CN109781345B - Refrigerating system refrigerant leakage detection method based on data driving and model - Google Patents
Refrigerating system refrigerant leakage detection method based on data driving and model Download PDFInfo
- Publication number
- CN109781345B CN109781345B CN201910053039.2A CN201910053039A CN109781345B CN 109781345 B CN109781345 B CN 109781345B CN 201910053039 A CN201910053039 A CN 201910053039A CN 109781345 B CN109781345 B CN 109781345B
- Authority
- CN
- China
- Prior art keywords
- refrigerant
- mass
- quality
- flash
- compressor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Abstract
The invention discloses a refrigerating system refrigerant leakage detection method based on data driving and a model, which relates to the field of refrigerant leakage detection and comprises the following steps: (1) setting a temperature measuring point in a refrigeration system; (2) respectively establishing a refrigerant quality theoretical model of a heat exchanger partition, a compressor and a connecting pipeline; (3) when the refrigerating system contains a flash evaporator, establishing a local weighted linear regression model for calculating the refrigerant quality of the flash evaporator; (4) inputting the obtained temperature measuring point data into a heat exchanger partition, a compressor and connecting pipeline refrigerant quality theoretical model and a local weighted linear regression model for flash tank refrigerant quality calculation, and accumulating to obtain the refrigerant quality of the system; (5) comparing the refrigerant mass of a refrigeration system with an initial charge of the refrigeration system. The method can save the data amount required by leakage detection, accurately judge whether the refrigeration system leaks and calculate the leakage amount.
Description
Technical Field
The invention relates to the field of refrigerant leakage detection, in particular to a refrigeration system refrigerant leakage detection method based on data driving and a model.
Background
As the product life of a refrigeration system increases, refrigerant within the system may leak. The refrigerant leaks by 10 percent, and the product performance can be reduced by 27 percent at most. The performance of the refrigeration system is reduced due to refrigerant leakage, so that the working temperature of the cooled environment or equipment is increased continuously, and finally, the power consumption of the cooled equipment is increased, and serious consequences such as system shutdown can be caused. Therefore, in order to avoid system faults caused by refrigerant leakage, a precise refrigerant leakage detection method of the refrigeration system is established, and the method has very important significance for fault monitoring and fault isolation of the refrigeration system. The existing refrigerant leakage detection methods of the refrigeration system are classified into a direct measurement method and a prediction method. The direct measurement method mainly comprises the steps of soapy water leakage detection, bubble leakage detection, halogen detector leakage detection and helium mass spectrometer leakage detection. The direct measurement method has the defects of manual intervention, influence on system performance due to measurement operation and the like, so the application range of the direct measurement method is limited. The prediction method of the refrigerant leakage is mainly a data-driven method, and comprises a virtual sensor, a neural network, a support vector machine and the like. The method comprises the steps of acquiring a large amount of data, extracting characteristic values to learn a model, and inputting the data into the model to perform classification judgment of refrigerant leakage in the whole system. The data-driven system detection method only needs larger data volume and better hyper-parameter selection to learn out the model with lower error. For a refrigeration system with complex operation environment, less actual fault operation data and high data acquisition cost, the size of the available data set is limited by the characteristic. The prediction method using only data driving is limited in its applicability. If the refrigerating system contains a flash tank or a liquid storage tank, such as an intermediate air supply type heat pump system, a large commercial refrigerating system and the like, the refrigerant in the components exists in two phases, the liquid level of the refrigerant is difficult to measure, and an accurate theoretical model for calculating the mass of the refrigerant in the components cannot be established. A leak detection method based only on theoretical models is therefore not suitable for this type of refrigeration system.
The invention aims to solve the problem of how to establish an accurate refrigerant leakage detection method for a refrigeration system by using a small amount of operation data.
Accordingly, those skilled in the art have endeavored to develop a data-driven and model-based refrigerant leak detection method for a refrigeration system. The method can save the data amount required by leakage detection, accurately judge whether the refrigeration system leaks and calculate the leakage amount.
