Intelligent control method and system for water-cooled central air conditioner based on deep reinforcement learning

Abstract

The application relates to the technical field of intelligent control, and is a water-cooled central air conditioner intelligent regulation and control method and system based on deep reinforcement learning, specifically comprising the following steps: performing heat exchange abnormality analysis and heat exchange environment influence factor analysis on a condenser based on a condenser heat exchange efficiency model and a cooling water microorganism breeding condition; fusing heat exchange abnormality analysis results and heat exchange environment influence factor analysis results to generate a comprehensive energy efficiency threat index of the water-cooled central air conditioner; and performing energy efficiency level sorting on each region of the water-cooled central air conditioner system based on the comprehensive energy efficiency threat index, and generating a differentiated regulation and control strategy.The application solves the problems that, in the prior art, the water-cooled central air conditioner has low energy efficiency, regulation and control lag and insufficient fault early warning capability during operation, and is difficult to adapt to complex and changeable actual operation environments and energy saving requirements.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, specifically to an intelligent control method and system for water-cooled central air conditioning based on deep reinforcement learning. Background Technology

[0002] Currently, water-cooled central air conditioning systems are widely used in large buildings. However, their operational energy efficiency is affected by a combination of complex factors, including condenser fouling, microbial growth in cooling water, dynamic load changes, water quality fluctuations, and the system's hydraulic-thermal coupling characteristics. Existing control methods are mostly based on single parameters or simple thresholds, failing to fully consider the collaborative analysis of multi-source data and systematic energy efficiency threat assessment. For example, in terms of fouling monitoring, traditional methods often only focus on one type of fouling, either crystalline or biological, ignoring the interaction mechanism between the two under different operating conditions. Furthermore, there is a lack of in-depth modeling of the hydraulic-thermal coupling relationship of the piping network, resulting in the inability to accurately identify low-efficiency heat exchange areas and their dynamic evolution trends, making it difficult to predict future changes in thermal state. These technical problems lead to low energy efficiency, delayed control, and insufficient fault warning capabilities in water-cooled central air conditioning systems, making them unsuitable for adapting to complex and ever-changing actual operating environments and energy-saving requirements. Summary of the Invention

[0003] The technical problem to be solved by the present invention is that water-cooled central air conditioning systems in the prior art have low energy efficiency, lagging control and insufficient fault early warning capabilities, and are difficult to adapt to complex and ever-changing actual operating environments and energy-saving requirements. The present invention proposes a water-cooled central air conditioning intelligent control method and system based on deep reinforcement learning.

[0004] To achieve the above objectives, the technical solution of the intelligent control method for water-cooled central air conditioning based on deep reinforcement learning of the present invention includes the following steps:

[0005] Step 1: Obtain data on cooling water inlet temperature, condenser tube wall fouling, chilled water flow distribution, and cooling tower fan power consumption.

[0006] Step 2: Construct a condenser heat exchange efficiency model, and conduct anomaly analysis of condenser heat exchange based on the condenser heat exchange efficiency model and the microbial growth in cooling water;

[0007] Step 3: Analyze the factors affecting the heat exchange environment by combining the historical operating load variation trend of the water-cooled central air conditioning system, the spatiotemporal distribution characteristics of cooling water hardness, and the zonal structural characteristics of the water-cooled central air conditioning system.

[0008] Step 4: Integrate the results of the heat exchange anomaly analysis with the results of the heat exchange environment influencing factor analysis to generate the comprehensive energy efficiency threat index of the water-cooled central air conditioning system;

[0009] The results of the heat exchange anomaly analysis include: current health threat index and structural energy efficiency threat index; the results of the heat exchange environmental influencing factor analysis include: trend energy efficiency threat index;

[0010] Step 5: Based on the comprehensive energy efficiency threat index, rank the energy efficiency levels of each area of ​​the water-cooled central air conditioning system and generate differentiated control strategies.

[0011] Specifically, step one includes:

[0012] A11: Collect condenser tube wall temperature distribution data, simultaneously acquire cooling water pipeline spatial geometric parameters, use a high-speed camera system to record cooling water outlet flow field morphology data, and monitor changes in microbial concentration in the cooling water;

[0013] A12: Based on thermal resistance sensing data, quantify the deposition degree of crystallized and particulate fouling, and calculate the reduction in equivalent heat exchange area caused by crystallization and particulate deposition. ;

[0014] A13: Simultaneously, based on data on changes in microbial concentration in cooling water, the adhesion effect of biofouling is quantified, and the reduction in equivalent heat exchange area caused by biofouling is calculated. ;

[0015] A14: Collect historical load scheduling data, hydraulic characteristic parameters of the condenser piping network, cooling tower performance curve data, and historical ambient temperature and humidity data of the water-cooled central air conditioning system; construct a cooling tower performance-system load coupling model; and output the real-time cooling capacity coefficient of the cooling tower. .

[0016] Specifically, step two includes:

[0017] B11: Extract spatial geometric parameters of cooling water pipelines, flow field morphology data of cooling water outlet, and microbial concentration variation data in cooling water, using the theoretical heat transfer area of ​​the water-cooled central air conditioning system under factory conditions. After deducting the reduced area for the two types of fouling, the comprehensive equivalent heat exchange area under the current operating conditions is obtained. ;

[0018] B12: Introducing the biofouling impact factor and a dynamic correction coefficient based on the load change rate to obtain the real-time dynamic heat exchange rate. ;

[0019] B13: Analyze the distribution characteristics of cooling water velocity field, evaluate the thermal resistance based on multi-parameter coupling, calculate the fouling thermal resistance coefficient per unit area of ​​each heat exchange area, identify abnormal thermal resistance by continuously monitoring the change of the fouling thermal resistance coefficient per unit area of ​​the condenser pipeline, and calculate the regional energy efficiency index of each pipeline area.

