Method and system for start-stop control of multiple cold machines in an inaccurate measurement environment

Abstract

The application discloses a kind of measurement inaccurate environment under the start-stop control method and system of multiple chiller system, and relates to the technical field of multiple chiller system operation control.Through based on the determination of adjustment step and high error adjustment coefficient of chiller start-stop threshold correction direction coefficient, and then determine chiller capacity dynamic adjustment factor, it can comprehensively reflect the influence of system measurement uncertainty parameter on chiller rated capacity, realize the dynamic self-adaptive correction of chiller start-stop threshold;Through based on chiller rated capacity, current chiller opening number and chiller capacity dynamic adjustment factor determine the end cold load judgment threshold for opening and closing chiller, and based on the relative size relationship between current cold load measurement value and end cold load judgment threshold, determine the start-stop strategy of system chiller, it can improve the operation efficiency of system as far as possible under the premise of avoiding frequent start-stop of chiller group, to effectively guarantee the safe and reliable operation of multiple chiller system.

Description

Technical Field

[0001] This invention relates to the field of multi-cooler system operation control technology, and in particular to a start-up and shutdown control method and system for multi-cooler systems under conditions of inaccurate measurement. Background Technology

[0002] Multi-chill systems are widely used in large public buildings, centralized energy stations, and other similar settings to provide chilled water at suitable temperatures and flow rates to meet the cooling load demands of end users. Their operating energy consumption accounts for a significant portion of the building's total operating energy consumption, making it crucial to improve the operating efficiency of chiller groups to reduce the overall energy consumption of the central air conditioning system.

[0003] Due to sensor bias and measurement noise, multi-cooler systems operate in an environment of constant measurement inaccuracies, inevitably leading to discrepancies between measured and true operating parameters. Existing methods often employ fixed cooler switching thresholds for start-up and shutdown decisions. These thresholds perform poorly when operating parameters are inaccurately measured, potentially causing frequent cooler starts and stops, reducing system efficiency, and even triggering system safety incidents. Summary of the Invention

[0004] The purpose of this invention is to provide a start-up and shutdown control method and system for a multi-cooler system under conditions of inaccurate measurement, which can improve the operating efficiency of the system as much as possible while avoiding frequent start-up and shutdown of the cooler group, thereby effectively ensuring the safe and reliable operation of the multi-cooler system.

[0005] In a first aspect, the present invention provides a start-up and shutdown control method for a multi-cooler system under conditions of inaccurate measurement. Applied to a multi-cooler system, the method includes: during system operation, periodically acquiring online operating parameter values ​​corresponding to preset high-impact indicators in the multi-cooler system; comparing the online operating parameter values ​​with preset judgment thresholds to determine a high-error flag bit corresponding to the high-impact indicator, and determining a cooler start-up and shutdown threshold correction direction coefficient based on the high-error flag bit; determining an adjustment step size and a high-error adjustment coefficient based on the cooler start-up and shutdown threshold correction direction coefficient; determining a cooler capacity dynamic adjustment factor based on the cooler start-up and shutdown threshold correction direction coefficient, the adjustment step size, and the high-error adjustment coefficient; determining a terminal cooling load judgment threshold for starting and stopping the coolers based on the cooler rated capacity, the current number of coolers in operation, and the cooler capacity dynamic adjustment factor; and determining a system cooler start-up and shutdown strategy based on the relative magnitude relationship between the current cooling load measurement value and the terminal cooling load judgment threshold.

[0006] In some preferred embodiments of the present invention, the method further includes: acquiring historical datasets of operating parameters related to cooling load and historical datasets of cooling load in a multi-cooler system; training a data model between system operating parameters and cooling load using the historical datasets of operating parameters as input and the historical datasets of cooling load as output; selecting at least one set of data samples from the dataset as a benchmark parameter combination based on the number of coolers and the rated capacity of each cooler, and determining the proportional coefficient of each set of benchmark parameters; determining the sensitivity coefficient of each system operating parameter in each benchmark parameter combination based on a preset uncertainty, and determining the weighted sensitivity coefficient of each system operating parameter based on the proportional coefficient; sorting the operating parameters according to the weighted sensitivity coefficients, and selecting at least one system operating parameter as a high-impact indicator from largest to smallest.

[0007] In some preferred embodiments of the present invention, the system operating parameters include: chilled water supply temperature, chilled water return temperature, cooling water supply temperature, cooling water return temperature, bypass pipe flow rate, chilled water flow rate, terminal air supply temperature, and indoor temperature.

[0008] In some preferred embodiments of the present invention, selecting at least one set of data samples as a benchmark parameter combination from the dataset based on the number of chillers and the rated capacity of each chiller, and determining the proportional coefficient of each set of benchmark parameters includes: dividing the historical cooling load dataset into multiple data subsets; wherein the number of data subsets is equal to the total number of chillers; assigning all data samples whose cooling load values ​​are within the rated capacity of the first chiller to the first data subset, assigning all data samples whose cooling load values ​​are within the range from the rated capacity of the first chiller to the sum of the rated capacities of the first and second chillers to the second data subset, and so on, until all data subsets are divided; determining the average value of the cooling load values ​​in each data subset, and selecting the set of data samples whose cooling load values ​​are closest to the average value to form a benchmark parameter combination; and determining the proportional coefficient based on the benchmark parameter combination and the number of samples in the data subset.

[0009] In some preferred embodiments of the present invention, the sensitivity coefficient of each system operating parameter in each benchmark parameter combination is determined according to a preset uncertainty, and the weighted sensitivity coefficient of each system operating parameter is determined based on the proportional coefficient. This includes: for each benchmark parameter combination, after fine-tuning the value of one of the system operating parameters according to the uncertainty, substituting it into the data model to obtain the corresponding cooling load value, and determining the sensitivity coefficient of the system operating parameter using the following formula. ;in, For system operating parameters In the combination of reference parameters Sensitivity coefficient in For running parameters The original value; This is the original value of the cooling load. For running parameters The adjusted value For running parameters Fine-tune the cooling load value calculated by the input data model; determine the weighted sensitivity coefficient using the following formula: ;in, The weighted sensitivity coefficient for the running parameter j; The number of data subsets; is the scaling factor for the i-th set of reference parameters.

[0010] In some preferred embodiments of the present invention, comparing the online operating parameter values ​​with a preset judgment threshold to determine the high error flag corresponding to the high-impact index, and determining the chiller start-stop threshold correction direction coefficient based on the high error flag includes: if the online operating parameter value is greater than the judgment threshold, then setting the corresponding high error flag to 1; otherwise, setting it to 0; the chiller start-stop threshold correction direction coefficient is determined by the following formula: ;in, Correction of directional coefficient for chiller start / stop threshold; This is a flag indicating the cumulative high error. This represents the number of high error flags. This is the high error flag for the i-th system operating parameter.

[0011] In some preferred embodiments of the present invention, the adjustment step size, high error adjustment coefficient, and chiller capacity dynamic adjustment factor are determined by the following formulas: ; ; In the formula, The adjustment step size for the nth adjustment period. The preset attenuation base, This represents the cumulative number of times the chiller has been started and stopped to date. This is the high error adjustment coefficient for the nth adjustment period. This is a preset probability constant; This is the dynamic adjustment factor for the chiller capacity in the nth adjustment cycle.