Disclosure of Invention
In view of the above-mentioned drawbacks of the prior art, the technical problem to be solved by the present invention is how to establish an accurate refrigerant leak detection method for a refrigeration system using a small amount of operation data.
In order to achieve the above object, the present invention provides a data-driven and model-based refrigerant leakage detection method for a refrigeration system, which is characterized by comprising the following steps:
(1) setting a temperature measuring point in a refrigeration system, and reading experimental working condition data;
(2) establishing a heat exchanger partition refrigerant quality theoretical model based on a temperature signal, a compressor refrigerant quality theoretical model based on theory and an explicit expression, and a connecting pipeline refrigerant quality theoretical model based on theory and the explicit expression;
(3) when the refrigerating system comprises a flash tank, establishing a local weighted linear regression model based on the calculation of the refrigerant quality of the data-driven flash tank;
(4) after the data of the temperature measuring points are obtained, the data are input into the heat exchanger partition refrigerant quality theoretical model based on the temperature signals, the compressor refrigerant quality theoretical model based on the theory and the explicit expression, the connecting pipeline refrigerant quality theoretical model based on the theory and the explicit expression and the local weighted linear regression model calculated based on the data-driven flash evaporator refrigerant quality, and the refrigerant quality of the refrigerating system is obtained through accumulation;
(5) comparing the refrigerant quality of the refrigeration system with an initial charge of the refrigeration system.
Further, the temperature measuring points in the step (1) are arranged along the inlet, the outlet and the heat exchanger zones of the compressor, the connecting pipeline and the flash tank.
Further, the refrigerant mass of the evaporator includes a state point of an inlet and a state point of an outlet of the evaporator, and a state point at which a two-phase zone in the evaporator ends, which is represented by the following formula:
meavp＝ft(Tx...Tx+N,Ty)
＝ft(X5,X6,X7)
the refrigerant mass of the condenser includes a state point of an inlet and a state point of an outlet of the condenser and a state point of a start and a state point of an end of a two-phase region in the condenser, and is represented by the following formula:
mcond＝ft(Ti,Tj...Tj+M,Tk)
＝ft(X1,X2,X3,X4)
the parameters of the state points are represented by the following formula according to the temperature measuring points:
Xi＝(Ti,hi,Xi,ρi),i＝1,2,...,7
wherein, X is1Is the state point of the inlet of the condenser, the X2The state point, the X, for the initiation of the two-phase zone within the condenser3The point of state, the X, of the termination of the two-phase zone within the condenser4Is the state point of the outlet of the condenser, the X5For the state point of the inlet of the evaporator, the X6The state point, the X, of the termination of the two-phase region of the evaporator7Is the state point of the outlet of the evaporator, the h is the enthalpy of the state point, the X is the dryness of the state point, and the ρ is the enthalpy of the state point;
said X1And said X2In the specification X6And said X7Between them is a superheat zone, X3And said X4The refrigerant quality is represented by the following formula:
wherein, said mshIs the refrigerant mass of the superheat region, the mscThe refrigerant mass of the supercooling region, the pshIs the refrigerant density of the superheat region, the pscIs the refrigerant density of the subcooling zone, V is the volume of the superheat zone or the subcooling zone;
for the refrigerant mass in the two-phase region, a Hughmark model is adopted for calculation, and the volume of the superheat region or the supercooling region is calculated by the product of the length and the cross-sectional area of the superheat region or the supercooling region and is represented by the following formula:
Vj＝AjLj
wherein r is the refrigerant; a is air; the in is an inlet; the out is the outlet.
