meteorological data acquisition device is installed in the outdoor space and is used to collect outdoor temperature and outdoor relative humidity;

[0123] The industrial control computer is connected to the chiller unit data acquisition device and the outdoor meteorological data acquisition device via a network, and is used to classify, summarize and store the data collected by the chiller unit data acquisition device and the outdoor meteorological data acquisition device.

[0124] Furthermore, the chiller control module is also used to optimize the chiller's operating strategy by adopting a rotational working strategy when the data acquisition module has not yet collected a full year's worth of data.

[0125] The rotation strategy specifically includes:

[0126] S21. Based on the collected historical operating data of the chiller units, calculate the cumulative operating time t of each chiller unit after the most recent maintenance.

[0127] S22. Sort the chiller units in ascending order of cumulative running time t;

[0128] S23. When it is necessary to start the chiller unit, according to the order in S22, the unit with the shorter cumulative running time t shall be started first.

[0129] S24. After the chiller units have been running for a period of time, the cumulative running time t of each chiller unit is re-sorted. When the chiller units are started up next time, the chiller unit with the shorter cumulative running time t is selected to start according to the new sorting.

[0130] S25. If a chiller unit has undergone maintenance, the cumulative running time t of that chiller unit will be reset to zero and reordered.

[0131] Furthermore, in the chiller unit load distribution optimization module, the method for formulating the optimization working strategy is as follows:

[0132] S11. Based on the historical operating data of the chiller units, plot the load rate-operating energy efficiency curves of each chiller unit, and sort and mark the operating energy efficiency of all chiller units within different load rate distribution intervals.

[0133] S12. Based on the historical operating data of the chiller unit and outdoor meteorological data, establish and train a multi-step prediction model for building load to predict the hourly load of the building for the next 24 hours.

[0134] S13. Divide the 24 hours of a day into multiple time periods. Refer to the historical operating data of the chiller units, including the number of chiller units in operation and the rated cooling capacity of each chiller unit in each time period, to formulate a preliminary working strategy for the chiller units in each time period of the following day. The preliminary working strategy refers to the number of chiller units that should be turned on in each time period of the following day, and the rated cooling capacity that each chiller unit needs to achieve.

[0135] S14. Based on the primary working strategy determined in S13, the hourly load predicted for the next 24 hours in S12, and the ranking and marking results of the chiller unit operating efficiency obtained in S11, with the goal of minimizing the total energy consumption of the chiller units in operation, formulate an optimized working strategy for the chiller units in each time period of the next day; the optimized working strategy refers to which chiller units should be turned on in each time period of the next day.

[0136] Wherein, S11 includes:

[0137] S111. For each chiller unit, based on the collected load factor (PLR) and operating energy efficiency (COP)... i The following operational energy efficiency model was established, and the model parameters were solved using the least squares method:

[0138] COP i =β0 + β1PLR + β2PLR 2 ;

[0139] In the formula: β0, β1, and β2 are model parameters;

[0140] S112. Based on the operating energy efficiency model of each chiller unit, plot the load rate-operating energy efficiency curve of each chiller unit respectively;

[0141] S113. Based on the load rate-operating energy efficiency curves of each chiller unit, the load rate distribution range of all chiller units is obtained, and the load rate distribution range is further divided into multiple mutually exclusive sub-ranges; within each sub-range, the operating energy efficiency of all chiller units is sorted and marked.

[0142] S12 includes:

[0143] S121. Based on the LSTM prediction model, outdoor temperature, outdoor relative humidity and building work schedule are selected as feature variables, and the total load of the chillers running in the building is selected as the output result to establish a multi-step prediction model for building load; the collected historical operating data of the chillers and outdoor meteorological data are used to train the multi-step prediction model for building load.

[0144] S122. Use the trained building load multi-step prediction model to predict the hourly load of the building for the next 24 hours (i.e., the total load for each hour of the next 24 hours).

