Energy-saving control method of centrifugal water chilling unit

Abstract

The invention belongs to the technical field of refrigeration equipment control, and discloses an energy-saving control method of a centrifugal water chilling unit. Cooperative adjustment of the rotating speed of a compressor and the opening degree of a guide vane is achieved by obtaining operation state parameters and environment state parameters of a water chilling unit in real time, dynamically predicting the surge boundary, calculating the operation safety margin and generating an energy-saving optimization control vector. The method comprises the following steps: firstly, acquiring operation parameters such as compressor rotating speed, guide vane opening, condenser pressure and evaporator pressure and environment parameters such as environment temperature and cooling water temperature, and accurately predicting a surge boundary based on a pressure ratio, an operation condition vector and an environment correction coefficient; then analyzing the surge boundary change trend and the running state fluctuation characteristics, and determining the safety margin; and main characteristic components of the operation state are further extracted through characteristic decomposition, energy saving and safety requirements are balanced in combination with a dynamic tradeoff factor, and an optimization control vector is generated. According to the invention, the surge risk can be effectively prevented while the energy consumption is reduced, and the system adapts to complex and changeable operating environments.

Description

Technical Field

[0001] The present invention relates to the technical field of refrigeration equipment control, and more specifically, to an energy-saving control method for a centrifugal chiller. Background Art

[0002] As the core refrigeration equipment in large commercial buildings, industrial facilities, data centers and other places, centrifugal chillers are widely used due to their high efficiency, large capacity and stable performance. At present, the proportion of centrifugal chillers in building energy consumption is increasing day by day, and their operating efficiency directly affects the overall energy consumption and operating costs.

[0003] However, the existing control technology for centrifugal chillers has certain defects in practical applications, especially in variable load and multi-condition environments. Traditional control systems generally use fixed surge boundaries as the safety control benchmark, which cannot adapt to complex external conditions such as cooling water temperature fluctuations and ambient temperature changes, resulting in the system frequently entering the protection state or a significant drop in efficiency in extreme weather conditions such as high temperatures in summer or low temperatures in winter. In actual engineering, to avoid surge risks, maintenance personnel often have to adopt overly conservative safety margin settings, causing the unit to operate at operating points far from the high-efficiency region for a long time, resulting in a large amount of energy waste. Especially in large commercial buildings and industrial cooling systems, this waste is particularly significant. At the same time, the existing control strategies lack a coordinated optimization mechanism for compressor speed and guide vane opening, showing obvious energy efficiency degradation during partial load operation and being difficult to accurately respond to dynamic load changes such as morning and evening peaks and valley periods in office buildings. More seriously, when the system encounters sudden load fluctuations, due to insufficient prediction accuracy, the control system either overreacts and causes energy waste, or under-responds and leads to unstable refrigeration effects, and even causes surges in extreme cases, endangering equipment safety. This passive balance between energy conservation and safety severely restricts the improvement of building energy efficiency and the reduction of operating costs.

[0004] In view of this, the present invention proposes an energy-saving control method for a centrifugal chiller to solve the above problems. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art and to achieve the above object, the present invention provides the following technical solution: An energy-saving control method for a centrifugal chiller, comprising:

[0006] Obtain the operating state parameters and environmental state parameters of the centrifugal chiller during operation, where the operating state parameters include compressor speed, guide vane opening, condenser pressure, evaporator pressure, and the environmental state parameters include ambient temperature and cooling water temperature;

[0007] According to the operating state parameters and the environmental state parameters, obtain the predicted value of the surge boundary of the centrifugal chiller at each moment;

[0008] Based on the change trend of the predicted surge boundary value and the fluctuation characteristics of the operating state parameters within the analysis period at each moment, obtain the operating safety margin of the centrifugal chiller at each moment;

[0009] According to the operating safety margin and the dynamic distribution characteristics of the operating state parameters at each moment, obtain the energy-saving optimization control vector of the centrifugal chiller at each moment; adjust the compressor speed and the guide vane opening based on the energy-saving optimization control vector to achieve the energy-saving operation of the centrifugal chiller.

[0010] Further, the obtaining of the predicted surge boundary value of the centrifugal chiller at each moment includes:

[0011] Calculate the pressure ratio of the centrifugal chiller at each moment according to the condenser pressure and the evaporator pressure;

[0012] Construct the operating condition vector of the centrifugal chiller at each moment according to the pressure ratio, the compressor speed and the guide vane opening;

[0013] Based on the operating condition vector and a preset surge boundary model, obtain the initial surge boundary value of the centrifugal chiller at the current moment;

[0014] Calculate the environmental correction coefficient of the centrifugal chiller at each moment according to the ambient temperature and the cooling water temperature;

[0015] Use the environmental correction coefficient to correct the initial surge boundary value to obtain the predicted surge boundary value of the centrifugal chiller at each moment.

[0016] Further, the obtaining of the operating safety margin of the centrifugal chiller at each moment includes:

[0017] Statistically calculate the change rate of the predicted surge boundary value within the analysis period at each moment, denoted as the surge boundary change trend value;

[0018] Calculate the fluctuation variance of the operating state parameters within the analysis period at each moment, denoted as the operating state fluctuation value;

[0019] Construct the operating risk characteristic vector of the centrifugal chiller at each moment according to the surge boundary change trend value and the operating state fluctuation value;

[0020] Normalize the operating risk characteristic vector and combine it with a preset safety margin weight to obtain the operating safety margin of the centrifugal chiller at each moment.