[0020] B14: Based on the theory of pipe network hydraulic analysis, a topological relationship analysis is performed on the dynamic fouling accumulation process of adjacent pipes in a water-cooled central air conditioning system to identify potential low-efficiency heat exchange areas in the system, and the channel energy efficiency index of these areas is evaluated simultaneously. ;

[0021] B15: Employing a multi-index fusion algorithm, this system integrates the regional energy efficiency index corresponding to independent heat exchange areas with the channel energy efficiency index of identified low-efficiency heat exchange channels to comprehensively calculate the structural energy efficiency threat index of the water-cooled central air conditioning system.

[0022] Specifically, step B14 includes the following steps:

[0023] B141: Establish a kinetic model for fouling accumulation based on water quality and operational parameters, evaluate the fouling deposition rate on pipe surfaces, and output the fouling layer thickening rate. ;

[0024] B142: Based on the hydraulic analysis theory of pipe network, construct the pipe correlation matrix, analyze the hydraulic correlation and thermal coupling degree between each pipe, and calculate the mutual influence coefficient between pipes;

[0025] B143: Set a threshold for the mutual influence coefficient, identify and mark significantly associated pipes whose mutual influence coefficient exceeds the threshold, merge these pipes into a potentially inefficient heat exchange unit, and perform subsequent evaluation steps;

[0026] When the mutual influence coefficients of all associated pipelines of a certain pipeline are lower than the threshold, the channel energy efficiency index of that pipeline is taken as the benchmark value of 1.

[0027] Specifically, step B14 also includes the following steps:

[0028] B144: Calculate the synergistic effect coefficient of potentially inefficient heat exchange units composed of associated piping groups. The equivalent heat transfer performance of the system is corrected, a thermal coupling model of the pipe group is established, and the corrected equivalent heat transfer coefficient is obtained.

[0029] B145: Combining fouling thickening rate data and heat transfer coefficient correction results, a weighted fusion algorithm is used to evaluate the channel energy efficiency index of low-efficiency heat exchange channels. Specifically, the ratio of fouling thickening rate to the critical value and the ratio of critical heat transfer coefficient to the actual heat transfer coefficient are calculated separately. Then, these two ratios are weighted and summed using preset weighting coefficients to obtain the channel energy efficiency index characterizing the channel's energy efficiency status. .

[0030] Specifically, step three includes:

[0031] C11: Collect historical load time series data and future expected load scheduling plans for water-cooled central air conditioning within one month, construct the spectral characteristic function of load fluctuation, extract the dominant frequency components to predict the load fluctuation cycle, and input the predicted load cycle characteristics into the equivalent temperature gradient calculation algorithm to predict the equivalent temperature gradient of each region.

[0032] C12: Combining the equivalent temperature gradient prediction results and the thermal conductivity data of the system piping network, a heat flow prediction model based on the principle of energy conservation is established to predict the evolution trend of the dynamic heat exchange performance of the water-cooled central air conditioning system and obtain the predicted heat flow value for future periods. .

[0033] Specifically, step three also includes:

[0034] C13: Extract heat flow prediction data and simultaneously collect the pipe wall-water temperature difference in each heat exchange zone. And cooling water pH value parameters, and calculate environmental correction factors through heat transfer efficiency analysis;

[0035] C14: Integrating predicted heat flow values ​​and environmental correction factors, construct an energy efficiency threat prediction model and generate a trend energy efficiency threat index. ;

[0036] ;

[0037] in, The design heat flow rate for a water-cooled central air conditioning system; The environmental correction factor is calculated in step C13.

[0038] Specifically, in step four, the strategy for generating the comprehensive energy efficiency threat index of the water-cooled central air conditioning system includes:

[0039] D11: Extract the real-time cooling capacity coefficient of the cooling tower calculated in step A14, the real-time dynamic heat exchange output in step B12, and the average fouling thermal resistance coefficient of all pipes in the system to assess the current health status of the water-cooled central air conditioning system and obtain the current health threat index.

[0040] D12: Extract the current health threat index, structural energy efficiency threat index, and trend energy efficiency threat index. Based on preset weighting coefficients, the three threat indices are weighted and fused to obtain the comprehensive energy efficiency threat index.

[0041] Specifically, step five includes: ranking and regulating the energy efficiency of the water-cooled central air conditioning system based on the comprehensive energy efficiency threat index, including: comparing the comprehensive energy efficiency threat index with the energy efficiency anomaly threshold. If the comprehensive energy efficiency threat index is greater than or equal to the energy efficiency anomaly threshold, it indicates that there is an energy efficiency risk in the system in the next operating cycle, and the system needs to be cleaned and controlled according to the ranking result of the comprehensive energy efficiency threat index. If the comprehensive energy efficiency threat index is less than the energy efficiency anomaly threshold, there is no energy efficiency risk in the system in the next operating cycle, and the system continues to be continuously monitored.

[0042] In addition, the intelligent control system for water-cooled central air conditioning based on deep reinforcement learning of this invention includes the following modules:

[0043] The system includes a data acquisition module, a heat exchange analysis module, a heat exchange environment analysis module, an energy efficiency threat assessment module, and a control triggering module.

[0044] The data acquisition module is used to acquire cooling water inlet temperature, condenser tube wall fouling data, cold water flow distribution data, and cooling tower fan power consumption data.