[0145] S14 specifically includes:

[0146] When only one chiller unit needs to be started during a certain period of the next day, the load rate of that period is first calculated based on the prediction results of S12, and then the chiller unit with the highest energy efficiency under that load rate is selected as the target to be started based on the sorting and marking results of S11.

[0147] When n (n≥2) chiller units need to be started during a certain time period the following day, first calculate the total load Q for that time period based on the prediction results of S12. n Then, with the goal of minimizing the total energy consumption of n chillers, the PSO algorithm is used to solve for the load rate of each chiller when the total energy consumption of n chillers is minimized:

[0148] PLR = (PLR1, PLR2, ..., PLR) n );

[0149] The total energy consumption of n chiller units is:

[0150]

[0151] In the formula, PLR i Let COP be the load factor of the i-th chiller unit. i For the operating energy efficiency of the i-th chiller unit, Q i Let i be the rated cooling capacity of the i-th chiller unit, where i = 1, 2, ..., n;

[0152] The optimization constraints of the PSO algorithm are as follows:

[0153] 0 <PLR i <1 (i = 1, 2, ..., n);

[0154]

[0155] After determining the load rate of each chiller unit based on the optimization results of the PSO algorithm, the chiller unit with the highest energy efficiency under each load rate is selected as the start-up target based on the sorting and marking results of S11.

[0156] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. An energy-saving optimization control method for central air conditioning chiller units, characterized in that, include: The data acquisition module collects and stores historical operating data of the chiller units in the central air conditioning system, outdoor meteorological data, and building work schedules. The outdoor meteorological data includes outdoor temperature and outdoor relative humidity; the historical operating data of the chiller units includes the start-up time, shutdown time, maintenance records, load rate and operating energy efficiency of each chiller unit during operation. When the data acquisition module has not yet collected a full year's worth of data, a rotation work strategy is adopted to optimize the working strategy of the chiller units. The rotation work strategy is as follows: based on the running time of each chiller unit within the maintenance cycle, the chiller units with shorter running time are given priority to be turned on. The rotation strategy specifically includes: S21. Based on the collected historical operating data of the chiller units, calculate the cumulative operating time t of each chiller unit after the most recent maintenance. S22. Sort the chiller units in ascending order of cumulative running time t; S23. When it is necessary to start the chiller unit, according to the order in S22, the unit with the shorter cumulative running time t shall be started first. S24. After the chiller units have been running for a period of time, the cumulative running time t of each chiller unit is re-sorted. When the chiller units are started up next time, the chiller unit with the shorter cumulative running time t is selected to start according to the new sorting. S25. If a chiller unit has undergone maintenance, the cumulative running time t of that chiller unit will be reset to zero and reordered. After the data acquisition module has collected at least one year's worth of data, the following methods are used to optimize the operating strategy of the chiller unit: S11. Based on the historical operating data of the chiller units, plot the load rate-operating energy efficiency curves of each chiller unit, and sort and mark the operating energy efficiency of all chiller units within different load rate distribution intervals. S12. Based on the historical operating data of the chiller unit and outdoor meteorological data, establish and train a multi-step prediction model for building load to predict the hourly load of the building for the next 24 hours. S13. Divide the 24 hours of a day into multiple time periods. Refer to the historical operating data of the chiller units, including the number of chiller units in operation and the rated cooling capacity of each chiller unit in each time period, to formulate a preliminary working strategy for the chiller units in each time period of the following day. The preliminary working strategy refers to the number of chiller units that should be turned on in each time period of the following day, and the rated cooling capacity that each chiller unit needs to achieve. S14. Based on the primary working strategy determined in S13, the hourly load predicted for the next 24 hours in S12, and the ranking and labeling results of the chiller unit operating efficiency obtained in S11, with the goal of minimizing the total energy consumption of the chiller units in operation, formulate optimized working strategies for the chiller units in each time period of the next day; the optimized working strategy refers to which chiller units should be turned on in each time period of the next day. S15. In accordance with the optimized work strategy established in S14, start the corresponding chiller units at different times on the following day. S14 specifically includes: When only one chiller unit needs to be started during a certain period of the next day, the load rate of that period is first calculated based on the prediction results of S12, and then the chiller unit with the highest energy efficiency under that load rate is selected as the target to be started based on the sorting and marking results of S11. When n (n≥2) chiller units need to be started during a certain time period the following day, first calculate the total load Q for that time period based on the prediction results of S12. n Then, with the goal of minimizing the total energy consumption of n chillers, the PSO algorithm is used to solve for the load rate of each chiller when the total energy consumption of n chillers is minimized: PLR=(PLR1,PLR2,…,PLR n ); The total energy consumption of n chiller units is: In the formula, PLR i Let COP be the load factor of the i-th chiller unit. i For the operating energy efficiency of the i-th chiller unit, Q i Let i be the rated cooling capacity of the i-th chiller unit, where i = 1, 2, ..., n; The optimization constraints of the PSO algorithm are as follows: 0<PLR i <1(i=1,2,…,n); After determining the load rate of each chiller unit based on the optimization results of the PSO algorithm, the chiller unit with the highest energy efficiency under each load rate is selected as the start-up target based on the sorting and marking results of S11.