[0021] Further, obtaining the energy-saving optimization control vector of the centrifugal chiller at each moment includes:

[0022] Determining the operating safety boundary of the centrifugal chiller at each moment according to the operating safety margin;

[0023] Based on the dynamic distribution characteristics of the operating safety boundary and the operating state parameters, constructing the operating state characteristic matrix of the centrifugal chiller at each moment;

[0024] Performing eigenvalue decomposition on the operating state characteristic matrix, extracting the main eigencomponents, and obtaining the energy-saving optimization direction of the centrifugal chiller at each moment;

[0025] Generating the energy-saving optimization control vector of the centrifugal chiller at each moment according to the energy-saving optimization direction and the operating safety margin, where the energy-saving optimization control vector includes the compressor speed adjustment amount and the guide vane opening adjustment amount.

[0026] Further, adjusting the compressor speed and the guide vane opening based on the energy-saving optimization control vector includes:

[0027] Obtaining the actual operating condition parameters of the centrifugal chiller at the current moment;

[0028] Calculating the compressor speed target value and the guide vane opening target value of the centrifugal chiller at the next moment according to the energy-saving optimization control vector;

[0029] Generating a control instruction for the centrifugal chiller based on the actual operating condition parameters, the compressor speed target value, and the guide vane opening target value;

[0030] Adjusting the compressor speed and the guide vane opening of the centrifugal chiller through the control instruction.

[0031] Further, the obtaining method of the preset surge boundary model includes:

[0032] Collecting the historical operation data of the centrifugal chiller under different operating conditions, where the historical operation data includes the compressor speed, the guide vane opening, the condenser pressure, the evaporator pressure, and the surge occurrence state;

[0033] Constructing an operating condition sample set of the centrifugal chiller according to the historical operation data;

[0034] Training the operating condition sample set by using a machine learning algorithm to obtain the preset surge boundary model, and the surge boundary model is used to characterize the mapping relationship between the operating condition vector and the surge boundary value.

[0035] Further, the calculation method of the fluctuation variance of the operating state parameters includes:

[0036] Statistically analyze the time series data of the compressor speed, the guide vane opening, the condenser pressure, and the evaporator pressure within the analysis period at each moment;

[0037] Perform normalization processing on the time series data to obtain a normalized operating parameter sequence;

[0038] Calculate the variance of the normalized operating parameter sequence to obtain the operating state fluctuation value.

[0039] Further, the construction method of the operating state feature matrix includes:

[0040] Extract the operating state distribution characteristics of the centrifugal chiller at each moment according to the dynamic distribution characteristics of the operating state parameters. The operating state distribution characteristics include the joint distribution probability of the compressor speed and the guide vane opening;

[0041] Determine the safe operating constraint conditions of the centrifugal chiller at each moment according to the operating safety boundary;

[0042] Matrixify the operating state distribution characteristics and the safe operating constraint conditions to obtain the operating state feature matrix.

[0043] Further, the machine learning algorithm is a support vector machine algorithm or a deep neural network algorithm.

[0044] Further, calculating the target values of the compressor speed and the guide vane opening of the centrifugal chiller at the next moment according to the energy-saving optimization control vector includes:

[0045] Obtain the actual operating condition parameters of the centrifugal chiller at the current moment and the energy-saving optimization control vector;

[0046] Predict the operating condition trend vector of the centrifugal chiller at the next moment according to the actual operating condition parameters. The operating condition trend vector includes the compressor speed change trend and the guide vane opening change trend;

[0047] Calculate the initial target values of the compressor speed and the initial target value of the guide vane opening of the centrifugal chiller at the next moment according to the operating condition trend vector and the energy-saving optimization control vector;

[0048] Determine the dynamic trade-off factor of the centrifugal chiller at the next moment based on the operating safety margin and the environmental state parameters. The dynamic trade-off factor is used to characterize the priority between energy-saving optimization and operating safety;

[0049] Obtain the predicted deviation correction value of the centrifugal chiller at the next moment according to the deviation between the actual operating condition parameters and the historical operating data;

[0050] Use the dynamic trade-off factor to perform weighted adjustment on the initial compressor speed target value and the initial guide vane opening target value, and combine the predicted deviation correction value for correction to obtain the compressor speed target value and the guide vane opening target value of the centrifugal chiller at the next moment.

[0051] The technical effects and advantages of an energy-saving control method for a centrifugal chiller according to the present invention:

[0052] By accurately grasping the dynamically changing surge critical point, the unit of the present invention can operate safely in a high-efficiency region closer to the surge boundary, tapping the deep energy-saving potential that is difficult to reach by traditional control methods. The present invention no longer relies on a conservative static safety boundary, but instead perceives and accurately predicts the surge risk in real time, significantly reducing the energy waste caused by overprotection, while effectively reducing the risk of equipment damage caused by surges, and solving the traditional contradiction between energy saving and safety. In terms of energy efficiency, through the precise coordinated control of the compressor speed and the guide vane opening, the system always maintains the best energy efficiency ratio condition, and can maintain high-efficiency operation even under severe conditions of drastic fluctuations in the external environment, achieving a significant reduction in energy consumption. This adaptive optimization control strategy makes the chiller more sensitive and accurate in responding to load fluctuations, avoiding over-regulation and energy loss in traditional control. In the long-term operation, the present invention significantly reduces the frequency of surge events in the system, significantly extends the service life of key components, reduces maintenance costs and unplanned downtime, and at the same time realizes the continuous optimization of energy consumption. Brief Description of the Drawings

[0053] Figure 1 It is a step flowchart of an energy-saving control method for a centrifugal chiller according to the present invention. Detailed Embodiments

[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0056] The following will specifically describe the specific solution of an energy-saving control method for a centrifugal chiller provided by the present invention with reference to the accompanying drawings.