[0045] The heat exchange analysis module is used to construct a condenser heat exchange efficiency model and to perform heat exchange anomaly analysis of the condenser based on the condenser heat exchange efficiency model and the microbial growth in the cooling water.

[0046] The heat exchange environment analysis module is used to analyze the influencing factors of the heat exchange environment by combining the historical operating load change trend of the water-cooled central air conditioner, the spatiotemporal distribution characteristics of the cooling water hardness, and the zonal structure characteristics of the water-cooled central air conditioner.

[0047] The energy efficiency threat assessment module is used to integrate the results of heat exchange anomaly analysis with the results of heat exchange environmental influencing factor analysis to generate a comprehensive energy efficiency threat index for water-cooled central air conditioning.

[0048] The control triggering module ranks the energy efficiency levels of each area of ​​the water-cooled central air conditioning system based on the comprehensive energy efficiency threat index and generates differentiated control strategies.

[0049] Compared with the prior art, the technical effects of the present invention are as follows:

[0050] 1. This invention establishes a multi-mechanism coupling analysis model for condenser fouling, enabling simultaneous monitoring and quantitative assessment of crystalline fouling and biological fouling. This overcomes the limitations of traditional methods with a single perspective, significantly improves the accuracy and comprehensiveness of fouling status identification, provides a precise analytical basis for the maintenance of water-cooled central air conditioning, and effectively avoids the problem of decreased heat transfer efficiency caused by fouling accumulation.

[0051] 2. This invention adopts the hydraulic-thermal coupling analysis method of pipeline network. By constructing the pipeline correlation matrix and the synergistic effect evaluation model, it realizes the accurate identification and dynamic tracking of low-energy-efficiency areas of the system. It can promptly detect abnormal operating conditions and locate the root cause of the problem, which greatly improves the system's operational stability and energy efficiency management level.

[0052] 3. This invention establishes a load-environment coupled prediction model and a multi-parameter fusion energy efficiency threat assessment system, realizing full-cycle energy efficiency management from real-time monitoring to future prediction. This enables the system to have advanced early warning and adaptive control capabilities, which not only significantly improves system operating efficiency and reduces energy consumption, but also extends the service life of water-cooled central air conditioning. Attached Figure Description

[0053] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0054] Figure 1 This is a flowchart illustrating the intelligent control method for water-cooled central air conditioning based on deep reinforcement learning according to the present invention.

[0055] Figure 2 This is a schematic diagram of the intelligent control system for water-cooled central air conditioning based on deep reinforcement learning, as presented in this invention. Detailed Implementation

[0056] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0057] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0058] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0059] Example 1:

[0060] like Figure 1As shown in the figure, the intelligent control method for water-cooled central air conditioning based on deep reinforcement learning in this embodiment of the invention is as follows: Figure 1 As shown, the specific steps include the following:

[0061] Step 1: Obtain data on cooling water inlet temperature, condenser tube wall fouling, chilled water flow distribution, and cooling tower fan power consumption.

[0062] Step one includes:

[0063] A11: Collect condenser tube wall temperature distribution data, simultaneously acquire cooling water pipeline spatial geometric parameters, use a high-speed camera system to record cooling water outlet flow field morphology data, and monitor changes in microbial concentration in the cooling water;

[0064] A12: Based on thermal resistance sensing data, quantify the deposition degree of crystallized and particulate fouling, and calculate the reduction in equivalent heat exchange area caused by crystallization and particulate deposition. ;

[0065] For example, in this embodiment, a calculation strategy for the equivalent area reduction caused by crystallization and particulate fouling is provided, specifically as follows:

[0066] ;

[0067] in, This represents the total number of thermal resistance sensors. Let n be the basic heat exchange area of ​​the nth pipeline; The change in resistance of the nth pipeline reflects the increase in thermal resistance due to fouling on the pipe wall. The reference thermal resistance value; The average fouling thickness of the nth pipe; The baseline dirt thickness;

[0068] A13: Simultaneously, based on data on changes in microbial concentration in cooling water, the adhesion effect of biofouling is quantified, and the reduction in equivalent heat exchange area caused by biofouling is calculated. ;

[0069] For example, in this embodiment, a calculation strategy for the equivalent area reduction caused by biological fouling is provided, specifically as follows:

[0070] ;

[0071] in, G represents the theoretical heat transfer area of ​​the water-cooled central air conditioner under factory conditions; G represents the total number of microorganisms in the cooling water; g represents the microorganism species number. This represents the concentration change data for the g-th microorganism; This provides the average concentration variation data for type g microorganisms in all pipelines; The average equivalent radius of the g-th microorganism; This represents the coverage ratio of the g-th type of microorganism per unit volume;

[0072] It should be noted that load changes in water-cooled central air conditioning systems affect the condenser operating temperature and cooling water flow rate. When the water-cooled central air conditioning system is under high load, the condensing pressure and wall temperature increase, exacerbating the precipitation of fouling (i.e., crystallization). The formation rate of biofilm (i.e., high flow rate); at the same time, high flow rate affects the formation rate of biofilm (i.e., biofilm). It has a certain shearing and peeling effect.

[0073] When the water-cooled central air conditioning system is running at low load, the wall temperature decreases and the rate of crystallization of fouling decreases. However, the lower flow rate reduces the shear force, which is more conducive to the adhesion and growth of biofilm.

