Fault detection and diagnosis are essential to the air conditioning system of the data center for elevating reliability and reducing energy consumption.This study proposed a convolutional neural network(CNN)based data...Fault detection and diagnosis are essential to the air conditioning system of the data center for elevating reliability and reducing energy consumption.This study proposed a convolutional neural network(CNN)based data-driven fault detection and diagnosis model considering temporal dependency for composite air conditioning system that is capable of cooling the high heat flux in data centers.The input of fault detection and diagnosis model was an unsteady dataset generated by the experimentally validated transient mathematical model.The dataset concerned three typical faults,including refrigerant leakage,evaporator fan breakdown,and condenser fouling.Then,the CNN model was trained to construct a map between the input and system operating conditions.Further,the performance of the CNN model was validated by comparing it with the support vector machine and the neural network.Finally,the score-weighted class mapping activation method was utilized to interpret model diagnosis mechanisms and to identify key input features in various operating modes.The results demonstrated in the pump-driven heat pipe mode,the accuracy of the CNN model was 99.14%,increasing by around 8.5%compared with the other two methods.In the vapor compression mode,the accuracy of the CNN model achieved 99.9%and declined the miss rate of refrigerant leakage by at least 61%comparatively.The score-weighted class mapping activation results indicated the ambient temperature and the actuator-related parameters,such as compressor frequency in vapor compression mode and condenser fan frequency in pump-driven heat pipe mode,were essential features in system fault detection and diagnosis.展开更多
A three-dimensional finite element dynamic simulation platform of the ground source heat pump system(GSHPS)is established.According to the outlet temperature of ground heat exchangers(GHEs)required by the code in summ...A three-dimensional finite element dynamic simulation platform of the ground source heat pump system(GSHPS)is established.According to the outlet temperature of ground heat exchangers(GHEs)required by the code in summer and winter,the calculated minimum buried depth of GHEs meeting the requirements is 60 m,when the number of borehole is 9.By using the established platform,the annual operation performance and cost of the GSHPS under different buried pipe depths are studied.The results show that the deeper the buried depth of GHEs is,the better the heat exchange effect of GHEs is.Compared with the GHEs with 60 m buried depth,when the buried depth of GHEs is 65 m,70 m,75 m and 80 m,the average coefficient of performance(COP)of the unit increases by 4.1%,6.3%,7.7%and 8.2%in cooling period and 1.0%,1.6%,1.8%and 1.9%in heating period,respectively.Considering the performance and initial investment of the GHSPS comprehensively,the optimal buried depth of GHEs is 60 m.However,considering the performance the system and the total cost of the system running for 20 years comprehensively,the optimal buried depth of GHEs is 70 m.展开更多
To solve the problems of single heat source heat pump systems in severe cold regions,a dual-source hybrid heat pump unit(DSHHPU)is proposed.The mathematical models of the DSHHPU when charging R134a or its alternative ...To solve the problems of single heat source heat pump systems in severe cold regions,a dual-source hybrid heat pump unit(DSHHPU)is proposed.The mathematical models of the DSHHPU when charging R134a or its alternative refrigerants R32,R290 and R600a were established respectively,and the performance was simulated and analysed.The results showed that the four refrigerants have different performance characteristics in different aspects.In heat pipe mode,the heating capacity and evaporating pressure of R32 are 36.94%and 59.94%higher than those of R134a.The heating capacity and evaporating pressure of R290 are 5.73%and 22.99%lower than those of R134a.The heating capacity and evaporating pressure of R600a are 43.29%and 68.08%lower than those of R134a.In vapour compression heating mode,the discharge temperature of R32,R290 and R600a are 184.88,72.98 and 66.44%of that of R134a.The coefficient of performance(COP)of R32,R290 and R600a are 72.65,111.59 and 117.94%of that of R134a.Finally,the effects of radiation intensity and ambient temperature on key performance parameters of the different refrigerants were analysed.The research results provide a reference for research on refrigerant replacements for multi-heat source composite heat pump systems.展开更多
基金the support from the National Natural Science Foundation of China(Grant number 52176180)the support from“the open competition mechanism to select the best candidates”key technology project of Liaoning(Grant 2022JH1/10800008).