[0057] The present invention proposes an energy-saving control method for a centrifugal chiller. Please refer to Figure 1 , which shows a step flowchart of an energy-saving control method for a centrifugal chiller provided by an embodiment of the present invention. The method includes:

[0058] Step S1: Obtain the operating state parameters and environmental state parameters of the centrifugal chiller during operation. The operating state parameters include compressor speed, guide vane opening, condenser pressure, and evaporator pressure. The environmental state parameters include ambient temperature and cooling water temperature.

[0059] During the operation of the centrifugal chiller, the operating state parameters and environmental state parameters are collected in real time by sensors installed on the centrifugal chiller. The operating state parameters include compressor speed, guide vane opening, condenser pressure, and evaporator pressure. The environmental state parameters include ambient temperature and cooling water temperature. The sensors are used to continuously collect the operating state parameters and environmental state parameters of the centrifugal chiller at each moment, and the collected operating state parameters and environmental state parameters are denoised and data-normalized.

[0060] It should be noted that the sensors installed on the centrifugal chiller include but are not limited to speed sensors, pressure sensors, temperature sensors, etc., to ensure that the collection requirements of the operating state parameters and environmental state parameters can be fully covered. The sampling frequencies of the operating state parameters and environmental state parameters are the same.

[0061] In an implementation manner of the embodiment of the present invention, the sampling frequency is set to once every 1 second.

[0062] In an implementation manner of the embodiment of the present invention, the Kalman filter algorithm is selected for denoising, and the Z-score normalization method is used for data normalization. The specific methods of the above preprocessing are not introduced here, and they are all technical means well-known to those skilled in the art. Other data collection devices and data preprocessing algorithms can also be selected, which are not limited here.

[0063] The following steps are all analyzed using the preprocessed operating state parameters and environmental state parameters.

[0064] Step S2: Obtain the predicted surge boundary value of the centrifugal chiller at each moment according to the operating state parameters and environmental state parameters.

[0065] During the operation of a centrifugal chiller, the accurate prediction of the surge boundary is crucial for energy-saving operation and operation safety. However, the changes in environmental state parameters such as ambient temperature and cooling water temperature will cause dynamic changes in the surge boundary. Traditional control methods are usually based on fixed surge boundaries and are difficult to adapt to complex and variable environmental conditions, affecting the energy-saving effect and operation stability. Since the prediction of the surge boundary requires comprehensive consideration of the dynamic effects of operation state parameters and environmental state parameters, based on the multi-dimensional analysis of operation state parameters and environmental state parameters, the predicted value of the surge boundary of the centrifugal chiller at each moment is determined to improve the accuracy of energy-saving control in complex and variable environments.

[0066] Preferably, in some possible implementation manners of the embodiments of the present invention, obtaining the predicted value of the surge boundary of the centrifugal chiller at each moment includes: calculating the pressure ratio of the centrifugal chiller at each moment according to the condenser pressure and the evaporator pressure; constructing an operation condition vector of the centrifugal chiller at each moment according to the pressure ratio, the compressor speed and the guide vane opening; obtaining the initial surge boundary value of the centrifugal chiller at the current moment based on the operation condition vector and a preset surge boundary model; calculating the environmental correction coefficient of the centrifugal chiller at each moment according to the ambient temperature and the cooling water temperature; and correcting the initial surge boundary value by using the environmental correction coefficient to obtain the predicted value of the surge boundary of the centrifugal chiller at each moment.

[0067] The pressure ratio reflects the pressure relationship between the condenser and the evaporator during the operation of the centrifugal chiller and is an important indicator for judging the surge boundary; the compressor speed and the guide vane opening directly affect the operation conditions of the centrifugal chiller. By constructing an operation condition vector, the operation state of the centrifugal chiller can be comprehensively characterized; the preset surge boundary model is used to characterize the mapping relationship between the operation condition vector and the surge boundary value; the changes in the ambient temperature and the cooling water temperature will affect the thermodynamic performance of the centrifugal chiller. By introducing the environmental correction coefficient, the initial surge boundary value can be dynamically adjusted to improve the prediction accuracy.

[0068] In the embodiments of the present invention, the pressure ratio PR is expressed by the formula:

[0069] ; where P_c is the condenser pressure and P_e is the evaporator pressure.

[0070] In the embodiments of the present invention, the operation condition vector V is expressed by the formula:

[0071] ; where PR is the pressure ratio, N is the compressor speed, and θ is the guide vane opening.

[0072] In the embodiments of the present invention, the environmental correction coefficient α is expressed by the formula:

[0073] ; where T_env is the environmental temperature, T_ref is the reference environmental temperature, T_cw is the cooling water temperature, T_cw_ref is the reference cooling water temperature, and k1 and k2 are preset correction weights.

[0074] In an implementation of the embodiment of the present invention, the reference environmental temperature T_ref is set to 25 degrees Celsius, the reference cooling water temperature T_cw_ref is set to 30 degrees Celsius, the correction weight k1 is set to 0.02, and k2 is set to 0.03.

[0075] In the embodiment of the present invention, the predicted surge boundary value S_pred is expressed by the formula:

[0076] ; where S_init is the initial surge boundary value and α is the environmental correction coefficient.

[0077] Step S3: Based on the change trend of the predicted surge boundary value and the fluctuation characteristics of the operating state parameters within the analysis period at each moment, obtain the operating safety margin of the centrifugal chiller at each moment.