[0074] A14: Collect historical load scheduling data, hydraulic characteristic parameters of the condenser piping network, cooling tower performance curve data, and historical ambient temperature and humidity data of the water-cooled central air conditioning system; construct a cooling tower performance-system load coupling model; and output the real-time cooling capacity coefficient of the cooling tower. .

[0075] In this embodiment, the load scheduling history includes: hourly cooling load value of the air conditioning system, load change rate and duration; the cooling tower performance curve data includes: cooling capacity at different ambient wet-bulb temperatures and inlet water temperatures, and the relationship curve between cooling tower fan power consumption and approximation degree.

[0076] For example, in this embodiment, the specific strategy for calculating the real-time cooling capacity coefficient using the cooling tower performance-system load coupling model is as follows:

[0077] ;

[0078] in, The real-time cooling capacity coefficient represents the percentage of the cooling tower's actual cooling efficiency relative to its optimal capacity under current operating conditions.

[0079] This is the actual heat dissipation calculated in real time based on the cooling water flow rate and the temperature difference between the inlet and outlet.

[0080] The theoretical maximum heat dissipation is obtained from the cooling tower performance curve based on the current ambient wet-bulb temperature and water flow rate.

[0081] The approximation is the difference between the currently measured cooling water outlet temperature and the ambient wet-bulb temperature.

[0082] The optimal approximation of the cooling tower's highest efficiency under current operating conditions is determined by performance curves and experimental data;

[0083] The load change rate influence function is an empirical coefficient used to characterize the impact of cooling tower response lag on instantaneous efficiency when the system load changes drastically.

[0084] In this embodiment, the optimal approximation and other parameters are dynamically updated by integrating the performance curves provided by the cooling tower manufacturer, the system's historical operating data, and machine learning fitting.

[0085] It should be noted that the real-time cooling capacity coefficient This comprehensively reflects the coupled influence between the cooling tower's own performance, environmental conditions, and dynamic changes in system load: when When the value approaches 1, it indicates that the cooling tower is operating at its optimal point, with high cooling efficiency and minimal impact on system energy efficiency; while when... A value significantly less than 1 indicates low cooling tower efficiency, which may be caused by fan failure, packing scaling, uneven water distribution, or sudden load changes.

[0086] Step 2: Construct a condenser heat exchange efficiency model, and conduct anomaly analysis of condenser heat exchange based on the condenser heat exchange efficiency model and the microbial growth in cooling water;

[0087] Step two includes:

[0088] B11: Extract spatial geometric parameters of cooling water pipelines, flow field morphology data of cooling water outlet, and microbial concentration variation data in cooling water, using the theoretical heat transfer area of ​​the water-cooled central air conditioning system under factory conditions. After deducting the reduced area for the two types of fouling, the comprehensive equivalent heat exchange area under the current operating conditions is obtained. The details are as follows:

[0089] ;

[0090] B12: Introducing the biofouling impact factor and a dynamic correction coefficient based on the load change rate to obtain the real-time dynamic heat exchange rate. ;

[0091] For example, in this embodiment, the method for calculating the real-time dynamic heat exchange rate based on the basic formula of convective heat transfer is as follows:

[0092] ;

[0093] in, This refers to the cooling water flow rate; In this embodiment, the equivalent average temperature difference is used. Take the difference between the inlet water temperature and the pipe wall temperature;

[0094] As a factor influencing biofouling, an S-curve model is adopted in this embodiment, which conforms to the nonlinear characteristics of microbial growth;

[0095] The dynamic correction coefficient for load changes is represented by a hyperbolic tangent function, which reflects the saturation effect of load changes.

[0096] The concentration of microorganisms monitored by the biosensor closest to the pipeline; is the critical adhesion concentration; s, a, b are model fitting parameters obtained through machine learning using historical data. The real-time load change rate of the system;

[0097] It should be noted that, in this embodiment, the method for calculating real-time dynamic heat exchange is based on the fundamental formula of convective heat transfer. Approximate volumetric flow rate, multiplied by density Obtain the mass flow rate, then multiply by the specific heat capacity. By considering the temperature difference, the maximum heat power that the fluid can carry can be obtained;

[0098] B13: Analyze the distribution characteristics of cooling water velocity field, evaluate the thermal resistance based on multi-parameter coupling, calculate the fouling thermal resistance coefficient per unit area of ​​each heat exchange area, identify abnormal thermal resistance by continuously monitoring the change of the fouling thermal resistance coefficient per unit area of ​​the condenser pipeline, and calculate the regional energy efficiency index of each pipeline area.

[0099] For example, in this embodiment, the method for calculating the fouling thermal resistance coefficient per unit area is as follows:

[0100] ;

[0101] in, The fouling thermal resistance coefficient per unit area of ​​the nth pipeline segment; This represents the thermal resistance coefficient per unit area of ​​the clean pipeline. The thickness of the fouling in the nth pipe segment is obtained using an ultrasonic thickness gauge. The thermal conductivity of the dirt layer; It is a biofouling factor, which is related to the type and concentration of microorganisms; The coupling coefficient;

[0102] It should be noted that this model also considers crystallization fouling (and...). Related) and biofouling (with) The coupling effect of related factors more accurately reflects the actual fouling formation mechanism;

[0103] For example, in this embodiment, the regional energy efficiency index The calculation method is as follows:

[0104] ;

[0105] in, The average fouling thermal resistance coefficient of all pipes in the system is used to compare and characterize the health of the spatial dimension. The average fouling thermal resistance coefficient of all pipelines during the same historical period of the system is used to characterize the health status over time. This is a correction function for the temperature difference between the inlet and outlet water.