[0012] In some preferred embodiments of the present invention, the threshold for determining the terminal cooling load is determined by the following formula: ; ;in, Determine the threshold for the terminal cooling load to start a chiller; The threshold for determining the terminal cooling load of a chiller to be shut down; This represents the number of chillers currently running in the system. For the first The rated capacity of the operating chiller. This is the preset dead zone coefficient for cold start-stop.

[0013] In some preferred embodiments of the present invention, the judgment threshold is determined by the following formula: ;in, High impact indicator The judgment threshold, High impact indicator The mean in historical datasets; High impact indicator Standard deviation in historical datasets, This is the preset confidence level coefficient.

[0014] Secondly, this invention provides a start-stop control system for a multi-cooler system under conditions of inaccurate measurement, applied to a multi-cooler system, comprising: a data acquisition module, used to periodically acquire online operating parameter values ​​corresponding to preset high-impact indicators in the multi-cooler system during system operation; a correction direction coefficient determination module, used to compare the online operating parameter values ​​with preset judgment thresholds, determine the high-error flag bit corresponding to the high-impact indicator, and determine the cooler start-stop threshold correction direction coefficient based on the high-error flag bit; a data processing module, used to determine the adjustment step size and high-error adjustment coefficient based on the cooler start-stop threshold correction direction coefficient; the data processing module is also used to determine the cooler capacity dynamic adjustment factor based on the cooler start-stop threshold correction direction coefficient, adjustment step size, and high-error adjustment coefficient; a cooling load judgment threshold determination module, used to determine the terminal cooling load judgment threshold for starting and stopping the coolers based on the cooler rated capacity, the current number of coolers in operation, and the cooler capacity dynamic adjustment factor; and a start-stop strategy determination module, used to determine the system cooler start-stop strategy based on the relative magnitude relationship between the current cooling load measurement value and the terminal cooling load judgment threshold.

[0015] This invention brings the following beneficial effects: This invention provides a start-up and shutdown control method and system for a multi-cooler system under conditions of inaccurate measurement. The method is applied to a multi-cooler system and includes: during system operation, periodically acquiring online operating parameter values ​​corresponding to preset high-impact indicators in the multi-cooler system; comparing the online operating parameter values ​​with preset judgment thresholds to determine a high-error flag bit corresponding to the high-impact indicator, and determining a cooler start-up / shutdown threshold correction direction coefficient based on the high-error flag bit; determining an adjustment step size and a high-error adjustment coefficient based on the cooler start-up / shutdown threshold correction direction coefficient; determining a cooler capacity dynamic adjustment factor based on the cooler start-up / shutdown threshold correction direction coefficient, adjustment step size, and high-error adjustment coefficient; determining a terminal cooling load judgment threshold for starting and stopping the coolers based on the cooler rated capacity, the current number of coolers in operation, and the cooler capacity dynamic adjustment factor; and determining a system cooler start-up / shutdown strategy based on the relative magnitude of the current cooling load measurement value and the terminal cooling load judgment threshold. This method can maximize system operating efficiency while avoiding frequent start-ups and shutdowns of the cooler group, thereby effectively ensuring the safe and reliable operation of the multi-cooler system. Attached Figure Description

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

[0017] Figure 1 A schematic diagram of a multi-cooler system provided in an embodiment of the present invention; Figure 2 A flowchart of a start-up and shutdown control method for a multi-cooler system under conditions of inaccurate measurement, provided as an embodiment of the present invention; Figure 3 This is a schematic diagram comparing the daily operating energy consumption of a multi-cooler system under different control methods provided in the embodiments of the present invention. Figure 4 This is a schematic diagram comparing the supply of terminal cooling load demand in a multi-cooling system under different control methods provided in the embodiments of the present invention; Figure 5 This invention provides a schematic diagram of the start-stop control system for a multi-cooler system under conditions of inaccurate measurement. Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0018] Icons: 310 - Data acquisition module; 320 - Correction direction coefficient determination module; 330 - Data processing module; 340 - Cooling load judgment threshold determination module; 350 - Start-stop strategy determination module; 400 - Memory; 401 - Processor; 402 - Bus; 403 - Communication interface. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0020] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0021] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0022] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of this invention is in use. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention. In addition, the terms "first," "second," "third," etc., are only used to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0023] Furthermore, terms such as "horizontal," "vertical," and "sag" do not imply that components must be absolutely horizontal or suspended, but rather that they can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal relative to "vertical," and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0024] In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0025] The following detailed description of some embodiments of the present invention is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0026] This invention provides a start-up and shutdown control method for a multi-cooler system under conditions of inaccurate measurement, applicable to multi-cooler systems.

[0027] See Figure 1 The diagram shown is a schematic of a multi-chiller system according to an embodiment of the present invention. The system comprises five chiller units connected in parallel, each with a rated cooling capacity of 1600kW and a rated COP of 5.71. On the cooling side, high-temperature cooling water is pumped by cooling water pumps to a cooling tower for heat dissipation and then returned to the condenser end of the chiller units. Each cooling water pump has a rated power of 40kW and a rated flow rate of 324m³ / h. 3 / h, each cooling tower has a rated heat exchange capacity of 2194kW, a rated fan power of 22kW, and a rated fan flow rate of 4200m³ / h. 3 / min. On the chilled side, chilled water pumps deliver low-temperature chilled water generated at the evaporator end of the chiller unit to the air handling unit to cool the air conditioning supply air. Each chilled water pump has a rated power of 27kW and a rated flow rate of 288m³ / min. 3 / h. On the wind side, the air handling unit delivers low-temperature air to each end user, thereby maintaining the room temperature at the set value. Operational data from July to October were selected as the dataset, with data collection intervals of 30 minutes, totaling 5904 data samples.

[0028] See Figure 2 The flowchart shown in this embodiment of the invention provides a start-up and shutdown control method for a multi-cooler system under conditions of inaccurate measurement. The method includes: Step S102: During the system operation phase, periodically acquire the online operating parameter values ​​corresponding to the preset high-impact indicators in the multi-cooler system.

[0029] Specifically, during real-time operation, multi-cooling systems are constantly in an environment of inaccurate measurements due to sensor bias and measurement noise. The preset high-impact indicators refer to several system operating parameters that have the most significant impact on the measurement deviation of the terminal cooling load, determined in advance through screening.

[0030] In this embodiment, the high-impact indicators are four system operating parameters: chilled water supply temperature, bypass pipe flow rate, terminal air supply temperature, and indoor temperature. Every 30 minutes, the current measured values ​​of these high-impact indicators are read from the corresponding sensors or data acquisition devices and used as the online operating parameter values.

[0031] By periodically acquiring the online operating parameter values ​​corresponding to high-impact indicators, the status of the most critical measurement parameters for judging cooling load can be monitored in real time, providing a data basis for subsequent judgment of the direction and degree of measurement deviation.

[0032] Step S104: Compare the online operating parameter values ​​with the preset judgment thresholds to determine the high error flag corresponding to the high impact index, and determine the chiller start-stop threshold correction direction coefficient based on the high error flag.

[0033] Specifically, for each high-impact indicator, the online operating parameter value is compared with a preset judgment threshold. If the online operating parameter value is greater than the preset judgment threshold, it indicates that the current measured value of the operating parameter deviates significantly from its historical normal distribution range, and there is a large measurement error. In this case, the high error flag corresponding to the high-impact indicator is set to 1; otherwise, it indicates that the measured value of the operating parameter is within the normal range, and the high error flag is set to 0.