[0078] Since the operation safety of the centrifugal chiller needs to balance energy-saving optimization and surge risk, the accurate evaluation of the operating safety margin is crucial for the control strategy. Traditional control methods usually rely on a fixed safety margin and are difficult to adapt to the dynamic changes of the predicted surge boundary value and the fluctuation characteristics of the operating state parameters, resulting in either overly conservative or overly aggressive operation. Therefore, according to the change trend of the predicted surge boundary value and the fluctuation characteristics of the operating state parameters within the analysis period, analyze the operating risk of the centrifugal chiller to obtain the operating safety margin for guiding energy-saving optimization control.

[0079] Preferably, in some possible implementations of the embodiment of the present invention, obtaining the operating safety margin of the centrifugal chiller at each moment includes: counting the change rate of the predicted surge boundary value within the analysis period at each moment, denoted as the surge boundary change trend value; calculating the fluctuation variance of the operating state parameters within the analysis period at each moment, denoted as the operating state fluctuation value; constructing an operating risk characteristic vector of the centrifugal chiller at each moment according to the surge boundary change trend value and the operating state fluctuation value; normalizing the operating risk characteristic vector and combining it with the preset safety margin weight to obtain the operating safety margin of the centrifugal chiller at each moment.

[0080] The change rate of the predicted surge boundary reflects the dynamic change trend of the surge boundary. If the change rate is larger, the stability of the surge boundary is lower and the operation risk is higher. The fluctuation variance of the operating state parameters reflects the stability of the operating state. If the fluctuation variance is larger, the uncertainty of the operating state is higher and the operation risk is higher. The operation risk characteristic vector comprehensively characterizes the change trend of the surge boundary and the fluctuation characteristics of the operating state. By normalizing and combining the safety margin weights, the size of the operating safety margin can be quantified.

[0081] In the embodiment of the present invention, the change trend value ΔS of the surge boundary is expressed by the formula:

[0082] ; where S_pred(t) is the predicted value of the surge boundary at the current time t, S_pred(t - 1) is the predicted value of the surge boundary at the previous time of the current time t, and Δt is the time interval.

[0083] In the embodiment of the present invention, the fluctuation value σ of the operating state is expressed by the formula:

[0084] ; where X_i is the value of the operating state parameter at the i-th time within the analysis period, μ is the mean value of the operating state parameters within the analysis period, and n is the number of times within the analysis period.

[0085] In the embodiment of the present invention, the operation risk characteristic vector R is expressed by the formula:

[0086] ;

[0087] In the embodiment of the present invention, the operating safety margin M is expressed by the formula:

[0088] ; where ΔS is the change trend value of the surge boundary, σ is the fluctuation value of the operating state, w1 and w2 are preset safety margin weights, and Norm is a normalization function. It should be noted that the normalization process is used to map the operation risk characteristic vector to the interval [0, 1] to quantify the size of the operating safety margin.

[0089] In one implementation manner of the embodiment of the present invention, the analysis period contains 10 times, and each time is the last time within its analysis period.

[0090] In one implementation manner of the embodiment of the present invention, the preset safety margin weight w1 is set to 0.6 and w2 is set to 0.4.

[0091] It should be noted that in the embodiments of the present invention, the Norm function is used to normalize the operation risk feature vector. In the embodiments of the present invention, other normalization methods can also be selected, such as min-max normalization, Z-score standardization and other normalization methods, which are not limited herein.

[0092] Step S4: According to the dynamic distribution characteristics of the operation safety margin and operation state parameters at each moment, obtain the energy-saving optimization control vector of the centrifugal chiller at each moment; based on the energy-saving optimization control vector, adjust the compressor speed and guide vane opening to achieve the energy-saving operation of the centrifugal chiller.

[0093] Since the energy-saving operation of the centrifugal chiller needs to balance operation safety and energy consumption optimization, the accurate generation of the energy-saving optimization control vector is crucial for the control strategy. Traditional control methods usually rely on fixed operation safety boundaries or single optimization objectives, and it is difficult to adapt to the dynamic changes of the operation safety margin and the complex distribution characteristics of operation state parameters, resulting in insufficient energy-saving effects or increased operation risks. Therefore, according to the dynamic distribution characteristics of the operation safety margin and operation state parameters, analyze the energy-saving optimization direction of the centrifugal chiller, generate the energy-saving optimization control vector, and use it to adjust the compressor speed and guide vane opening to achieve energy-saving operation.

[0094] Preferably, in some possible implementation manners of the embodiments of the present invention, obtaining the energy-saving optimization control vector of the centrifugal chiller at each moment includes: determining the operation safety boundary of the centrifugal chiller at each moment according to the operation safety margin; constructing the operation state feature matrix of the centrifugal chiller at each moment based on the operation safety boundary and the dynamic distribution characteristics of the operation state parameters; performing eigen-decomposition on the operation state feature matrix, extracting the main feature components, and obtaining the energy-saving optimization direction of the centrifugal chiller at each moment; generating the energy-saving optimization control vector of the centrifugal chiller at each moment according to the energy-saving optimization direction and the operation safety margin, and the energy-saving optimization control vector includes the compressor speed adjustment amount and the guide vane opening adjustment amount.

[0095] The operation safety margin reflects the operation risk level of the centrifugal chiller at the current moment. By determining the operation safety boundary, constraints can be provided for energy-saving optimization; the dynamic distribution characteristics of the operation state parameters reflect the joint distribution law of parameters such as compressor speed and guide vane opening. By constructing the operation state feature matrix, the multi-dimensional characteristics of the operation state can be comprehensively characterized; eigen-decomposition is used to extract the main feature components of the operation state feature matrix to determine the energy-saving optimization direction; the energy-saving optimization control vector combines the energy-saving optimization direction and the operation safety margin to achieve the coordinated adjustment of the compressor speed and the guide vane opening.