2. The energy-saving optimization control method for central air conditioning chiller units according to claim 1, characterized in that, S11 includes: S111. For each chiller unit, based on the collected load factor (PLR) and operating energy efficiency (COP)... i The following operational energy efficiency model was established, and the model parameters were solved using the least squares method: COP i =β0+β1PLR+β2PLR 2 ; In the formula: β0, β1, and β2 are model parameters; S112. Based on the operating energy efficiency model of each chiller unit, plot the load rate-operating energy efficiency curve of each chiller unit respectively; S113. Based on the load rate-operating energy efficiency curves of each chiller unit, the load rate distribution range of all chiller units is obtained, and the load rate distribution range is further divided into multiple mutually exclusive sub-ranges; within each sub-range, the operating energy efficiency of all chiller units is sorted and marked.

3. The energy-saving optimization control method for central air conditioning chiller units according to claim 1, characterized in that, S12 includes: S121. Based on the LSTM prediction model, outdoor temperature, outdoor relative humidity and building work schedule are selected as feature variables, and the total load of the chillers running in the building is selected as the output result to establish a multi-step prediction model for building load; the collected historical operating data of the chillers and outdoor meteorological data are used to train the multi-step prediction model for building load. S122. Use the trained building load multi-step prediction model to predict the 24-hour hourly load of the building for the next day.

4. An energy-saving optimization control system for a central air conditioning chiller unit, used to implement the energy-saving optimization control method for a central air conditioning chiller unit as described in any one of claims 1 to 3, characterized in that, include: The data acquisition module is used to collect and store historical operating data of the chiller units in the central air conditioning system, outdoor meteorological data, and building work schedules. The outdoor meteorological data includes outdoor temperature and outdoor relative humidity; the historical operating data of the chiller units includes the start-up time, shutdown time, maintenance records, load rate and operating energy efficiency of each chiller unit during operation. The chiller unit load distribution optimization module is used to provide optimized working strategies for the chiller unit after the data acquisition module has collected at least one year's worth of data, so as to reduce the overall energy consumption of the chiller unit. The chiller control module is used to control each chiller to start or stop at different times according to the optimized working strategy provided by the chiller load distribution optimization module, so as to distribute the total load to the chillers that are turned on. In the chiller unit load distribution optimization module, the method for formulating the optimization strategy is as follows: S11. Based on the historical operating data of the chiller units, plot the load rate-operating energy efficiency curves of each chiller unit, and sort and mark the operating energy efficiency of all chiller units within different load rate distribution intervals. S12. Based on the historical operating data of the chiller unit and outdoor meteorological data, establish and train a multi-step prediction model for building load to predict the hourly load of the building for the next 24 hours. S13. Divide the 24 hours of a day into multiple time periods. Refer to the historical operating data of the chiller units, including the number of chiller units in operation and the rated cooling capacity of each chiller unit in each time period, to formulate a preliminary working strategy for the chiller units in each time period of the following day. The preliminary working strategy refers to the number of chiller units that should be turned on in each time period of the following day, and the rated cooling capacity that each chiller unit needs to achieve. S14. Based on the primary working strategy determined in S13, the hourly load predicted for the next 24 hours in S12, and the ranking and marking results of the chiller unit operating efficiency obtained in S11, with the goal of minimizing the total energy consumption of the chiller units in operation, formulate an optimized working strategy for the chiller units in each time period of the next day; the optimized working strategy refers to which chiller units should be turned on in each time period of the next day.