Further, the calculation of the theoretical model of compressor refrigerant quality based on theoretical and explicit expressions in the step (2) is as follows:
the refrigerant in the compressor includes the refrigerant in the refrigeration oil, the refrigerant in the accumulator, and the refrigerant in the shell cavity, represented by the formula:
mcomp＝mr_oil+mr_shell+mr_accum
＝ft(Tp,Tq)
wherein, said mcompIs the total mass of the refrigerant in the compressor in g; m isr_ollIs the refrigerant mass in the refrigeration oil in g; m isr_shellIs the refrigerant mass in the shell cavity in g; m isr_accumIs the refrigerant mass in the accumulator in g; p is the temperature measurement point of the inlet of the compressor; q is the temperature measurement point at the outlet of the compressor;
the refrigerant mass in the refrigeration oil is calculated from the refrigeration oil mass and the solubility of the refrigerant in oil, and is represented by the following formula:
s＝f(Tcom,Pc)＝ft(Tp,Tq,Tc)
moil＝ρoilVoil＝ft(Tp,Tq)
wherein s is the solubility of the refrigerant, and m isoilFor the quality of the cooling oil, the TcomIs the temperature of the compressor, the PcFor the condensation pressure, the value of poilFor the cooling oil density, the VoilIs the volume of the cooling oil;
the mass of refrigerant in the accumulator and the housing cavity can be calculated from the volume of the container and the density of the refrigerant, and is represented by the following equation:
mr_accum＝ρin·Vaccum＝ft(Tp)
mr_shell＝ρout·Vshell＝ft(Tq)
wherein, the VaccumAnd said VshellVolumes of the reservoir and the housing cavity, respectively, the pinAnd the rhooutThe refrigerant densities of the inlet and the outlet of the compressor, respectively.
Further, the calculation of the theoretical model of the refrigerant quality of the connecting line based on theoretical and explicit expressions in the step (2) is as follows:
the mass of refrigerant in the connecting line is calculated from the density of refrigerant in the connecting line and is represented by the following formula:
mpipe,j＝ft(ρpipe,j)＝ft(Tpipe,j)
＝ρpipe,j·Lpipe,j·(πd2)/4
wherein, said mpipeFor the refrigerant mass in the connecting line, said LpipeIs the length of the connecting line, d is the diameter of the connecting line, ppipe,jThe density of the refrigerant for the j-th of the connecting line.
Further, the calculation of the locally weighted linear regression model based on data-driven flash tank refrigerant mass calculation in step (3) is as follows:
Mflash＝Mtest-Mcomp_test-Mcond_test-Meavp_test-Mpipe_test
wherein, M isflashIs the refrigerant mass matrix, M, within the flash tanktestIs the initial charge matrix, M, of the flash tank refrigeration systemcomp_testIs the refrigerant mass matrix, M, obtained by the theoretical model of compressor refrigerant mass based on theory and explicit expressioncond_testIs the refrigerant mass matrix, M, of the condenser obtained from the theoretical model of refrigerant mass in the zones of the heat exchanger based on temperature signalseavp_testIs the refrigerant mass matrix, M, of the evaporator obtained by the theoretical model of refrigerant mass of the heat exchanger partitions based on temperature signalspipe_testThe refrigerant quality matrix of the connecting pipeline is obtained by the theoretical model of the refrigerant quality of the connecting pipeline based on theory and display expression;
the parameter matrix of the flash tank is represented by the following formula:
wherein said θ is said flashA machine model parameter matrix; the above-mentioned
An input matrix composed of the characteristic quantities of the flash tank; w is a weight matrix, W (i, i) is the ith term on the diagonal of the weight matrix, and k is a custom parameter;
a predicted value of the refrigerant mass of the flash tank is represented by the following formula:
mflash_pred＝θTxflash
wherein, said mflash_predIs the mass of the flash tank, said x, from the temperature measurement pointsflashAnd the input vector of the flash tank is obtained according to the conversion of the temperature measuring points.
Further, if the flash tank is not present in the refrigeration system, the MflashIs 0.
Further, the refrigerant mass in the refrigeration system in the step (4) is calculated as follows:
wherein, the sigma miIs the sum of the refrigerant masses in the refrigeration system in g; m iscompIs the refrigerant mass in the compressor in g; m iscondIs the refrigerant mass in g in the condenser; m iseavpIs the refrigerant mass in the evaporator in g; m ispipe,jIs the refrigerant mass in g in the jth of said connecting lines; m isflashIs the refrigerant mass in g in the flash tank; the t is the current detection time; the n is the total number of the temperature measuring points; and T is the temperature value of the temperature measuring point and has the unit of ℃.