[0106] B14: Based on the theory of pipe network hydraulic analysis, a topological relationship analysis is performed on the dynamic fouling accumulation process of adjacent pipes in a water-cooled central air conditioning system to identify potential low-efficiency heat exchange areas in the system, and the channel energy efficiency index of these areas is evaluated simultaneously. ;

[0107] B15: Employing a multi-index fusion algorithm, this system integrates the regional energy efficiency index corresponding to independent heat exchange areas with the channel energy efficiency index of identified low-efficiency heat exchange channels to comprehensively calculate the structural energy efficiency threat index of the water-cooled central air conditioning system.

[0108] For example, in this embodiment, the structural energy efficiency threat index The calculation strategy is as follows:

[0109] ;

[0110] in, The entropy weights of each indicator; The coefficient represents the inter-regional coupling effect; M represents the total number of independent heat exchange regions. This represents the total number of low-efficiency heat exchange channels;

[0111] Step B14 includes the following steps:

[0112] B141: Establish a kinetic model for fouling accumulation based on water quality and operational parameters, evaluate the fouling deposition rate on pipe surfaces, and output the fouling layer thickening rate. ;

[0113] For example, in this embodiment, the dirt layer thickening rate The calculation method is as follows:

[0114] ;

[0115] in, This is the growth coefficient of crystallized fouling, which is related to water hardness; The Langerile saturation index reflects the scaling tendency of cooling water; The critical saturation index; The biofouling growth coefficient; The reaction order is determined through fitting experimental data. It should be noted that when... When the value is greater than 1, it indicates the existence of a critical point effect, meaning that dirt growth accelerates after the threshold is exceeded.

[0116] It should be noted that this model considers both chemical crystallization and biological adhesion as the two main fouling mechanisms, which is consistent with the actual operation of air conditioning systems.

[0117] B142: Based on the hydraulic analysis theory of pipe network, construct the pipe correlation matrix, analyze the hydraulic correlation and thermal coupling degree between each pipe, and calculate the mutual influence coefficient between pipes;

[0118] For example, in this embodiment, the mutual influence coefficient The calculation method is as follows:

[0119] ;

[0120] Where v is the parameter index participating in the coupling analysis, v=1,2,3 correspond to pipeline flow rate, inlet and outlet temperature difference and pipeline flow resistance, respectively; This represents the average value of the parameters in the coupling analysis when the label index is e; Weights of the parameters involved in the coupling analysis;

[0121] It should be noted that flow rate differences reflect the degree of hydraulic imbalance, which directly affects the heat exchange performance of each pipeline; temperature differences reflect the thermal imbalance, which affects the overall energy efficiency of the system; and flow resistance differences lead to uneven flow distribution, which exacerbates the deterioration of system performance.

[0122] B143: Set a threshold for the mutual influence coefficient, identify and mark significantly associated pipes whose mutual influence coefficients exceed the threshold, merge these pipes into a potentially inefficient heat exchange unit, and perform subsequent evaluation steps;

[0123] When the mutual influence coefficients of all associated pipelines of a certain pipeline are lower than the threshold, the channel energy efficiency index of that pipeline is taken as the benchmark value of 1.

[0124] B144: Calculate the synergistic effect coefficient of potentially inefficient heat exchange units composed of associated piping groups. The equivalent heat transfer performance of the system is corrected, a thermal coupling model of the pipe group is established, and the corrected equivalent heat transfer coefficient is obtained.

[0125] For example, in this embodiment, the synergy effect coefficient is calculated as follows:

[0126] ;

[0127] in, The coupling coefficient; This represents the number of associated pipelines in the associated pipeline group; This refers to the temperature difference deviation between related pipelines; The average temperature difference;

[0128] For example, in this embodiment, the method for correcting the equivalent heat transfer coefficient is as follows:

[0129] ;

[0130] in, These are the initial heat transfer coefficient and the corrected equivalent heat transfer coefficient, respectively.

[0131] It should be noted that, in this embodiment, the method for correcting the equivalent heat transfer coefficient is based on the fundamental principles of heat transfer, specifically as follows: ;

[0132] B145: Combining fouling thickening rate data and heat transfer coefficient correction results, a weighted fusion algorithm is used to evaluate the channel energy efficiency index of low-efficiency heat exchange channels. Specifically, the ratio of fouling thickening rate to the critical value and the ratio of critical heat transfer coefficient to the actual heat transfer coefficient are calculated separately. Then, these two ratios are weighted and summed using preset weighting coefficients to obtain the channel energy efficiency index characterizing the channel's energy efficiency status. .

[0133] Step 3: Analyze the factors affecting the heat exchange environment by combining the historical operating load variation trend of the water-cooled central air conditioning system, the spatiotemporal distribution characteristics of cooling water hardness, and the zonal structural characteristics of the water-cooled central air conditioning system.

[0134] Step three includes:

[0135] C11: Collect historical load time series data and future expected load scheduling plans for water-cooled central air conditioning within one month, construct the spectral characteristic function of load fluctuation, extract the dominant frequency components to predict the load fluctuation cycle, and input the predicted load cycle characteristics into the equivalent temperature gradient calculation algorithm to predict the equivalent temperature gradient of each region.

[0136] For example, in this embodiment, the historical load time series data includes: load rise and fall rate, high load duration, load change magnitude, etc.

[0137] For example, in this embodiment, the spectral characteristic function The calculation method is as follows:

[0138] ;

[0139] in, The system load value at time t; f is the frequency variable; T is the total length of the time series; The imaginary unit;

[0140] By converting the load time series to the frequency domain using Fourier transform, the main frequency components are extracted, and the periodic patterns of load fluctuations are identified, providing a basis for predicting future load change patterns.