5. The energy-saving optimization control system for central air conditioning chiller units according to claim 4, characterized in that, The data acquisition module includes: a chiller unit data acquisition device, an outdoor meteorological data acquisition device, and an industrial control computer; The chiller unit data acquisition device is connected to each chiller unit and is used to collect the start-up time, shutdown time, maintenance records, load rate and operating energy efficiency of each chiller unit during operation. The outdoor meteorological data acquisition device is installed in the outdoor space and is used to collect outdoor temperature and outdoor relative humidity; The industrial control computer is connected to the chiller unit data acquisition device and the outdoor meteorological data acquisition device via a network, and is used to classify, summarize and store the data collected by the chiller unit data acquisition device and the outdoor meteorological data acquisition device.

6. The energy-saving optimization control system for central air conditioning chiller units according to claim 4, characterized in that, In the chiller unit load distribution optimization module, S14 specifically includes: When only one chiller unit needs to be started during a certain period of the next day, the load rate of that period is first calculated based on the prediction results of S12, and then the chiller unit with the highest energy efficiency under that load rate is selected as the target to be started based on the sorting and marking results of S11. When n (n≥2) chiller units need to be started during a certain time period the following day, first calculate the total load Q for that time period based on the prediction results of S12. n Then, with the goal of minimizing the total energy consumption of n chillers, the PSO algorithm is used to solve for the load rate of each chiller when the total energy consumption of n chillers is minimized: PLR=(PLR1,PLR2,…,PLR n ); The total energy consumption of n chiller units is: In the formula, PLR i Let COP be the load factor of the i-th chiller unit. i For the operating energy efficiency of the i-th chiller unit, Q i Let i be the rated cooling capacity of the i-th chiller unit, where i = 1, 2, ..., n; The optimization constraints of the PSO algorithm are as follows: 0<PLR i <1(i=1,2,…,n); After determining the load rate of each chiller unit based on the optimization results of the PSO algorithm, the chiller unit with the highest energy efficiency under each load rate is selected as the start-up target based on the sorting and marking results of S11.

7. The energy-saving optimization control system for central air conditioning chiller units according to claim 4, characterized in that, The chiller control module is also used to optimize the chiller's operating strategy by adopting a rotational working strategy when the data acquisition module has not yet collected a full year's worth of data. The rotation strategy specifically includes: S21. Based on the collected historical operating data of the chiller units, calculate the cumulative operating time t of each chiller unit after the most recent maintenance. S22. Sort the chiller units in ascending order of cumulative running time t; S23. When it is necessary to start the chiller unit, according to the order in S22, the unit with the shorter cumulative running time t shall be started first. S24. After the chiller units have been running for a period of time, the cumulative running time t of each chiller unit is re-sorted. When the chiller units are started up next time, the chiller unit with the shorter cumulative running time t is selected to start according to the new sorting. S25. If a chiller unit has undergone maintenance, the cumulative running time t of that chiller unit will be reset to zero and reordered.

Images