Further, the refrigerant quality in the refrigeration system in the step (5) is judged as follows:
wherein, the sigma miIs the sum of the refrigerant masses within the refrigeration system in g; m isCharging deviceA charge of the refrigerant mass within the refrigeration system.
Further, the refrigerant leakage detection method is also applicable to a refrigeration system with a liquid storage tank.
The method can save the data amount required by leakage detection, accurately judge whether the refrigeration system leaks and calculate the leakage amount.
The conception, the specific structure and the technical effects of the present invention will be further described with reference to the accompanying drawings to fully understand the objects, the features and the effects of the present invention.
Drawings
FIG. 1 is a computational flow diagram of a data-driven and model-based refrigerant leak detection method for a refrigeration system according to a preferred embodiment of the present invention;
FIG. 2 is a schematic view of the temperature measurement station arrangement of the present invention;
FIG. 3 is a calculated result error graph of the present invention.
Detailed Description
The technical contents of the preferred embodiments of the present invention will be more clearly and easily understood by referring to the drawings attached to the specification. The present invention may be embodied in many different forms of embodiments and the scope of the invention is not limited to the embodiments set forth herein.
As shown in fig. 1, the method comprises the following specific steps:
(1) arranging temperature measuring points in the refrigerating system, and reading the existing experimental working condition data;
(2) establishing a heat exchanger partition refrigerant quality theoretical model based on a temperature signal, and a compressor and connecting pipeline refrigerant quality theoretical model based on theory and an explicit expression;
(3) when the refrigerating system contains a flash evaporator, establishing a local weighted linear regression model based on the calculation of the refrigerant quality of the data-driven flash evaporator;
(4) after new temperature measuring point data are obtained, inputting the data into a theoretical model of a compressor, a heat exchanger, a connecting pipeline and a data model of a flash tank, calculating the mass of the refrigerant in each part, and accumulating to obtain the mass of the refrigerant in the current system;
(5) and comparing the refrigerant quality of the current system with the initial charging amount in the system to obtain the refrigerant leakage condition of the refrigeration system.
As shown in fig. 2, in step (1), a plurality of temperature measuring points may be arranged along the way of the inlet and outlet of each component and the heat exchanger. When the valve A and the valve B in the system are opened and the valve C, the valve D and the valve E are closed, the refrigerating system does not contain a flash tank; when the valves A and B are closed and the valves C, D and E are opened, the refrigerating system contains a flash tank. The arrangement of the temperature measuring points in the two types of systems remains unchanged.
In the step (2), the calculation process of the heat exchanger partition refrigerant quality theoretical model based on the temperature signal specifically comprises the following steps:
for the evaporator, the refrigerant mass is calculated physically from 3 typical state points, including the evaporator inlet-outlet state point and the two-phase region termination state point in the evaporator, as shown in the following formula:
meavp＝ft(Tx...Tx+N,Ty)
＝ft(X5,X6,X7)
for the condenser, the refrigerant mass is calculated from the physical properties of 4 typical state points, including the inlet and outlet state points of the condenser and the starting and ending state points of the two-phase region in the condenser, as shown in the following formula:
mcond＝ft(Ti,Tj...Tj+M,Tk)
＝ft(X1,X2,X3,X4)
the parameters of each state point are obtained according to the measured temperature, as shown in the following formula:
Xi＝(Ti,hi,Xi,ρi),i＝1,2,...,7
in the formula: x1Is the condenser inlet state point; x2Is the initial state point of a two-phase area in the condenser; x3Is a two-phase region termination state point; x4Is the condenser outlet state point; x5Is an evaporator inlet state point; x6Is a termination state point of a two-phase region in the evaporator; x7Is an evaporator port state point; h is the enthalpy value of the state point; x is the dryness of the state point; ρ is the enthalpy value of the state point.
X1And X2Between, X6And X7Between them is a superheat zone, X3And X4The super-cooling areas are all single-phase areas, wherein the refrigerant quality can be directly calculated by the following companies:
in the formula: m isshMass of refrigerant in superheat zone; m isscIs the subcooling zone refrigerant mass; rhoshThe superheat zone refrigerant density; rhoscIs the subcooling zone refrigerant density; v is the phase region volume.