[0034] After obtaining the high error flags of all high-impact indicators, calculate a cumulative high error flag. Its value is the sum of the high error flag bits, that is Where n is the number of chillers currently in operation, and m is the number of high error flags (i.e., the number of high-impact indicators). This is the high error flag for the i-th system operating parameter.

[0035] Then, the directional coefficient for correcting the chiller start / stop threshold is determined based on the accumulated high error flag. If the cumulative high error flag is not 0 (i.e., at least one high-impact indicator's measurement value deviates significantly), then... Setting it to -1 indicates that the dynamic adjustment factor for chiller capacity should be updated towards a decreasing direction to cautiously address potential overestimation of cooling load demand due to inaccurate measurements; if the cumulative high error flag is 0 (i.e., all high-impact indicators are within the normal range), then... Setting it to 1 indicates that the dynamic adjustment factor of the chiller capacity should be updated in a direction of increasing to meet the actual cooling load demand.

[0036] Furthermore, in some preferred embodiments of the present invention, the judgment threshold is determined by the following formula: ;in, High impact indicator The judgment threshold High impact indicator The mean in historical datasets; High impact indicator Standard deviation in historical datasets, This is the preset confidence level coefficient.

[0037] Specifically, this threshold is used to define whether the online operating parameter values ​​of high-impact indicators have deviated significantly. Statistical analysis of historical datasets of high-impact indicators is performed to calculate their mean as a baseline, and their standard deviation as a measure of fluctuation range. Confidence coefficient. Used to adjust the strictness of the judgment threshold: when When a larger value is chosen, the judgment threshold is higher, meaning that only when the online operating parameter value deviates significantly from the mean will it be considered to have a high error. This method is suitable for scenarios with high tolerance for measurement noise; when When a smaller value is selected, the judgment threshold is lower, making it more sensitive to the detection of measurement deviations. This is suitable for scenarios requiring high measurement accuracy and rapid response. In this embodiment, Set to 3. As an optional implementation, It can also be set to different values ​​such as 2 or 2.5 to be flexibly adjusted according to the sensitivity of the actual system to environments where measurements are inaccurate.

[0038] Therefore, by determining the judgment threshold based on the historical data mean and standard deviation and setting an adjustable confidence coefficient, it is possible to adapt to the tolerance requirements of different systems for measurement deviations, take into account detection sensitivity and false alarm rate, and improve the accuracy of high error flag judgment.

[0039] Step S106: Determine the adjustment step size and high error adjustment coefficient based on the directional coefficient for correcting the start-stop threshold of the chiller.

[0040] Specifically, in determining the directional coefficient for the chiller start-stop threshold... Next, two key parameters need to be determined: adjusting the step size. and high error adjustment coefficient These factors together determine the magnitude of each correction to the chiller start-stop threshold.

[0041] The step size is adjusted to evaluate the update step size of the chiller start-stop threshold. If the current correction direction coefficient satisfies the previous correction direction coefficient... If the directional coefficients of the chiller start-stop threshold corrections are opposite in sign, it indicates that the calculation of the chiller capacity dynamic adjustment factor is in the process of convergence. At this time, multiplying by a decay factor to reduce the adjustment step size can improve the convergence speed and stability. Conversely, if the directional coefficients are opposite in sign, it indicates that the calculation of the chiller capacity dynamic adjustment factor is still unstable. Therefore, the adjustment step size should be kept unchanged to approach the target value as soon as possible.

[0042] A high error adjustment factor is used to quantify the magnitude of the correction effect of inaccurate measurement parameters on the chiller start-up and shutdown thresholds. If operating parameter values ​​significantly deviate from their historical distribution range (i.e., ,or If the cumulative high error flag is used to the total number of high error flags, then the high error adjustment coefficient is used. This ratio directly reflects the proportion of high-impact indicators that are in an abnormal state. If there is no deviation, then a preset fixed probability constant d is used.

[0043] Step S108: Determine the dynamic adjustment factor of the chiller capacity based on the chiller start / stop threshold correction direction coefficient, adjustment step size, and high error adjustment coefficient.

[0044] Specifically, the dynamic adjustment factor of the chiller capacity It is used to comprehensively reflect the impact of system measurement uncertainties on the chiller's rated capacity; it is a dynamic parameter that is updated iteratively on a cycle-by-cycle basis. In each adjustment cycle, based on the current adjustment step size... High error adjustment coefficient And the directional coefficient of the chiller start-stop threshold correction Dynamic adjustment factor for chiller capacity in the previous cycle After making corrections, a new dynamic adjustment factor for chiller capacity is obtained. This dynamic adjustment factor, by introducing a step-decreasing mechanism and high error ratio feedback, can achieve dynamic adaptive correction of the chiller start-up and shutdown thresholds while avoiding frequent start-ups and shutdowns of chiller groups.

[0045] Furthermore, in some preferred embodiments of the present invention, the adjustment step size, high error adjustment coefficient, and chiller capacity dynamic adjustment factor are determined by the following formulas: ; ; In the formula, The adjustment step size for the nth adjustment period. The preset attenuation base, This represents the cumulative number of times the chiller has been started and stopped to date. This is the high error adjustment coefficient for the nth adjustment period. This is a preset probability constant; This is the dynamic adjustment factor for the chiller capacity in the nth adjustment cycle.

[0046] Specifically, when When the direction of the two corrections reverses, it indicates that the system is in the process of convergence and adjustment, and the step size should be adjusted at this time. In the previous step Multiply by the attenuation factor ,in, The number of start-stop cycles for the chiller is accumulated. As the number of start-stop cycles increases, the attenuation effect accumulates, causing the adjustment step size to gradually decrease and avoid oscillation. When Values When there is a greater number of high-impact indicators that deviate from the target value, the larger the adjustment coefficient and the greater the correction range, in order to reflect the degree of measurement inaccuracy more quickly; when there is no deviation, the adjustment coefficient is a constant value. This makes the correction behavior smoother and more stable. Finally, the dynamic adjustment factor of the chiller capacity. In the previous period value Based on this, along the correction direction With step size Adjustments will be made.

[0047] As an optional implementation method, the attenuation base It can also be set to other values ​​between 0 and 1, such as 0.8 or 0.95, to adjust the convergence speed; fix the probability constant. It can also be set to 0.3 or 0.7, etc., to adjust the benchmark correction magnitude under no deviation conditions.

[0048] Therefore, by introducing a convergence judgment mechanism for correcting opposite signs and deviation proportional feedback, the correction step size and amplitude can be dynamically adjusted. This enables a rapid response when deviations are measured and a stable approximation during the convergence process, thereby achieving efficient and stable calculation of the dynamic adjustment factor of the chiller capacity and effectively avoiding threshold oscillations.

[0049] In this embodiment, the attenuation base Set it to 0.9, and set the fixed probability constant d to 0.5.

[0050] Step S110: Determine the terminal cooling load judgment threshold for starting and stopping the chillers based on the rated capacity of the chiller, the current number of chillers in operation, and the chiller capacity dynamic adjustment factor.