[0141] For example, in this embodiment, the method for predicting and calculating the equivalent temperature gradient is as follows:

[0142] ;

[0143] in, The predicted equivalent temperature gradient for the j-th heat exchange region; To predict load changes; Let J be the thermal resistance of the j-th region;

[0144] C12: Combining the equivalent temperature gradient prediction results and the thermal conductivity data of the system piping network, a heat flow prediction model based on the principle of energy conservation is established to predict the evolution trend of the dynamic heat exchange performance of the water-cooled central air conditioning system and obtain the predicted heat flow value for future periods. ;

[0145] For example, in this embodiment, the method for calculating the predicted heat flow value is based on the fundamental equation of heat transfer: ;in, The overall heat transfer coefficient; For heat transfer area; The logarithmic mean temperature difference;

[0146] C13: Extract heat flow prediction data and simultaneously collect the pipe wall-water temperature difference in each heat exchange zone. And cooling water pH value parameters, and calculate environmental correction factors through heat transfer efficiency analysis;

[0147] For example, in this embodiment, the method for calculating the environmental correction factor is as follows:

[0148] Based on predicted heat flow data, combined with real-time collected parameters of pipe wall-water temperature difference and cooling water pH value in each heat exchange zone, the ratio of actual heat transfer to theoretical maximum heat transfer is calculated. Relative deviations of the pipe wall-water temperature difference from the optimal temperature difference and the relative deviations of the cooling water pH value from the optimal pH value are introduced as correction terms. Multiplying this ratio by the two correction terms yields the final environmental correction factor. ;

[0149] C14: Integrating predicted heat flow values ​​and environmental correction factors, construct an energy efficiency threat prediction model and generate a trend energy efficiency threat index. ;

[0150] ;

[0151] in, The design heat flow rate for a water-cooled central air conditioning system; The environmental correction factor is calculated in step C13.

[0152] Step 4: Integrate the results of the heat exchange anomaly analysis with the results of the heat exchange environment influencing factor analysis to generate the comprehensive energy efficiency threat index of the water-cooled central air conditioning system;

[0153] The results of the heat exchange anomaly analysis include: current health threat index and structural energy efficiency threat index; the results of the heat exchange environmental influencing factor analysis include: trend energy efficiency threat index;

[0154] In step four, the strategy for generating the comprehensive energy efficiency threat index of the water-cooled central air conditioning system includes:

[0155] D11: Extract the real-time cooling capacity coefficient of the cooling tower calculated in step A14, the real-time dynamic heat exchange output in step B12, and the average fouling thermal resistance coefficient of all pipes in the system to assess the current health status of the water-cooled central air conditioning system and obtain the current health threat index.

[0156] For example, in this embodiment, the current health threat index The acquisition strategy is as follows: ;

[0157] in, The real-time cooling capacity coefficient of the cooling tower is calculated in step A14; This refers to the real-time dynamic heat exchange output in step B12. This represents the average fouling thermal resistance coefficient of all current piping in the system. This is the critical thermal resistance of the pipeline;

[0158] D12: Extract the current health threat index, structural energy efficiency threat index, and trend energy efficiency threat index. Based on preset weighting coefficients, the three threat indices are weighted and fused to obtain the comprehensive energy efficiency threat index.

[0159] Step 5: Based on the comprehensive energy efficiency threat index, rank the energy efficiency levels of each area of ​​the water-cooled central air conditioning system and generate differentiated control strategies.

[0160] Step five includes: ranking and regulating the energy efficiency of the water-cooled central air conditioning system based on the comprehensive energy efficiency threat index. This includes comparing the comprehensive energy efficiency threat index with the energy efficiency anomaly threshold. If the comprehensive energy efficiency threat index is greater than or equal to the energy efficiency anomaly threshold, it indicates that the system has an energy efficiency risk in the next operating cycle, and the system needs to be cleaned and controlled according to the ranking result of the comprehensive energy efficiency threat index. If the comprehensive energy efficiency threat index is less than the energy efficiency anomaly threshold, the system does not have an energy efficiency risk in the next operating cycle, and the system continues to be continuously monitored.

[0161] Example 2:

[0162] like Figure 2 As shown in the figure, the intelligent control system for water-cooled central air conditioning based on deep reinforcement learning in this embodiment of the invention, such as Figure 2 As shown, it includes the following modules:

[0163] The system includes a data acquisition module, a heat exchange analysis module, a heat exchange environment analysis module, an energy efficiency threat assessment module, and a control triggering module.

[0164] The data acquisition module is used to acquire cooling water inlet temperature, condenser tube wall fouling data, cold water flow distribution data, and cooling tower fan power consumption data.

[0165] The heat exchange analysis module is used to construct a condenser heat exchange efficiency model and to perform heat exchange anomaly analysis of the condenser based on the condenser heat exchange efficiency model and the microbial growth in the cooling water.

[0166] The heat exchange environment analysis module is used to analyze the influencing factors of the heat exchange environment by combining the historical operating load change trend of the water-cooled central air conditioner, the spatiotemporal distribution characteristics of cooling water hardness, and the zonal structure characteristics of the water-cooled central air conditioner.

[0167] The energy efficiency threat assessment module is used to integrate the results of heat exchange anomaly analysis with the results of heat exchange environmental influencing factor analysis to generate a comprehensive energy efficiency threat index for water-cooled central air conditioning.