For the refrigerant quality in the two-phase region, a Hughmark model calculation can be used. The phase volume can be calculated from the product of the phase length and the cross-sectional area:
Vj＝AjLj
in the formula: r is a refrigerant; a is air; in is an inlet; out is the outlet; the Heat transfer coefficient was calculated from the correlation (Kim B, Sohn B. an experimental study of Flow wiring in a rectangular channel with offset strip fins. International Journal of Heat and Flow,2006,27(3): 514-.
In the step (2), the calculation process of the compressor refrigerant quality theoretical model based on the theoretical and explicit expressions specifically includes:
the refrigerant in the compressor is distributed primarily in the refrigeration oil, in the accumulator and in the housing cavity, so the refrigerant mass in the compressor can be summed from the refrigerant mass in three sections:
mcomp＝mr_oil+mr_shell+mr_accum
＝ft(Tp,Tq)
in the formula: m iscompIs the total mass of the refrigerant in the compressor, and the unit is g; m isr_ollIs the mass of refrigerant dissolved in the refrigeration oil, in g; m isr_shellIs the refrigerant mass in the shell cavity in units of g; m isr_accumIs the refrigerant mass in the accumulator in g; p is a compressor inlet measuring point; and q is a compressor outlet measuring point.
The refrigerant mass in the refrigeration oil can be calculated from the refrigeration oil mass and the solubility of the refrigerant in the oil:
s＝f(Tcom,Pc)＝ft(Tp,Tq,Tc)
moil＝ρoilVoil＝ft(Tp,Tq)
in the formula: s is the refrigerant solubility; m isoilThe quality of the cooling oil; t iscomIs the compressor temperature; pcIs the condensing pressure; rhooilIs the cooling oil density; voilIs the volume of cooling oil.
The mass of refrigerant in the accumulator and the housing cavity can be calculated from the volume of the vessel and the density of the refrigerant:
mr_accum＝ρin·Vaccum＝ft(Tp)
mr_shell＝ρout·Vshell＝ft(Tq)
in the formula: vaccumAnd VshellThe volumes of the reservoir and the housing cavity, respectively; rhoinAnd ρoutCompressor inlet and outlet refrigerant densities, respectively.
In the step (2), the calculation process of the theoretical model of the refrigerant quality of the connecting pipeline based on the theory and the explicit expression specifically comprises the following steps:
because the external surface area of the connecting pipe is small, the heat exchange with the external environment can be ignored, and the connecting pipe can be treated as a heat insulation pipe, namely, the density of the refrigerant in the pipe is uniform. The mass of refrigerant in the tube can therefore be calculated from the density of the refrigerant in the tube:
mpipe,j＝ft(ρpipe,j)＝ft(Tpipe,j)
＝ρpipe,j·Lpipe,j·(πd2)/4
in the formula: m ispipeThe total mass of the refrigerant in the connecting pipeline; l ispipeIs the length of the connecting tube; d is the diameter of the connecting pipe; g is a gas phase; f is a liquid phase; rhopipe,jThe jth connecting tube refrigerant density.
In the step (3), the calculation process of the local weighted linear regression model based on the data-driven flash tank refrigerant mass calculation specifically includes:
obtaining a refrigerant mass matrix of the flash tank according to the total refrigerant mass of the existing experimental data and the calculation results of the heat exchanger, the compressor and the connecting pipeline model:
Mflash＝Mtest-Mcomp_test-Mcond_test-Meavp_test-Mpipe_test
in the formula: mflashCalculating a refrigerant mass matrix in the flash tank; mtestAn initial charge matrix of the system in the experimental data; mcomp_testThe method comprises the following steps of calculating a compressor refrigerant quality matrix through a compressor model according to experimental data; mcond_testAnd Meavp_testThe mass matrix of the refrigerants of the condenser and the evaporator is obtained through calculation of a heat exchanger model according to experimental data; mpipe_testThe method is a connected pipeline refrigerant quality matrix calculated by a connected pipeline model according to experimental data.