文摘Fault detection and diagnosis are essential to the air conditioning system of the data center for elevating reliability and reducing energy consumption.This study proposed a convolutional neural network(CNN)based data-driven fault detection and diagnosis model considering temporal dependency for composite air conditioning system that is capable of cooling the high heat flux in data centers.The input of fault detection and diagnosis model was an unsteady dataset generated by the experimentally validated transient mathematical model.The dataset concerned three typical faults,including refrigerant leakage,evaporator fan breakdown,and condenser fouling.Then,the CNN model was trained to construct a map between the input and system operating conditions.Further,the performance of the CNN model was validated by comparing it with the support vector machine and the neural network.Finally,the score-weighted class mapping activation method was utilized to interpret model diagnosis mechanisms and to identify key input features in various operating modes.The results demonstrated in the pump-driven heat pipe mode,the accuracy of the CNN model was 99.14%,increasing by around 8.5%compared with the other two methods.In the vapor compression mode,the accuracy of the CNN model achieved 99.9%and declined the miss rate of refrigerant leakage by at least 61%comparatively.The score-weighted class mapping activation results indicated the ambient temperature and the actuator-related parameters,such as compressor frequency in vapor compression mode and condenser fan frequency in pump-driven heat pipe mode,were essential features in system fault detection and diagnosis.
基金The authors gratefully acknowledge the support from the Natural Science Foundation of China(grant No.51778115)the Fundamental Research Funds for the Central Universities(grant No.N182502043).
文摘A three-dimensional finite element dynamic simulation platform of the ground source heat pump system(GSHPS)is established.According to the outlet temperature of ground heat exchangers(GHEs)required by the code in summer and winter,the calculated minimum buried depth of GHEs meeting the requirements is 60 m,when the number of borehole is 9.By using the established platform,the annual operation performance and cost of the GSHPS under different buried pipe depths are studied.The results show that the deeper the buried depth of GHEs is,the better the heat exchange effect of GHEs is.Compared with the GHEs with 60 m buried depth,when the buried depth of GHEs is 65 m,70 m,75 m and 80 m,the average coefficient of performance(COP)of the unit increases by 4.1%,6.3%,7.7%and 8.2%in cooling period and 1.0%,1.6%,1.8%and 1.9%in heating period,respectively.Considering the performance and initial investment of the GHSPS comprehensively,the optimal buried depth of GHEs is 60 m.However,considering the performance the system and the total cost of the system running for 20 years comprehensively,the optimal buried depth of GHEs is 70 m.
基金The authors gratefully acknowledge the support from the Natural Science Foundation of China(grant No.51778115)the Fundamen-tal Research Funds for the Central Universities(grant No.N182502043).
文摘To solve the problems of single heat source heat pump systems in severe cold regions,a dual-source hybrid heat pump unit(DSHHPU)is proposed.The mathematical models of the DSHHPU when charging R134a or its alternative refrigerants R32,R290 and R600a were established respectively,and the performance was simulated and analysed.The results showed that the four refrigerants have different performance characteristics in different aspects.In heat pipe mode,the heating capacity and evaporating pressure of R32 are 36.94%and 59.94%higher than those of R134a.The heating capacity and evaporating pressure of R290 are 5.73%and 22.99%lower than those of R134a.The heating capacity and evaporating pressure of R600a are 43.29%and 68.08%lower than those of R134a.In vapour compression heating mode,the discharge temperature of R32,R290 and R600a are 184.88,72.98 and 66.44%of that of R134a.The coefficient of performance(COP)of R32,R290 and R600a are 72.65,111.59 and 117.94%of that of R134a.Finally,the effects of radiation intensity and ambient temperature on key performance parameters of the different refrigerants were analysed.The research results provide a reference for research on refrigerant replacements for multi-heat source composite heat pump systems.