[0168] The control triggering module ranks the energy efficiency levels of each area of ​​the water-cooled central air conditioning system based on the comprehensive energy efficiency threat index and generates differentiated control strategies.

[0169] Example 3:

[0170] This embodiment provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0171] The processor executes the aforementioned intelligent control method for water-cooled central air conditioning based on deep reinforcement learning by calling the computer program stored in memory.

[0172] The electronic device can vary considerably depending on its configuration or performance. It may include one or more Central Processing Units (CPUs) and one or more memories, wherein the memory stores at least one computer program, which is loaded and executed by the processor to implement the deep reinforcement learning-based intelligent control method for water-cooled central air conditioning provided in the above-described embodiment. The electronic device may also include other components for implementing its functions; for example, it may have wired or wireless network interfaces and input / output interfaces for data input and output. Details will not be elaborated upon in this embodiment.

[0173] Example 4:

[0174] This embodiment proposes a computer-readable storage medium on which an erasable and rewritable computer program is stored.

[0175] When the computer program runs on the computer device, it enables the computer device to execute the aforementioned intelligent control method for water-cooled central air conditioning based on deep reinforcement learning.

[0176] For example, computer-readable storage media can be read-only memory (ROM), random access memory (RAM), compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage devices.

[0177] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0178] It should be understood that determining B based on A does not mean determining B solely based on A; it also means determining B based on A and / or other information.

[0179] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the flow or function according to the embodiments of the present invention is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired network and / or wireless network. A computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives (SSDs).

[0180] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0181] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0182] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only one method, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0183] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0184] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0185] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0186] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent control of water-cooled central air conditioning based on deep reinforcement learning, characterized in that, The method includes: Step 1: Obtain data on cooling water inlet temperature, condenser tube wall fouling, chilled water flow distribution, and cooling tower fan power consumption. Step 2: Construct a condenser heat exchange efficiency model, and conduct anomaly analysis of condenser heat exchange based on the condenser heat exchange efficiency model and the microbial growth in cooling water; Step two includes: B11: Extract spatial geometric parameters of cooling water pipelines, flow field morphology data of cooling water outlet, and microbial concentration variation data in cooling water, using the theoretical heat transfer area of ​​the water-cooled central air conditioning system under factory conditions. After deducting the reduced area for the two types of fouling, the comprehensive equivalent heat exchange area under the current operating conditions is obtained. ; B12: Introducing the biofouling impact factor and a dynamic correction coefficient based on the load change rate to obtain the real-time dynamic heat exchange rate. ; B13: Analyze the distribution characteristics of cooling water velocity field, evaluate the thermal resistance based on multi-parameter coupling, calculate the fouling thermal resistance coefficient per unit area of ​​each heat exchange area, identify abnormal thermal resistance by continuously monitoring the change of the fouling thermal resistance coefficient per unit area of ​​the condenser pipeline, and calculate the regional energy efficiency index of each pipeline area. B14: Based on the theory of pipe network hydraulic analysis, a topological relationship analysis is performed on the dynamic fouling accumulation process of adjacent pipes in a water-cooled central air conditioning system to identify potential low-efficiency heat exchange areas in the system, and the channel energy efficiency index of these areas is evaluated simultaneously. ; Step B14 includes the following steps: B141: Establish a kinetic model for fouling accumulation based on water quality and operational parameters, evaluate the fouling deposition rate on pipe surfaces, and output the fouling layer thickening rate. ; B142: Based on the hydraulic analysis theory of pipe network, construct the pipe correlation matrix, analyze the hydraulic correlation and thermal coupling degree between each pipe, and calculate the mutual influence coefficient between pipes; B143: Set a threshold for the mutual influence coefficient, identify and mark significantly associated pipes whose mutual influence coefficient exceeds the threshold, merge these pipes into a potentially inefficient heat exchange unit, and perform subsequent evaluation steps; When the mutual influence coefficients of all associated pipelines of a certain pipeline are lower than the threshold, the channel energy efficiency index of that pipeline is taken as the benchmark value of 1. B144: Calculate the synergistic effect coefficient of potentially inefficient heat exchange units composed of associated piping groups. The equivalent heat transfer performance of the system is corrected, a thermal coupling model of the pipe group is established, and the corrected equivalent heat transfer coefficient is obtained. B145: Combining fouling thickening rate data and heat transfer coefficient correction results, a weighted fusion algorithm is used to evaluate the channel energy efficiency index of low-efficiency heat exchange channels. Specifically, the ratio of fouling thickening rate to the critical value and the ratio of critical heat transfer coefficient to the actual heat transfer coefficient are calculated separately. Then, these two ratios are weighted and summed using preset weighting coefficients to obtain the channel energy efficiency index characterizing the channel's energy efficiency status. ; B15: Employing a multi-index fusion algorithm, this algorithm integrates the regional energy efficiency index corresponding to independent heat exchange areas and the channel energy efficiency index of identified low-efficiency heat exchange channels to comprehensively calculate the structural energy efficiency threat index of the water-cooled central air conditioning system. Step 3: Analyze the factors affecting the heat exchange environment by combining the historical operating load variation trend of the water-cooled central air conditioning system, the spatiotemporal distribution characteristics of cooling water hardness, and the zonal structural characteristics of the water-cooled central air conditioning system. Step 4: Integrate the results of the heat exchange anomaly analysis with the results of the heat exchange environment influencing factor analysis to generate the comprehensive energy efficiency threat index of the water-cooled central air conditioning system; The results of the heat exchange anomaly analysis include: current health threat index and structural energy efficiency threat index; the results of the heat exchange environmental influencing factor analysis include: trend energy efficiency threat index; Step 5: Based on the comprehensive energy efficiency threat index, rank the energy efficiency levels of each area of ​​the water-cooled central air conditioning system and generate differentiated control strategies.