[0051] Specifically, the dynamic adjustment factor of the chiller capacity for the current cycle is obtained. Then, combining the inherent rated capacity of each chiller and the number of chillers currently in operation, two key judgment thresholds are calculated: the terminal cooling load judgment threshold used to start a chiller. and the threshold for determining the terminal cooling load used to shut down a chiller. .

[0052] The start-up threshold determines at what level the terminal cooling load demand increases that an additional chiller needs to be started, while the shutdown threshold determines at what level the terminal cooling load demand decreases that a chiller needs to be shut down. Both thresholds are based on the total rated capacity of the currently operating chillers, dynamically adjusted using a chiller capacity dynamic adjustment factor, and incorporate a chiller start-up / shutdown dead zone coefficient. A dead zone is created between the start and stop thresholds to prevent the cold unit from frequently starting and stopping near the threshold.

[0053] Furthermore, in some preferred embodiments of the present invention, the threshold for determining the terminal cooling load is determined by the following formula: ; ;in, Determine the threshold for the terminal cooling load to start a chiller; The threshold for determining the terminal cooling load of a chiller to be shut down; This represents the number of chillers currently running in the system. For the first The rated capacity of the operating chiller. This is the preset dead zone coefficient for cold start-stop.

[0054] Specifically, the threshold for determining when to start a chiller. Equal to the dynamic adjustment factor of chiller capacity Multiply by all currently running The sum of the rated capacities of the two chillers. Because The threshold will change dynamically based on the degree of measurement inaccuracy, and will be adjusted accordingly: when a high-impact indicator shows a large measurement deviation, The system will adjust towards a decrease, thus lowering the activation threshold and preventing premature activation of the new chiller unit when measurements might overestimate the cooling load; when there is no measurement deviation... Adjust the threshold in the direction of increase to make the opening threshold closer to the actual rated capacity level. The threshold for shutting down a chiller. A different base is used than the activation threshold, i.e., using The calculation is performed by summing the rated capacities of the chillers and subtracting the dead zone coefficient. Achieve hysteresis control.

[0055] In this embodiment, Set to 0.05. As an optional implementation, It can also be set to different values ​​such as 0.03 or 0.08 to adjust the dead zone width according to the system's tolerance for the frequency of start-stop.

[0056] Therefore, by combining the dynamic adjustment factor of the chiller capacity with the rated total capacity of the operating chiller and introducing a start-stop dead zone, the start-stop threshold can be flexibly adjusted according to the measurement accuracy. At the same time, hysteresis control is used to avoid frequent start-stop of the chiller near the threshold, thereby improving the stability of system operation.

[0057] Step S112: Based on the relative magnitude of the current cooling load measurement value and the terminal cooling load judgment threshold, determine the start-up and shutdown strategy of the system chiller.

[0058] Specifically, in each control cycle, the current terminal cooling load is measured and compared with two judgment thresholds to determine the next chiller start-up and shutdown operation.

[0059] If the current cooling load measurement is higher than the threshold value for judging the terminal cooling load of turning on one chiller. If the measured return water temperature of the chilled water main is detected to be higher than the safe operating limit (this condition serves as an additional protection condition to ensure the safe operation of the system), an instruction is issued to start one chiller to meet the increased cooling load demand or to prevent safety hazards caused by excessively high return water temperature; if the current cooling load measurement value is lower than the terminal cooling load judgment threshold for shutting down one chiller. If the load is too low, a command is issued to shut down one chiller to avoid running too many chillers and reducing system energy efficiency. If the measured cooling load is between two thresholds, the current number of chillers in operation remains unchanged, maintaining the current operating status. The entire control process is repeated every 30 minutes to achieve periodic closed-loop control.

[0060] Furthermore, in some preferred embodiments of the present invention, before executing the above-described multi-cooler system start-up and shutdown control method, the method further includes: acquiring historical datasets of operating parameters related to cooling load and historical datasets of cooling load in the multi-cooler system; training a data model between system operating parameters and cooling load using the historical datasets of operating parameters as input and the historical datasets of cooling load as output; selecting at least one set of data samples from the dataset as a reference parameter combination based on the number of coolers and the rated capacity of each cooler, and determining the proportional coefficient of each set of reference parameters; determining the sensitivity coefficient of each system operating parameter in each reference parameter combination based on a preset uncertainty, and determining the weighted sensitivity coefficient of each system operating parameter based on the proportional coefficient; sorting the operating parameters according to the weighted sensitivity coefficients, and selecting at least one system operating parameter as a high-impact indicator from largest to smallest.

[0061] Specifically, in this embodiment, before the system is put into real-time operation, it is necessary to use historical operating data to complete the training of the data model and the screening of high-impact indicators.

[0062] First, historical datasets of operating parameters and cooling loads related to the cooling load in the multi-cooling system are obtained over a historical period. Using the system operating parameters as input and the cooling load as output, an artificial neural network is used to train a data model. This model can predict the corresponding terminal cooling load value based on any set of operating parameter inputs, thus providing a computational tool for subsequent sensitivity analysis. See Table 1 for a schematic diagram of artificial neural network parameter settings provided in this embodiment of the invention.

[0063] Table 1

[0064] As an optional implementation, the data model can also employ random forest algorithm, long short-term memory neural network, or convolutional neural network. For example, when using random forest algorithm, the number of decision trees can be set to 100, and the maximum depth to 10; when using long short-term memory neural network, the number of hidden layer nodes can be set to 64, and the time step to 5. Different data models can be flexibly selected according to the nonlinear characteristics and time-series dependencies of the actual system to better fit the mapping relationship between system operating parameters and cooling load.

[0065] Then, based on the number of chillers and the rated capacity of each chiller, the selection of benchmark parameter combinations and the determination of proportional coefficients are performed. Specifically, the historical cooling load dataset is divided into multiple data subsets according to the total number of chillers. In this embodiment, the total number of chillers is 5, therefore the dataset is divided into five data subsets. In this embodiment, the classification ranges for cooling load values ​​in the five data subsets are [0, 1600), [1600, 3200), [3200, 4800), [4800, 6400), and [6400, 8000]. All data samples with cooling load values ​​within the rated capacity of the first chiller (1600kW) are assigned to the first data subset. All data samples with cooling load values ​​between the rated capacity of the first chiller and the sum of the rated capacities of the first and second chillers (3200kW) are assigned to the second data subset, and so on, until all data subsets are divided. Calculate the average cooling load value for each subset, and select the data sample whose cooling load value is closest to this average value to form a baseline parameter combination. See Table 2 for a schematic diagram of the baseline parameter combination calculation results.

[0066] Table 2

[0067] Finally, the scaling factor for each group's baseline parameter is determined based on the sample size of each subset. The calculation formula is: ;in, The number of data subsets; The scaling factor for the i-th set of reference parameters. Let be the number of samples in the i-th subset.

[0068] In this embodiment, the sample sizes for the five subsets are 1353, 548, 1547, 1659, and 797, respectively, with corresponding proportionality coefficients of 0.229, 0.093, 0.262, 0.281, and 0.135. These proportionality coefficients reflect the probability weights of each load range in actual operation, making the subsequent sensitivity analysis results more consistent with the distribution characteristics of the actual operating conditions of the system.