[0096] In the embodiments of the present invention, the operation safety boundary B is expressed by the formula:

[0097] ; where S_pred is the predicted value of the surge boundary and M is the operating safety margin.

[0098] In the embodiments of the present invention, the construction method of the operating state feature matrix A will be further described in the subsequent steps.

[0099] In the embodiments of the present invention, the energy-saving optimization control vector C is expressed by the formula:

[0100] ; where ΔN is the adjustment amount of the compressor speed and Δθ is the adjustment amount of the guide vane opening.

[0101] In one implementation manner of the embodiments of the present invention, the feature decomposition uses the singular value decomposition (SVD) method to extract the main feature components, and other feature decomposition methods such as the principal component analysis (PCA) can also be used, which is not limited herein.

[0102] Adjusting the compressor speed and the guide vane opening based on the energy-saving optimization control vector includes: obtaining the actual operating condition parameters of the centrifugal chiller at the current moment; calculating the target value of the compressor speed and the target value of the guide vane opening of the centrifugal chiller at the next moment according to the energy-saving optimization control vector; generating a control instruction for the centrifugal chiller based on the actual operating condition parameters, the target value of the compressor speed, and the target value of the guide vane opening; and adjusting the compressor speed and the guide vane opening of the centrifugal chiller through the control instruction to achieve energy-saving operation.

[0103] The actual operating condition parameters reflect the actual operating state of the centrifugal chiller at the current moment. By combining the energy-saving optimization control vector, the target value of the compressor speed and the target value of the guide vane opening at the next moment can be accurately calculated; the control instruction is used to convert the calculation result into an executable control signal, thereby achieving energy-saving operation.

[0104] In the embodiments of the present invention, the actual operating condition parameters include the compressor speed, the guide vane opening, the condenser pressure, and the evaporator pressure.

[0105] In the embodiments of the present invention, the generation method of the control instruction will be further described in the subsequent steps.

[0106] Step S5: Obtaining a preset surge boundary model.

[0107] Preferably, in some possible implementation manners of the embodiments of the present invention, the method for obtaining the preset surge boundary model includes: collecting historical operation data of the centrifugal chiller under different operating conditions, where the historical operation data includes compressor speed, guide vane opening, condenser pressure, evaporator pressure, and surge occurrence status; constructing an operating condition sample set of the centrifugal chiller according to the historical operation data; training the operating condition sample set using a machine learning algorithm to obtain the preset surge boundary model, and the surge boundary model is used to represent the mapping relationship between the operating condition vector and the surge boundary value.

[0108] The historical operation data reflects the actual performance of the centrifugal chiller under different operating conditions. By constructing the operating condition sample set, training data can be provided for the machine learning algorithm; the preset surge boundary model can dynamically predict the surge boundary value during operation by learning the mapping relationship between the operating condition vector and the surge boundary value.

[0109] In the embodiments of the present invention, the operating condition sample set D is expressed by the formula:

[0110] ; where V_i is the i-th operating condition vector, S_i is the corresponding surge boundary value, and n is the number of samples.

[0111] In one implementation manner of the embodiments of the present invention, the machine learning algorithm is a support vector machine algorithm or a deep neural network algorithm.

[0112] In one implementation manner of the embodiments of the present invention, the collected historical operation data covers an operating time of at least 1000 hours and includes at least 10 different operating conditions.

[0113] Step S6: Calculation of the fluctuation variance of the operating state parameters.

[0114] Preferably, in some possible implementation manners of the embodiments of the present invention, the method for calculating the fluctuation variance of the operating state parameters includes: statistically analyzing the time series data of the compressor speed, guide vane opening, condenser pressure, and evaporator pressure within each analysis period; performing standardization processing on the time series data to obtain a standardized operating parameter sequence; calculating the variance of the standardized operating parameter sequence to obtain the operating state fluctuation value.

[0115] The time series data reflects the dynamic changes of the operating state parameters within the analysis period. Through standardization processing, the influence of different parameter dimensions can be eliminated; the fluctuation variance is used to quantify the instability of the operating state parameters and is an important indicator for evaluating the operating safety margin.

[0116] In the embodiments of the present invention, the standardized operating parameter sequence X_std is expressed by the formula:

[0117] ; where X is the time series data, μ is the mean of the time series data, and σ is the standard deviation of the time series data.

[0118] In the embodiment of the present invention, the running state fluctuation value σ is expressed by the formula:

[0119] ; where X_std_i is the i-th value of the standardized operation parameter sequence, and n is the number of time instances within the analysis period.

[0120] In one implementation manner of the embodiment of the present invention, the standardization process adopts the Z-score standardization method, and other standardization methods such as maximum-minimum normalization can also be used, which is not limited herein.

[0121] Step S7: Construction of the running state feature matrix.

[0122] Preferably, in some possible implementation manners of the embodiment of the present invention, the method for constructing the running state feature matrix includes: extracting the running state distribution features of the centrifugal chiller at each moment according to the dynamic distribution characteristics of the running state parameters, where the running state distribution features include the joint distribution probability of the compressor speed and the guide vane opening; determining the safe operation constraint conditions of the centrifugal chiller at each moment according to the running safety boundary; and performing matrix representation on the running state distribution features and the safe operation constraint conditions to obtain the running state feature matrix.