2. The intelligent control method for water-cooled central air conditioning based on deep reinforcement learning according to claim 1, characterized in that, Step one includes: A11: Collect condenser tube wall temperature distribution data, simultaneously acquire cooling water pipeline spatial geometric parameters, use a high-speed camera system to record cooling water outlet flow field morphology data, and monitor changes in microbial concentration in the cooling water; A12: Based on thermal resistance sensing data, quantify the deposition degree of crystallized and particulate fouling, and calculate the reduction in equivalent heat exchange area caused by crystallization and particulate deposition. ; A13: Simultaneously, based on data on changes in microbial concentration in cooling water, the adhesion effect of biofouling is quantified, and the reduction in equivalent heat exchange area caused by biofouling is calculated. ; A14: Collect historical load scheduling data, hydraulic characteristic parameters of the condenser piping network, cooling tower performance curve data, and historical ambient temperature and humidity data of the water-cooled central air conditioning system; construct a cooling tower performance-system load coupling model; and output the real-time cooling capacity coefficient of the cooling tower. .

3. The intelligent control method for water-cooled central air conditioning based on deep reinforcement learning according to claim 2, characterized in that, Step three includes: C11: Collect historical load time series data and future expected load scheduling plans for water-cooled central air conditioning within one month, construct the spectral characteristic function of load fluctuation, extract the dominant frequency components to predict the load fluctuation cycle, and input the predicted load cycle characteristics into the equivalent temperature gradient calculation algorithm to predict the equivalent temperature gradient of each region. C12: Combining the equivalent temperature gradient prediction results and the thermal conductivity data of the system piping network, a heat flow prediction model based on the principle of energy conservation is established to predict the evolution trend of the dynamic heat exchange performance of the water-cooled central air conditioning system and obtain the predicted heat flow value for future periods. .

4. The intelligent control method for water-cooled central air conditioning based on deep reinforcement learning according to claim 3, characterized in that, Step three also includes: C13: Extract heat flow prediction data and simultaneously collect the pipe wall-water temperature difference in each heat exchange zone. And cooling water pH value parameters, and calculate environmental correction factors through heat transfer efficiency analysis; C14: Integrating predicted heat flow values ​​and environmental correction factors, construct an energy efficiency threat prediction model and generate a trend energy efficiency threat index. ; ； in, The design heat flow rate for a water-cooled central air conditioning system; The environmental correction factor is calculated in step C13.

5. The intelligent control method for water-cooled central air conditioning based on deep reinforcement learning according to claim 4, characterized in that, In step four, the strategy for generating the comprehensive energy efficiency threat index of the water-cooled central air conditioning system includes: D11: Extract the real-time cooling capacity coefficient of the cooling tower calculated in step A14, the real-time dynamic heat exchange output in step B12, and the average fouling thermal resistance coefficient of all pipes in the system to assess the current health status of the water-cooled central air conditioning system and obtain the current health threat index. D12: Extract the current health threat index, structural energy efficiency threat index, and trend energy efficiency threat index. Based on preset weighting coefficients, the three threat indices are weighted and fused to obtain the comprehensive energy efficiency threat index.

6. The intelligent control method for water-cooled central air conditioning based on deep reinforcement learning according to claim 5, characterized in that, Step five includes: ranking and regulating the energy efficiency of the water-cooled central air conditioning system based on the comprehensive energy efficiency threat index. This includes comparing the comprehensive energy efficiency threat index with the energy efficiency anomaly threshold. If the comprehensive energy efficiency threat index is greater than or equal to the energy efficiency anomaly threshold, it indicates that the system has an energy efficiency risk in the next operating cycle, and the system needs to be cleaned and controlled according to the ranking result of the comprehensive energy efficiency threat index. If the comprehensive energy efficiency threat index is less than the energy efficiency anomaly threshold, the system does not have an energy efficiency risk in the next operating cycle, and the system continues to be continuously monitored.

7. A water-cooled central air conditioning intelligent control system based on deep reinforcement learning, used to implement the water-cooled central air conditioning intelligent control method based on deep reinforcement learning as described in any one of claims 1-6, characterized in that, The system includes: The system includes a data acquisition module, a heat exchange analysis module, a heat exchange environment analysis module, an energy efficiency threat assessment module, and a control triggering module. The data acquisition module is used to acquire cooling water inlet temperature, condenser tube wall fouling data, cold water flow distribution data, and cooling tower fan power consumption data. The heat exchange analysis module is used to construct a condenser heat exchange efficiency model and to perform heat exchange anomaly analysis of the condenser based on the condenser heat exchange efficiency model and the microbial growth in the cooling water. The heat exchange environment analysis module is used to analyze the influencing factors of the heat exchange environment by combining the historical operating load change trend of the water-cooled central air conditioner, the spatiotemporal distribution characteristics of cooling water hardness, and the zonal structure characteristics of the water-cooled central air conditioner. The energy efficiency threat assessment module is used to integrate the results of heat exchange anomaly analysis with the results of heat exchange environmental influencing factor analysis to generate a comprehensive energy efficiency threat index for water-cooled central air conditioning. The control triggering module ranks the energy efficiency levels of each area of ​​the water-cooled central air conditioning system based on the comprehensive energy efficiency threat index and generates differentiated control strategies.

Images