[0069] Next, the sensitivity coefficient is calculated based on a preset uncertainty (set to 5% in this embodiment). For each baseline parameter combination, one of the system operating parameters is adjusted by 5%, while keeping other parameters unchanged. This adjusted value is then substituted into the trained data model to obtain the corresponding predicted cooling load. The sensitivity coefficient of this operating parameter in that baseline parameter combination is then calculated. ;in, For system operating parameters In the combination of reference parameters Sensitivity coefficient in For running parameters The original value; This is the original value of the cooling load. For running parameters The value after adjustment by 5%, For running parameters Adjust by 5% and input the cooling load value calculated by the data model.

[0070] This sensitivity coefficient reflects the degree of relative change in cooling load caused by a unit relative change in operating parameter j under the load level represented by the baseline parameter combination i. The larger the coefficient, the more sensitive the parameter is to the influence of cooling load.

[0071] Then, the sensitivity coefficients of each operating parameter under each combination of baseline parameters are weighted and summed according to their corresponding scaling factors to obtain the weighted sensitivity coefficient of that operating parameter: ;in, The weighted sensitivity coefficient for the running parameter j; The number of data subsets; is the scaling factor for the i-th set of reference parameters.

[0072] Furthermore, in some preferred embodiments of the present invention, the system operating parameters include: chilled water supply temperature, chilled water return temperature, cooling water supply temperature, cooling water return temperature, bypass pipe flow rate, chilled water flow rate, terminal air supply temperature, and indoor temperature.

[0073] Specifically, the eight categories of system operating parameters mentioned above cover the main measurable physical quantities on the chiller side, cooling side, and air side of a multi-chiller system. They respectively reflect the operating status of different stages of the refrigeration cycle and the terminal heat exchange effect. Chilled water supply temperature and chilled water return temperature reflect the cooling capacity of the chiller side and the heat absorption at the terminal; cooling water supply temperature and cooling water return temperature reflect the heat dissipation efficiency of the cooling side; bypass pipe flow rate and chilled water flow rate reflect the water system's distribution characteristics; and terminal supply air temperature and indoor temperature are directly related to the terminal air conditioning effect and user comfort. Depending on the different multi-chiller system configurations and sensor availability, all of these system operating parameters can be selected, or some parameters can be selected for inclusion in the data model and sensitivity analysis, as long as a reasonable mapping relationship exists between them and the terminal cooling load.

[0074] Therefore, by selecting a variety of operating parameters covering the refrigeration side, cooling side, and air side, the operating status of each link in the multi-cooling system can be comprehensively reflected, providing rich input features for the data model and improving the accuracy of cooling load prediction.

[0075] In this embodiment, the weighted sensitivity coefficients of the eight operating parameters are calculated as follows: chilled water supply temperature 0.985, chilled water return temperature 0.563, cooling water supply temperature 0.489, cooling water return temperature 0.512, bypass pipe flow rate 0.946, chilled water flow rate 0.732, terminal air supply temperature 0.889, and indoor temperature 0.824.

[0076] Finally, based on the weighted sensitivity coefficients, the parameters are sorted from largest to smallest, and the top-ranked parameters are selected as high-impact indicators. In this embodiment, the top four high-impact indicators are selected: chilled water supply temperature, bypass flow rate, terminal air supply temperature, and indoor temperature. As an optional implementation, the number of high-impact indicators can be flexibly adjusted according to actual needs. For example, the top three parameters—chilled water supply temperature, bypass flow rate, and terminal air supply temperature—can be selected to simplify the scale of monitoring parameters; alternatively, the top five or even all eight parameters can be selected to more comprehensively capture the impact of measurement inaccuracies, but this will correspondingly increase the computational complexity.

[0077] Therefore, by using historical data to train the data model, and dividing the load range based on the number of chillers and rated capacity, and calculating the weighted sensitivity coefficient, it is possible to select the high-impact indicators that are most sensitive to measurement uncertainty for different multi-chiller systems, thereby improving the pertinence and effectiveness of subsequent dynamic threshold correction and laying the foundation for reliable control during online operation.

[0078] Furthermore, in some preferred embodiments of the present invention, selecting at least one set of data samples as a benchmark parameter combination from the dataset based on the number of chillers and the rated capacity of each chiller, and determining the proportional coefficient of each set of benchmark parameters includes: dividing the historical cooling load dataset into multiple data subsets; wherein the number of data subsets is equal to the total number of chillers; assigning all data samples whose cooling load values ​​are within the rated capacity of the first chiller to the first data subset, assigning all data samples whose cooling load values ​​are within the range from the rated capacity of the first chiller to the sum of the rated capacities of the first and second chillers to the second data subset, and so on, until all data subsets are divided; determining the average value of the cooling load values ​​in each data subset, and selecting a set of data samples whose cooling load values ​​are closest to the average value to form a benchmark parameter combination; and determining the proportional coefficient based on the benchmark parameter combination and the number of samples in the data subset.

[0079] Specifically, this method of selecting benchmark parameter combinations is designed based on the load distribution characteristics of chiller groups in actual operation. When the terminal cooling load demand is low and can be met by only one chiller, the state characteristics of the system operating parameters differ significantly from those of multiple chillers operating simultaneously. Therefore, the dataset needs to be stratified according to load level. The historical cooling load dataset is divided into multiple data subsets of equal size based on the total number of chillers. Each subset corresponds to a load range from one chiller to all chillers being put into operation. In this way, the samples in each subset correspond to similar numbers of chillers in operation and system operating conditions. A set of data samples with cooling load values ​​closest to the average value of that subset is selected as a typical representative of that load range, so that the benchmark parameter combination can effectively cover the entire operating range from low load to high load. The proportional coefficient is determined based on the proportion of the number of samples in each subset to the total number of samples, reflecting the frequency of different load ranges in actual operation, thereby ensuring that the calculated results of the weighted sensitivity coefficient match the actual operating conditions.

[0080] Therefore, by dividing the load range according to the number of chillers and determining the proportional coefficient based on the sample size, the combination of benchmark parameters can represent typical operating conditions under different load levels, and the weight allocation is consistent with the actual operating distribution, thus ensuring the representativeness and reliability of the sensitivity analysis results.

[0081] Furthermore, in some preferred embodiments of the present invention, the sensitivity coefficient of each system operating parameter in each benchmark parameter combination is determined according to the preset uncertainty, and the weighted sensitivity coefficient of each system operating parameter is determined based on the proportional coefficient, including: for each benchmark parameter combination, after fine-tuning one of the system operating parameter values ​​according to the uncertainty, substituting it into the data model to obtain the corresponding cooling load value, and determining the sensitivity coefficient of the system operating parameter through the following formula; ;in, For system operating parameters In the combination of reference parameters Sensitivity coefficient in For running parameters The original value; This is the original value of the cooling load. For running parameters The adjusted value For running parameters Fine-tune the cooling load value calculated by the input data model; determine the weighted sensitivity coefficient using the following formula: ;in, The weighted sensitivity coefficient for the running parameter j; The number of data subsets; is the scaling factor for the i-th set of reference parameters.

[0082] Specifically, the preset uncertainty is used to simulate the measurement deviation amplitude that the sensor may experience in actual operation, and in this embodiment it is set to 5%. In the sensitivity coefficient calculation, each system operating parameter in each reference parameter combination is finely adjusted by the same amount (increased or decreased by 5%), while keeping the other parameters unchanged. Then, the ratio of the rate of change of the predicted cooling load value to the rate of change of the parameter is observed to measure the sensitivity of the parameter to the influence of the cooling load.