The solution mode of the parameter matrix of the flash tank is as follows:
in the formula: theta is a flash tank model parameter matrix;
an input matrix composed of flash tank characteristic quantities; w is the weight matrix, W (i, i) is the ith term on the diagonal of the weight matrix; and k is a custom parameter.
The predicted value of refrigerant mass in the flash tank is:
mflash_pred＝θTxflash
in the formula: m isflash_predCalculating the mass of the flash tank according to the new temperature measuring point; x is the number offlashThe input vector of the flash tank is obtained according to the conversion of the new temperature measuring point.
If there is no flash tank or receiver in the system, the refrigerant mass for that portion is 0.
In the step (4), the calculated value of the mass of the refrigerant in the current system is as follows:
in the formula, sigma miIs the sum of the refrigerant masses in each part, and the unit is g; m iscompIs the mass of the refrigerant in the compressor, and the unit is g; m iscondIs the mass of the refrigerant in the condenser, and the unit is g; m iseavpIs the mass of refrigerant in the evaporator, in g; m ispipe,jThe mass of the refrigerant in the jth pipeline is g; m isflashIs the mass of refrigerant in the flash tank, in g; t is the current detection time; n is the total number of temperature measuring points; t is the temperature value of the temperature measuring point, and the unit is ℃.
In the step e, the criterion for determining whether the refrigerant leakage occurs in the refrigeration system may be represented as:
in the formula, sigma miIs the sum of the refrigerant masses in each part, and the unit is g; m isCharging deviceThe refrigerant system is charged.
In order to verify the effect of the method, a verification operation was performed using steady-state experimental data provided by a test manufacturer. The experimental working conditions comprise 24 groups of experimental working conditions and cover different working conditions with the charging amounts of 2.03kg, 2.45kg, 2.655kg, 2.855kg, 3.055kg, 3.255kg and 3.455 kg. The verified system configuration adopts the test system configuration of a test manufacturer, and the calculation working condition is subjected to simulation calculation according to the test working condition.
The comparison of the calculated value and the experimental value of the refrigerant quality of the system by the refrigerant leakage detection method is shown in fig. 3. As can be seen in fig. 3, the prediction error of the detection method herein for the refrigerant mass in the system is around ± 2%. Thus, a refrigerant leak detection method is proven to be viable when the refrigerant mass in the refrigerant system varies by more than 5% from the initial system charge.
From the analysis, the invention establishes a refrigeration system refrigerant leakage detection method based on data driving and a model by adopting a method based on data driving and a model. In the invention, the heat exchanger refrigerant quality theoretical model is a heat exchanger partition model based on temperature signals; the refrigerant quality theoretical model of the compressor and the connecting pipeline is a theoretical model based on theory and an explicit expression; the refrigerant mass calculation model of the flash tank is a local weighted linear regression model based on data drive; and the leakage condition of the refrigerant in the system is obtained by accumulating and comparing the results of the refrigerant quality calculation models of all parts. The verification result shows that the detection method has high calculation accuracy and achieves good effect.
The foregoing detailed description of the preferred embodiments of the invention has been presented. It should be understood that numerous modifications and variations could be devised by those skilled in the art in light of the present teachings without departing from the inventive concepts. Therefore, the technical solutions available to those skilled in the art through logic analysis, reasoning and limited experiments based on the prior art according to the concept of the present invention should be within the scope of protection defined by the claims.