[0123] The dynamic distribution characteristics of the running state parameters reflect the joint variation law of parameters such as the compressor speed and the guide vane opening during the operation of the centrifugal chiller. By extracting the running state distribution features, the multi-dimensional characteristics of the running state can be comprehensively characterized; the running safety boundary provides the constraint conditions for energy-saving optimization. By determining the safe operation constraint conditions, it can be ensured that the energy-saving optimization is carried out within the safe range; the running state feature matrix integrates the running state distribution features and the safe operation constraint conditions through matrix representation, providing a data basis for subsequent feature decomposition.

[0124] In the embodiment of the present invention, the running state distribution feature P is expressed by the formula:

[0125] ; where P(N) is the probability distribution of the compressor speed, and P(θ|N) is the conditional probability distribution of the guide vane opening given the compressor speed.

[0126] In the embodiment of the present invention, the safe operation constraint conditions are expressed by the formula:

[0127] ; where S_pred is the predicted value of the surge boundary, and B is the running safety boundary.

[0128] In the embodiment of the present invention, the operation state feature matrix A is expressed by the formula:

[0129] ; where P(N, θ) is the operation state distribution feature, and S_pred - B is the safe operation constraint condition.

[0130] In an implementation manner of the embodiment of the present invention, the probability distribution of the operation state distribution feature is calculated by the kernel density estimation (KDE) method, and other probability distribution estimation methods such as histogram estimation can also be used, which are not limited herein.

[0131] Step S8: Calculate the target value of the compressor speed and the target value of the guide vane opening of the centrifugal chiller at the next moment according to the energy-saving optimization control vector.

[0132] Preferably, in some possible implementation manners of the embodiment of the present invention, calculating the target value of the compressor speed and the target value of the guide vane opening of the centrifugal chiller at the next moment according to the energy-saving optimization control vector includes: obtaining the actual operation condition parameters and the energy-saving optimization control vector of the centrifugal chiller at the current moment; predicting the operation condition trend vector of the centrifugal chiller at the next moment according to the actual operation condition parameters, where the operation condition trend vector includes the compressor speed change trend and the guide vane opening change trend; calculating the initial target value of the compressor speed and the initial target value of the guide vane opening of the centrifugal chiller at the next moment according to the operation condition trend vector and the energy-saving optimization control vector; determining the dynamic trade-off factor of the centrifugal chiller at the next moment based on the operation safety margin and the environmental state parameters, where the dynamic trade-off factor is used to represent the priority between energy-saving optimization and operation safety; obtaining the predicted deviation correction value of the centrifugal chiller at the next moment according to the deviation between the actual operation condition parameters and the historical operation data; using the dynamic trade-off factor to perform weighted adjustment on the initial target value of the compressor speed and the initial target value of the guide vane opening, and combining the predicted deviation correction value for correction to obtain the target value of the compressor speed and the target value of the guide vane opening of the centrifugal chiller at the next moment.

[0133] The actual operating condition parameters reflect the actual operating state of the centrifugal chiller at the current moment. By predicting the operating condition trend vector, the dynamic change trend of the operating state can be considered, and the adaptability of the target value calculation can be improved; the energy-saving optimization control vector provides the adjustment direction for energy-saving optimization. By combining the operating condition trend vector, the initial compressor speed target value and the initial guide vane opening target value can be calculated; the changes in the operating safety margin and environmental state parameters will affect the priority between energy-saving optimization and operating safety. By introducing a dynamic trade-off factor, an adaptive balance between the two can be achieved; the deviation between the actual operating condition parameters and the historical operating data reflects the error of the prediction model. By introducing a prediction deviation correction value, the accuracy of the target value calculation can be further improved.

[0134] In the embodiment of the present invention, the operating condition trend vector T is expressed by the formula:

[0135] ; where ΔN_trend is the compressor speed change trend, and Δθ_trend is the guide vane opening change trend.

[0136] In the embodiment of the present invention, the compressor speed change trend ΔN_trend and the guide vane opening change trend Δθ_trend are predicted by the linear regression method, and other prediction methods such as time series analysis can also be used, which are not limited herein.

[0137] In the embodiment of the present invention, the initial compressor speed target value N_init and the initial guide vane opening target value θ_init are expressed by the formula:

[0138] ;

[0139] ;

[0140] where N_curr is the compressor speed at the current moment, θ_curr is the guide vane opening at the current moment, ΔN is the compressor speed adjustment amount in the energy-saving optimization control vector, and Δθ is the guide vane opening adjustment amount in the energy-saving optimization control vector.

[0141] In the embodiment of the present invention, the dynamic trade-off factor β is expressed by the formula:

[0142] ; where M is the operating safety margin, T_env is the ambient temperature, T_ref is the reference ambient temperature, T_cw is the cooling water temperature, T_cw_ref is the reference cooling water temperature, k3 and k4 are preset correction weights, and exp is the exponential function with the natural constant as the base. It should be noted that the dynamic trade-off factor β is used to characterize the priority between energy-saving optimization and operating safety. When the operating safety margin M is small, the priority is biased towards operating safety; when the ambient temperature or the cooling water temperature deviates significantly from the reference value, the priority is adjusted appropriately to adapt to environmental changes.

[0143] In an implementation manner of the embodiment of the present invention, the reference ambient temperature T_ref is set to 25 degrees Celsius, the reference cooling water temperature T_cw_ref is set to 30 degrees Celsius, the correction weight k3 is set to 0.01, and k4 is set to 0.02.

[0144] In the embodiment of the present invention, the prediction deviation correction value E is expressed by the formula:

[0145] ; where E_N is the prediction deviation correction value of the compressor speed, and E_θ is the prediction deviation correction value of the guide vane opening. The prediction deviation correction value is obtained by calculating the deviation between the actual operating condition parameters and the historical operating data. The specific method is:

[0146] ;

[0147] ;

[0148] where N_act_i is the actual compressor speed at the i-th historical moment, N_pred_i is the predicted compressor speed at the i-th historical moment, θ_act_i is the actual guide vane opening at the i-th historical moment, θ_pred_i is the predicted guide vane opening at the i-th historical moment, and m is the number of historical moments.