[0083] Sensitivity coefficient A larger value indicates that, within the load range represented by the baseline parameter combination i, a small measurement deviation in operating parameter j will lead to a larger deviation in the cooling load calculation. Therefore, this parameter has a high requirement for measurement accuracy within this load range. Weighted Sensitivity Coefficient The sensitivity of operating parameter j across all load intervals is synthesized by weighting the probability of samples appearing in each interval, thus yielding the overall sensitivity of the parameter to measurement uncertainty throughout the entire system operating range.

[0084] The weighted sensitivity coefficients calculated in this embodiment are ranked as follows: chilled water supply temperature (0.985) > bypass flow rate (0.946) > terminal air supply temperature (0.889) > indoor temperature (0.824) > chilled water flow rate (0.732) > chilled water return temperature (0.563) > cooling water return temperature (0.512) > cooling water supply temperature (0.489). This ranking indicates that the accuracy of chilled water supply temperature measurement has the greatest impact on cooling load calculation, and therefore it is listed as the most important high-impact indicator.

[0085] Therefore, by quantifying the impact of fine-tuning of each operating parameter on the predicted cooling load in each typical load range and weighting and synthesizing them, the measurement sensitivity of different operating parameters in the full operating range of the system can be accurately assessed, providing a quantitative basis for screening high-impact indicators.

[0086] Furthermore, in some preferred embodiments of the present invention, comparing the online operating parameter values ​​with a preset judgment threshold to determine the high error flag corresponding to the high-impact index, and determining the chiller start-stop threshold correction direction coefficient based on the high error flag includes: if the online operating parameter value is greater than the judgment threshold, then setting the corresponding high error flag to 1; otherwise, setting it to 0; the chiller start-stop threshold correction direction coefficient is determined by the following formula: ;in, Correction of directional coefficient for chiller start / stop threshold; This is a flag indicating the cumulative high error. This represents the number of high error flags. This is the high error flag for the i-th system operating parameter.

[0087] Specifically, the high error flag provides a simple binary judgment for the real-time status of each high-impact indicator. When the online operating parameter value of a high-impact indicator exceeds the judgment threshold determined based on its historical statistical characteristics, it indicates that the current measurement value of the indicator has deviated from the normal range, possibly due to sensor drift, noise interference, or other reasons causing measurement inaccuracies. In this case, its high error flag is set to 1. Cumulative high error flag. Sum the high error flags of all high-impact indicators, with values ​​ranging from 0 to m.

[0088] when When the measured values ​​of all high-impact indicators within the current cycle are within the normal range, the measurement results of the terminal cooling load are highly reliable. At this time, the chiller start / stop threshold correction direction coefficient can be set. This guides the dynamic adjustment factor of the chiller capacity to be updated towards an increasing direction, making the chiller start-up and shutdown thresholds closer to the rated capacity level; when When this occurs, it indicates that at least one high-impact indicator has deviated significantly from its measured value, suggesting a potential overestimation of the terminal cooling load measurement. In this case, setting... This guides the dynamic adjustment factor of the chiller capacity to be updated in a decreasing direction, thereby lowering the chiller start-up and shutdown threshold and avoiding unnecessary chiller startup due to excessively high measured values.

[0089] Therefore, by comparing the online operating parameter values ​​with the judgment threshold based on historical statistical characteristics and calculating the cumulative high error flag to determine the correction direction coefficient, the occurrence and direction of measurement deviation can be identified in real time, providing an accurate decision basis for subsequent correction and avoiding misjudgments caused by measurement deviation.

[0090] Furthermore, to verify the actual control performance of the multi-chill system start-up and shutdown control method provided in this embodiment, a traditional deterministic start-up and shutdown control method was used for performance comparison. This method does not consider any measurement inaccuracies, directly uses the measured value of the cooling load as the strategy input, and compares it with the rated capacity of multiple chillers to determine the current number of chillers to start / stop. To simulate a real uncertainty environment, sensor bias and measurement noise were randomly added to the raw data of a certain week to form a test data verification set with measurement errors.

[0091] See Figure 3 The diagram illustrates a comparison of daily operating energy consumption of a multi-chiller system under different control methods provided in this embodiment of the invention. As can be seen from the diagram, the traditional method ignores the existence of uncertainties in the actual measurement environment, and under certain operating conditions, it overestimates the actual cooling load demand, leading to an excessive number of chillers operating, resulting in energy waste and increased overall operating energy consumption of the multi-chiller system. In contrast, the method provided in this embodiment corrects the chiller start-up and shutdown judgment by evaluating measurement uncertainty parameters and dynamically adjusting chiller start-up and shutdown thresholds, thereby effectively mitigating the impact of measurement uncertainty on the multiple operation of the chiller group, and reducing its overall operating energy consumption by 6.6%.

[0092] See Figure 4 The diagram illustrates a comparison of the supply of terminal cooling load demand in a multi-cooler system under different control methods provided in this embodiment of the invention. Specifically, the load shortfall rate is used for quantification, which is the proportion of the shortfall in cooling load demand to the total cooling load demand during system operation. As shown in the diagram, the method provided in this embodiment can effectively reduce the underestimation of cooling load demand caused by uncertain environmental conditions, thereby ensuring the supply of terminal cooling load demand and improving the energy supply reliability of the multi-cooler system. In contrast, traditional methods are difficult to effectively estimate the number of cooler units in operation under uncertain environmental conditions, and in some operating conditions, insufficient supply of terminal cooling load demand may occur due to an insufficient number of units in operation.

[0093] This invention provides a start-up and shutdown control method for a multi-cooler system under conditions of inaccurate measurement. Applied to a multi-cooler system, the method includes: during system operation, periodically acquiring online operating parameter values ​​corresponding to preset high-impact indicators in the multi-cooler system; comparing the online operating parameter values ​​with preset judgment thresholds to determine a high-error flag bit corresponding to the high-impact indicator, and determining a cooler start-up and shutdown threshold correction direction coefficient based on the high-error flag bit; determining an adjustment step size and a high-error adjustment coefficient based on the cooler start-up and shutdown threshold correction direction coefficient; determining a cooler capacity dynamic adjustment factor based on the cooler start-up and shutdown threshold correction direction coefficient, adjustment step size, and high-error adjustment coefficient; determining a terminal cooling load judgment threshold for starting and stopping the coolers based on the cooler rated capacity, the current number of coolers in operation, and the cooler capacity dynamic adjustment factor; and determining a system cooler start-up and shutdown strategy based on the relative magnitude of the current cooling load measurement value and the terminal cooling load judgment threshold. This method can maximize system operating efficiency while avoiding frequent start-ups and shutdowns of the cooler group, thereby effectively ensuring the safe and reliable operation of the multi-cooler system.