Claims (4)
1. A data-driven and model-based refrigerant leak detection method for a refrigeration system, comprising the steps of:
(1) setting a temperature measuring point in a refrigeration system, and reading experimental working condition data;
(2) establishing a heat exchanger partition refrigerant quality theoretical model based on a temperature signal, a compressor refrigerant quality theoretical model based on theory and an explicit expression, and a connecting pipeline refrigerant quality theoretical model based on theory and the explicit expression;
(3) when the refrigerating system comprises a flash tank, establishing a local weighted linear regression model based on the calculation of the refrigerant quality of the data-driven flash tank;
(4) after the data of the temperature measuring points are obtained, the data are input into the heat exchanger partition refrigerant quality theoretical model based on the temperature signals, the compressor refrigerant quality theoretical model based on the theory and the explicit expression, the connecting pipeline refrigerant quality theoretical model based on the theory and the explicit expression and the local weighted linear regression model calculated based on the data-driven flash evaporator refrigerant quality, and the refrigerant quality of the refrigerating system is obtained through accumulation;
(5) comparing the refrigerant quality of the refrigeration system with an initial charge of the refrigeration system;
the temperature measuring points in the step (1) are arranged along the path of the inlet and the outlet of the compressor, the connecting pipeline and the flash tank and the heat exchanger partition; and the arrangement of the temperature measuring points in the refrigerating system is kept unchanged whether the refrigerating system contains a flash tank or not;
the calculation of the heat exchanger partition refrigerant quality theoretical model based on the temperature signals in the step (2) is as follows:
the refrigerant mass of the evaporator includes a state point of an inlet and a state point of an outlet of the evaporator, and a state point at which a two-phase region in the evaporator ends, and is represented by the following formula:
meavp＝ft(Tx...Tx+N,Ty)
＝ft(X5,X6,X7)
the refrigerant mass of the condenser includes a state point of an inlet and a state point of an outlet of the condenser and a state point of a start and a state point of an end of a two-phase region in the condenser, and is represented by the following formula:
mcond＝ft(Ti,Tj...Tj+M,Tk)
＝ft(X1,X2,X3,X4)
the parameters of the state points are represented by the following formula according to the temperature measuring points:
Xi＝(Ti,hi,Xi,ρi),i＝1,2,...,7
wherein, X is1Is the state point of the inlet of the condenser, the X2The state point, the X, for the initiation of the two-phase zone within the condenser3The point of state, the X, of the termination of the two-phase zone within the condenser4Is the state point of the outlet of the condenser, the X5For the state point of the inlet of the evaporator, the X6The state point, the X, of the termination of the two-phase region of the evaporator7The state point for the outlet of the evaporator, the h is the enthalpy of the state point, the X is the dryness of the state point, and the ρ is the density of the state point;
said X1And said X2Between which is a superheat zone, X6And said X7Between them is a superheat zone, X3And said X4The refrigerant quality is represented by the following formula:
wherein, said mshIs the refrigerant mass of the superheat region, the mscThe refrigerant mass of the supercooling region, the pshIs the refrigerant density of the superheat region, the pscIs the refrigerant density of the subcooling zone, V is the volume of the superheat zone or the subcooling zone;
for the refrigerant mass in the two-phase region, a Hughmark model is adopted for calculation, and the volume of the superheat region or the supercooling region is calculated by the product of the length and the cross-sectional area of the superheat region or the supercooling region and is represented by the following formula:
Vj＝AjLj
wherein r is the refrigerant; a is air; the in is an inlet; the out is an outlet;
the theoretical model of compressor refrigerant mass based on theoretical and explicit expressions in the step (2) is calculated as follows:
the refrigerant in the compressor includes the refrigerant in the refrigeration oil, the refrigerant in the accumulator, and the refrigerant in the shell cavity, represented by the formula:
mcomp＝mr_oil+mr_shell+mr_accum
＝ft(Tp，Tq)
wherein, said mcompIs the total mass of the refrigerant in the compressor in g; m isr_oilIs the refrigerant mass in the refrigeration oil in g; m isr_shellIs the refrigerant mass in the shell cavity in g; m isr_accumIs the refrigerant mass in the accumulator in g; p is the temperature measurement point of the inlet of the compressor; q is the temperature measurement point at the outlet of the compressor;