[0149] In the embodiment of the present invention, the compressor speed target value N_target and the guide vane opening target value θ_target are expressed by the formula:

[0150] ;

[0151] ; where N_safe is the safe compressor speed corresponding to the operating safety boundary, θ_safe is the safe guide vane opening corresponding to the operating safety boundary, β is the dynamic trade-off factor, E_N is the prediction deviation correction value of the compressor speed, and E_θ is the prediction deviation correction value of the guide vane opening.

[0152] In an implementation of an embodiment of the present invention, the safe compressor speed N_safe and the safe guide vane opening θ_safe are obtained through the inverse mapping calculation of the operating safety boundary B, or can be obtained by querying a preset safe operation table, which is not limited herein.

[0153] It should be noted that the method for generating the control instruction is as follows:

[0154] The control instruction adopts a structured data format, including fields such as a unique identifier, a timestamp, a target execution device, an instruction type, and parameters. The parameter part includes the compressor speed setting value, the guide vane opening setting value, the change rate limit, and the execution priority. A checksum is appended at the end of the control instruction for data integrity verification.

[0155] When generating the control instruction, first calculate the difference between the target value and the current value to obtain the compressor speed adjustment amount and the guide vane opening adjustment amount respectively. To prevent the device from being impacted too much, the system will limit the single adjustment amplitude. When the calculated adjustment amount exceeds the maximum allowable value, it will be limited within the maximum allowable range. Based on the limited adjustment amount, calculate the new setting value. At the same time, determine the change rate limit based on the operating safety margin. The smaller the safety margin, the lower the change rate, to ensure system stability.

[0156] The change rate limit is calculated based on the basic change rate and the operating safety margin. An exponential decay function is used to make the change rate have a non-linear relationship with the operating safety margin. When the operating safety margin is small, the change rate approaches zero for slow adjustment; when the operating safety margin is large, the change rate approaches the basic change rate for allowing faster adjustment. The basic change rate is usually set to 10 Hertz per second (for compressor speed) and 1% per second (for guide vane opening).

[0157] The execution priority is determined according to the urgency of the operating state, and is divided into four levels: emergency, high, normal, and low. When the pressure ratio exceeds the critical value or the evaporation temperature is lower than the minimum safety value, it is set to the emergency priority; when the operating safety margin is lower than the threshold, it is set to the high priority; when the current energy efficiency ratio is lower than 90% of the target energy efficiency ratio, it is set to the normal priority; in other cases, it is set to the low priority. The priority determines the execution order and resource allocation of the control instruction in the control system.

[0158] The control instruction is transmitted to the control unit of the centrifugal chiller through an industrial fieldbus (such as Modbus, Profibus, or BACnet). A standard master-slave architecture is adopted, with the control system as the master station and the centrifugal chiller as the slave station. The transmission process includes steps such as control instruction encoding, data packet transmission, instruction decoding, checksum verification, return of confirmation information, and status update. The transmission protocol ensures the reliable transfer and correct execution of the control instruction.

[0159] Through the above control instruction generation method, the present invention can achieve precise adjustment of the compressor speed and guide vane opening of a centrifugal chiller, and realize energy-saving operation on the premise of ensuring operation safety.

[0160] Thus, the present invention is completed.

[0161] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments or perform equivalent replacements on some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

[0162] It should be noted that in this text, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or device. Without further limitations, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element.

[0163] In the description of the present invention, it should be understood that the terms "first", "second", etc. are only used for descriptive distinction and cannot be construed as indicating or implying relative importance.

[0164] In the description of the present invention, unless otherwise specified, the meaning of "a plurality" is two or more.

[0165] In the description of the present invention, the meaning of "several" is one or more, and the meaning of "a large number" is two or more.

[0166] In the description of this specification, the description with reference to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0167] For the formulas in this specification, the dimensional quantities are removed and only the numerical values are calculated. The formulas are obtained by collecting a large amount of data and performing software simulations to obtain a formula that is closest to the actual situation. The preset parameters and threshold values in the formulas are set by those skilled in the art according to the actual situation.

[0168] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the claims and their equivalents.

Claims

1. An energy-saving control method for a centrifugal chiller, characterized in that, Including: Obtain the operating state parameters and environmental state parameters of the centrifugal chiller during operation. The operating state parameters include compressor speed, guide vane opening, condenser pressure, and evaporator pressure. The environmental state parameters include environmental temperature and cooling water temperature. According to the operating state parameters and the environmental state parameters, obtain the predicted surge boundary value of the centrifugal chiller at each moment. Based on the change trend of the predicted surge boundary value and the fluctuation characteristics of the operating state parameters within the analysis period at each moment, obtain the operating safety margin of the centrifugal chiller at each moment. According to the operating safety margin and the dynamic distribution characteristics of the operating state parameters at each moment, obtain the energy-saving optimization control vector of the centrifugal chiller at each moment. Adjust the compressor speed and the guide vane opening based on the energy-saving optimization control vector to achieve the energy-saving operation of the centrifugal chiller.