[0094] Based on the above embodiments, this invention provides a start-stop control system for a multi-cooler system under conditions of inaccurate measurement, applicable to multi-cooler systems. See [link to relevant documentation]. Figure 5 The diagram shown is a structural schematic of a start-stop control system for a multi-cooler system under conditions of inaccurate measurement, provided by an embodiment of the present invention. The device includes: The data acquisition module 310 is used to periodically acquire online operating parameter values ​​corresponding to preset high-impact indicators in the multi-cooler system during the system operation phase. The correction direction coefficient determination module 320 is used to compare the online operating parameter values ​​with the preset judgment threshold, determine the high error flag bit corresponding to the high impact index, and determine the chiller start-stop threshold correction direction coefficient based on the high error flag bit. Data processing module 330 is used to determine the adjustment step size and high error adjustment coefficient based on the directional coefficient of the chiller start-stop threshold correction. The data processing module 330 is also used to determine the dynamic adjustment factor of the chiller capacity based on the chiller start-stop threshold correction direction coefficient, adjustment step size and high error adjustment coefficient; The cooling load judgment threshold determination module 340 is used to determine the terminal cooling load judgment threshold for starting and stopping the chillers based on the rated capacity of the chiller, the current number of chillers in operation, and the dynamic adjustment factor of the chiller capacity. The start-stop strategy determination module 350 is used to determine the start-stop strategy of the system chiller based on the relative magnitude relationship between the current cooling load measurement value and the terminal cooling load judgment threshold.

[0095] Furthermore, in some preferred embodiments of the present invention, the apparatus further includes: a high-impact index determination module, used to acquire historical datasets of operating parameters and historical datasets of cooling load related to the multi-cooler system, and to train a data model between system operating parameters and cooling load using the historical dataset of operating parameters as input and the historical dataset of cooling load as output; to select at least one set of data samples from the dataset as a benchmark parameter combination based on the number of coolers and the rated capacity of each cooler, and to determine the proportional coefficient of each set of benchmark parameters; to determine the sensitivity coefficient of each system operating parameter in each benchmark parameter combination based on a preset uncertainty, and to determine the weighted sensitivity coefficient of each system operating parameter based on the proportional coefficient; to sort the operating parameters according to the weighted sensitivity coefficient, and to select at least one system operating parameter as a high-impact index from largest to smallest.

[0096] Furthermore, in some preferred embodiments of the present invention, the system operating parameters include: chilled water supply temperature, chilled water return temperature, cooling water supply temperature, cooling water return temperature, bypass pipe flow rate, chilled water flow rate, terminal air supply temperature, and indoor temperature.

[0097] Furthermore, in some preferred embodiments of the present invention, the high-impact index determination module is used to divide the historical cooling load dataset into multiple data subsets; wherein the number of data subsets is equal to the total number of chillers; all data samples whose cooling load values ​​are within the rated capacity of the first chiller are assigned to the first data subset, all data samples whose cooling load values ​​are within the range from the rated capacity of the first chiller to the sum of the rated capacities of the first and second chillers are assigned to the second data subset, and so on, until all data subsets are divided; the average value of the cooling load values ​​in each data subset is determined, and a set of data samples whose cooling load values ​​are closest to the average value is selected to form a benchmark parameter combination; a proportional coefficient is determined based on the benchmark parameter combination and the number of samples in the data subsets.

[0098] Furthermore, in some preferred embodiments of the present invention, the high-impact index determination module is used to, for each combination of reference parameters, fine-tune one of the system operating parameter values ​​according to the uncertainty, substitute it into the data model to obtain the corresponding cooling load value, and determine the sensitivity coefficient of the system operating parameter through the following formula; ;in, For system operating parameters In the combination of reference parameters Sensitivity coefficient in For running parameters The original value; This is the original value of the cooling load. For running parameters The adjusted value For running parameters Fine-tune the cooling load value calculated by the input data model; determine the weighted sensitivity coefficient using the following formula: ;in, The weighted sensitivity coefficient for the running parameter j; The number of data subsets; is the scaling factor for the i-th set of reference parameters.

[0099] Furthermore, in some preferred embodiments of the present invention, the high-impact index determination module is used to set the corresponding high-error flag to 1 if the online operating parameter value is greater than the judgment threshold, and otherwise set it to 0; the chiller start-stop threshold correction direction coefficient is determined by the following formula: ;in, Correction of directional coefficient for chiller start / stop threshold; This is a flag indicating the cumulative high error. This represents the number of high error flags. This is the high error flag for the i-th system operating parameter.

[0100] Furthermore, in some preferred embodiments of the present invention, the data processing module 330 is used to determine the adjustment step size, the high error adjustment coefficient, and the chiller capacity dynamic adjustment factor using the following formula: ; ; In the formula, The adjustment step size for the nth adjustment period. The preset attenuation base, This represents the cumulative number of times the chiller has been started and stopped to date. This is the high error adjustment coefficient for the nth adjustment period. This is a preset probability constant; This is the dynamic adjustment factor for the chiller capacity in the nth adjustment cycle.

[0101] Furthermore, in some preferred embodiments of the present invention, the cooling load judgment threshold determination module 340 is used to determine the terminal cooling load judgment threshold using the following formula: ; ;in, Determine the threshold for the terminal cooling load to start a chiller; The threshold for determining the terminal cooling load of a chiller to be shut down; This represents the number of chillers currently running in the system. For the first The rated capacity of the operating chiller. This is the preset dead zone coefficient for cold start-stop.

[0102] Furthermore, in some preferred embodiments of the present invention, the correction direction coefficient determination module 320 is used to determine the judgment threshold using the following formula: ;in, High impact indicator The judgment threshold, High impact indicator The mean in historical datasets; High impact indicator Standard deviation in historical datasets, This is the preset confidence level coefficient.

[0103] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the multi-cooler system start-stop control system under inaccurate measurement conditions described above can be referred to the corresponding process in the embodiments of the multi-cooler system start-stop control method under inaccurate measurement conditions, and will not be repeated here.

[0104] This invention also provides an electronic device for controlling the start-up and shutdown of a multi-cooler system operating under conditions of inaccurate measurement; see also Figure 6 The schematic diagram of an electronic device provided by the embodiment of the present invention shown above includes a memory 400 and a processor 401. The memory 400 is used to store one or more computer instructions, which are executed by the processor 401 to realize the above-mentioned start-up and shutdown control method for a multi-cooler system under inaccurate measurement conditions.

[0105] Furthermore, Figure 6 The electronic device shown also includes a bus 402 and a communication interface 403. The processor 401, the communication interface 403 and the memory 400 are connected via the bus 402.

[0106] The memory 400 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 403 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc. The bus 402 may be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 6 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0107] Processor 401 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 401 or by instructions in software form. Processor 401 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software module can reside in a readily available storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory 400, and processor 401 reads information from memory 400 and, in conjunction with its hardware, completes the steps of the method described in the foregoing embodiments.

[0108] This invention also provides a computer-readable storage medium storing computer-executable instructions. When these computer-executable instructions are called and executed by a processor, they cause the processor to implement the above-described start-stop control method for a multi-cooler system under conditions of inaccurate measurement. For specific implementation details, please refer to the method embodiments, which will not be repeated here.

[0109] The computer program product of the start-stop control method, device and electronic device for multi-cooler system under inaccurate measurement environment provided in the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods in the preceding method embodiments. For specific implementation, please refer to the method embodiments, which will not be repeated here.

[0110] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system and / or device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0111] Furthermore, in the description of the embodiments of the present invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.