2. The energy-saving control method of a centrifugal chiller according to claim 1, characterized in that, The obtaining of the predicted surge boundary value of the centrifugal chiller at each moment includes: According to the condenser pressure and the evaporator pressure, calculate the pressure ratio of the centrifugal chiller at each moment. According to the pressure ratio, the compressor speed, and the guide vane opening, construct the operating condition vector of the centrifugal chiller at each moment. Based on the operating condition vector and a preset surge boundary model, obtain the initial surge boundary value of the centrifugal chiller at the current moment. According to the environmental temperature and the cooling water temperature, calculate the environmental correction coefficient of the centrifugal chiller at each moment. Use the environmental correction coefficient to correct the initial surge boundary value to obtain the predicted surge boundary value of the centrifugal chiller at each moment.

3. The energy-saving control method of a centrifugal chiller according to claim 1, characterized in that, The obtaining of the operating safety margin of the centrifugal chiller at each moment includes: Statistical analysis of the change rate of the predicted surge boundary value within the analysis period at each moment, denoted as the surge boundary change trend value. Calculate the fluctuation variance of the operating state parameters within the analysis period at each moment, denoted as the operating state fluctuation value. According to the surge boundary change trend value and the operating state fluctuation value, construct the operating risk characteristic vector of the centrifugal chiller at each moment. Normalize the operating risk characteristic vector and combine it with a preset safety margin weight to obtain the operating safety margin of the centrifugal chiller at each moment.

4. An energy-saving control method for a centrifugal chiller according to claim 1, characterized in that, The obtaining of the energy-saving optimization control vector of the centrifugal chiller at each moment includes: According to the operating safety margin, determine the operating safety boundary of the centrifugal chiller at each moment. Based on the operating safety boundary and the dynamic distribution characteristics of the operating state parameters, construct the operating state characteristic matrix of the centrifugal chiller at each moment. Perform eigenvalue decomposition on the operating state characteristic matrix, extract the main eigencomponents, and obtain the energy-saving optimization direction of the centrifugal chiller at each moment. According to the energy-saving optimization direction and the operating safety margin, generate the energy-saving optimization control vector of the centrifugal chiller at each moment. The energy-saving optimization control vector includes the compressor speed adjustment amount and the guide vane opening adjustment amount.

5. The energy-saving control method of a centrifugal chiller according to claim 1, wherein Adjusting the compressor speed and the guide vane opening based on the energy-saving optimization control vector includes: Obtaining the actual operating condition parameters of the centrifugal chiller at the current moment; Calculating the target value of the compressor speed and the target value of the guide vane opening of the centrifugal chiller at the next moment according to the energy-saving optimization control vector; Generating a control instruction for the centrifugal chiller based on the actual operating condition parameters, the target value of the compressor speed, and the target value of the guide vane opening; Adjusting the compressor speed and the guide vane opening of the centrifugal chiller through the control instruction.

6. The energy-saving control method of a centrifugal chiller according to claim 2, characterized in that, The method for obtaining the preset surge boundary model includes: Collecting the historical operating data of the centrifugal chiller under different operating conditions, where the historical operating data includes the compressor speed, the guide vane opening, the condenser pressure, the evaporator pressure, and the surge occurrence state; Constructing an operating condition sample set of the centrifugal chiller according to the historical operating data; Training the operating condition sample set using a machine learning algorithm to obtain the preset surge boundary model, and the surge boundary model is used to characterize the mapping relationship between the operating condition vector and the surge boundary value.

7. The energy-saving control method of a centrifugal chiller according to claim 3, characterized in that The calculation method for the fluctuation variance of the operating state parameters includes: Statistical time series data of the compressor speed, the guide vane opening, the condenser pressure, and the evaporator pressure within the analysis period at each moment; Performing standardization processing on the time series data to obtain a standardized operating parameter sequence; Calculating the variance of the standardized operating parameter sequence to obtain the operating state fluctuation value.

8. The energy-saving control method of a centrifugal chiller according to claim 4, characterized in that, The construction method for the operating state feature matrix includes: Extracting the operating state distribution characteristics of the centrifugal chiller at each moment according to the dynamic distribution characteristics of the operating state parameters, where the operating state distribution characteristics include the joint distribution probability of the compressor speed and the guide vane opening; Determining the safe operating constraint conditions of the centrifugal chiller at each moment according to the operating safety boundary; Performing matrix representation on the operating state distribution characteristics and the safe operating constraint conditions to obtain the operating state feature matrix.

9. The energy-saving control method of a centrifugal chiller according to claim 6, characterized in that, The machine learning algorithm is a support vector machine algorithm or a deep neural network algorithm.

10. The energy-saving control method of a centrifugal chiller according to claim 5, characterized in that, The calculating the target value of the compressor speed and the target value of the guide vane opening of the centrifugal chiller at the next moment according to the energy-saving optimization control vector includes: Obtaining the actual operating condition parameters of the centrifugal chiller at the current moment and the energy-saving optimization control vector; Predicting the operating condition trend vector of the centrifugal chiller at the next moment according to the actual operating condition parameters, where the operating condition trend vector includes the compressor speed change trend and the guide vane opening change trend; Calculating the initial target value of the compressor speed and the initial target value of the guide vane opening of the centrifugal chiller at the next moment according to the operating condition trend vector and the energy-saving optimization control vector; Determining the dynamic trade-off factor of the centrifugal chiller at the next moment based on the operating safety margin and the environmental state parameters, and the dynamic trade-off factor is used to characterize the priority between energy-saving optimization and operating safety; Obtain the predicted deviation correction value of the centrifugal chiller at the next moment according to the deviation between the actual operating condition parameters and the historical operation data; Use the dynamic trade-off factor to perform weighted adjustment on the initial compressor speed target value and the initial guide vane opening target value, and combine the predicted deviation correction value for correction to obtain the compressor speed target value and the guide vane opening target value of the centrifugal chiller at the next moment.

Images