[0112] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0113] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A start-stop control method for a multi-cooler system under conditions of inaccurate measurement, characterized in that, Applied to multi-cooler systems, the method includes: During system operation, online operating parameter values ​​corresponding to preset high-impact indicators in the multi-cooler system are periodically acquired. Specifically, historical datasets of operating parameters and cooling loads related to the cooling load in the multi-cooler system are acquired. Using the historical datasets of operating parameters as input and the historical datasets of cooling loads as output, a data model between system operating parameters and cooling load is trained. Based on the number of coolers and the rated capacity of each cooler, at least one set of data samples is selected from the dataset as a baseline parameter combination, and the proportional coefficient of each set of baseline parameters is determined. Based on a preset uncertainty, the sensitivity coefficient of each system operating parameter in each baseline parameter combination is determined, and a weighted sensitivity coefficient of each system operating parameter is determined based on the proportional coefficient. The operating parameters are sorted according to the weighted sensitivity coefficients, and at least one system operating parameter is selected as the high-impact indicator from largest to smallest. The online operating parameter values ​​are compared with preset judgment thresholds to determine the high error flag corresponding to the high impact index, and the chiller start-stop threshold correction direction coefficient is determined based on the high error flag. The adjustment step size and high error adjustment coefficient are determined based on the refrigeration start / stop threshold correction direction coefficient. The dynamic adjustment factor of the chiller capacity is determined based on the chiller start / stop threshold correction direction coefficient, the adjustment step size, and the high error adjustment coefficient. The threshold for determining the terminal cooling load for starting and stopping the chillers is determined based on the rated capacity of the chiller, the current number of chillers in operation, and the dynamic adjustment factor of the chiller capacity. Based on the relative magnitude of the current cooling load measurement value and the terminal cooling load judgment threshold, the start-up and shutdown strategy of the system chiller is determined.

2. The method according to claim 1, characterized in that, The system operating parameters include: chilled water supply temperature, chilled water return temperature, cooling water supply temperature, cooling water return temperature, bypass pipe flow rate, chilled water flow rate, terminal air supply temperature, and indoor temperature.

3. The method according to claim 1, characterized in that, Based on the number of chillers and the rated capacity of each chiller, select at least one set of data samples from the dataset as a baseline parameter combination, and determine the scaling factors for each set of baseline parameters, including: The historical cooling load dataset is divided into multiple data subsets; the number of data subsets is equal to the total number of chillers. All data samples with cooling load values ​​within the rated capacity of the first chiller are assigned to the first data subset. All data samples with cooling load values ​​between the rated capacity of the first chiller and the sum of the rated capacities of the first and second chillers are assigned to the second data subset. This process is repeated until all data subsets are divided. Determine the average value of the cooling load in each of the data subsets, and select a set of data samples whose cooling load values ​​are closest to the average value to form the reference parameter combination; The scaling factor is determined based on the combination of the benchmark parameters and the number of samples in the data subset.

4. The method according to claim 3, characterized in that, Based on a preset uncertainty, the sensitivity coefficient of each system operating parameter in each of the aforementioned benchmark parameter combinations is determined, and the weighted sensitivity coefficient of each system operating parameter is determined based on the proportionality coefficient, including: For each of the aforementioned baseline parameter combinations, after fine-tuning one of the system operating parameter values ​​according to the uncertainty, the value is substituted into the data model to obtain the corresponding cooling load value, and the sensitivity coefficient of the system operating parameter is determined by the following formula; ； in, For system operating parameters In the combination of reference parameters Sensitivity coefficient in For running parameters The original value; This is the original value of the cooling load. For running parameters The adjusted value For running parameters Fine-tune the cooling load value calculated by the input data model; The weighted sensitivity coefficient is determined using the following formula: ； in, The weighted sensitivity coefficient for the running parameter j; The number of the data subsets; is the scaling factor for the i-th set of reference parameters.

5. The method according to claim 1, characterized in that, The online operating parameter values ​​are compared with preset judgment thresholds to determine the high-error flag bit corresponding to the high-impact index, and the chiller start-stop threshold correction direction coefficient is determined based on the high-error flag bit, including: If the online operating parameter value is greater than the judgment threshold, then the corresponding high error flag is set to 1; otherwise, it is set to 0. The chiller start / stop threshold correction direction coefficient is determined using the following formula: ； in, The direction coefficient is corrected for the chiller start / stop threshold. This is a flag indicating the cumulative high error. This represents the number of high error flags. This is the high error flag for the i-th system operating parameter.

6. The method according to claim 5, characterized in that, The adjustment step size, the high error adjustment coefficient, and the chiller capacity dynamic adjustment factor are determined using the following formulas: ； ； ； In the formula, The adjustment step size is the adjustment step size for the nth adjustment period. The preset attenuation base, This represents the cumulative number of times the chiller has been started and stopped to date. The high error adjustment coefficient is the one used in the nth adjustment period. This is a preset probability constant; The dynamic adjustment factor for the chiller capacity during the nth adjustment cycle.

7. The method according to claim 6, characterized in that, The threshold for judging the terminal cooling load is determined by the following formula: ； ； in, The threshold value for determining the terminal cooling load to start a chiller; The threshold for determining the terminal cooling load of a chiller is to be shut down. This represents the number of chillers currently running in the system. For the first The rated capacity of the operating chiller. This is the preset dead zone coefficient for cold start-stop.

8. The method according to claim 5, characterized in that, The judgment threshold is determined by the following formula: ； in, High impact indicator The aforementioned judgment threshold, High impact indicator The mean in historical datasets; High impact indicator Standard deviation in historical datasets, This is the preset confidence level coefficient.

9. A start-stop control system for a multi-cooler system under conditions of inaccurate measurement, characterized in that, Applied to multi-cooling systems, including: The data acquisition module is used to periodically acquire online operating parameter values ​​corresponding to preset high-impact indicators in the multi-cooler system during system operation. Specifically, it acquires historical datasets of operating parameters and historical datasets of cooling load related to the cooling load in the multi-cooler system, using the historical datasets of operating parameters as input and the historical datasets of cooling load as output to train a data model between system operating parameters and cooling load. Based on the number of coolers and the rated capacity of each cooler, it selects at least one set of data samples from the dataset as a benchmark parameter combination and determines the proportional coefficient of each benchmark parameter set. Based on a preset uncertainty, it determines the sensitivity coefficient of each system operating parameter in each benchmark parameter combination and determines the weighted sensitivity coefficient of each system operating parameter based on the proportional coefficient. Finally, it sorts the operating parameters according to the weighted sensitivity coefficients and selects at least one system operating parameter from largest to smallest as the high-impact indicator. The correction direction coefficient determination module is used to compare the online operating parameter value with a preset judgment threshold, determine the high error flag bit corresponding to the high impact index, and determine the chiller start-stop threshold correction direction coefficient based on the high error flag bit. The data processing module is used to determine the adjustment step size and high error adjustment coefficient based on the chiller start-stop threshold correction direction coefficient; The data processing module is also used to determine the dynamic adjustment factor of the chiller capacity based on the chiller start-stop threshold correction direction coefficient, the adjustment step size, and the high error adjustment coefficient; The cooling load judgment threshold determination module is used to determine the terminal cooling load judgment threshold for starting and stopping the chillers based on the rated capacity of the chiller, the current number of chillers in operation, and the dynamic adjustment factor of the chiller capacity. The start-stop strategy determination module is used to determine the start-stop strategy of the system chiller based on the relative magnitude relationship between the current cooling load measurement value and the terminal cooling load judgment threshold.

Images