the refrigerant mass in the refrigeration oil is calculated from the refrigeration oil mass and the solubility of the refrigerant in oil, and is represented by the following formula:
s＝f(Tcom，Pc)＝ft(Tp，Tq，Tc)
moil＝ρoilVoil＝ft(Tp，Tq)
wherein s is the solubility of the refrigerant, and m isoilFor the quality of the cooling oil, the TcomIs the temperature of the compressor, the PcFor the condensation pressure, the value of poilFor the cooling oil density, the VoilIs the volume of the cooling oil;
the mass of refrigerant in the accumulator and the housing cavity can be calculated from the volume of the container and the density of the refrigerant, and is represented by the following equation:
mr_accum＝ρin·Vaccum＝ft(Tp)
mr_shell＝ρout·Vshell＝ft(Tq)
wherein, the VaccumAnd said VshellVolumes of the reservoir and the housing cavity, respectively, the pinAnd the rhooutA refrigerant density of the inlet and the outlet of the compressor, respectively;
the calculation of the theoretical model of the refrigerant mass of the connecting pipeline based on the theory and the explicit expression in the step (2) is as follows:
the mass of refrigerant in the connecting line is calculated from the density of refrigerant in the connecting line and is represented by the following formula:
mpipe，j＝ft(ρpipe，j)＝ft(Tpipe，j)
＝ρpipe，j·Lpipe，j·(πd2)/4
wherein, said mpipeFor the refrigerant mass in the connecting line, said LpipeIs the length of the connecting line, d is the diameter of the connecting line, ppipe，j(ii) the density of the refrigerant for the jth of the connecting lines;
the calculation of the locally weighted linear regression model based on data driven flash tank refrigerant mass calculation in said step (3) is as follows:
Mflash＝Mtest-Mcomp_test-Mcond_test-Meavp_test-Mpipe_test
wherein, M isflashIs the refrigerant mass matrix, M, within the flash tanktestIs the initial charge matrix, M, of the refrigeration system with flash tankcomp_testIs the refrigerant mass matrix, M, obtained by the theoretical model of compressor refrigerant mass based on theory and explicit expressioncond_testIs the refrigerant mass matrix, M, of the condenser obtained from the theoretical model of refrigerant mass in the zones of the heat exchanger based on temperature signalseavp_testIs the refrigerant mass matrix, M, of the evaporator obtained by the theoretical model of refrigerant mass of the heat exchanger partitions based on temperature signalspipe_testThe refrigerant quality matrix of the connecting pipeline is obtained by the theoretical model of the refrigerant quality of the connecting pipeline based on theory and display expression;
the parameter matrix of the flash tank is represented by the following formula:
wherein θ is the flash tank model parameter matrix; the above-mentioned
Is an input matrix composed of characteristic quantities of the flash tank, the
A transpose matrix that is a flash tank refrigerant quality matrix; w is a weight matrix, W (i, i) is the ith term on the diagonal of the weight matrix, and k is a custom parameter;
a predicted value of the refrigerant mass of the flash tank is represented by the following formula:
mflash_pred＝θTxflash
wherein, said mflash_predIs the mass of the flash tank, said x, from the temperature measurement pointsflashIs the input vector of the flash tank obtained by the conversion of the temperature measuring points;
if the flash tank is not present in the refrigeration system, the MflashIs 0.
2. The data driven and model based refrigerant leak detection method of claim 1, wherein the refrigerant mass in the flash tank refrigerant system in step (4) is calculated as follows:
wherein, the sigma miIs the sum of the refrigerant masses in the refrigeration system in g; m iscompIs the refrigerant mass in the compressor in g; m iscondIs the refrigerant mass in g in the condenser; m iseavpIs the refrigerant mass in the evaporator in g; m ispipe，jIs the refrigerant mass in g in the jth of said connecting lines; m isflashIs the refrigerant in the flash tankMass in g; the t is the current detection time; the n is the total number of the temperature measuring points; and T is the temperature value of the temperature measuring point and has the unit of ℃.
3. The data driven and model based refrigerant leak detection method of claim 1 wherein the refrigerant quality determination in the refrigerant system in step (5) is as follows:
wherein, the sigma miIs the sum of the refrigerant masses within the refrigeration system in g; m isCharging deviceA charge of the refrigerant mass within the refrigeration system.
4. The data driven and model based refrigerant leak detection method of claim 1, wherein the refrigerant leak detection method is also applicable to a refrigerant system with a receiver tank.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910053039.2A CN109781345B (en) | 2019-01-21 | 2019-01-21 | Refrigerating system refrigerant leakage detection method based on data driving and model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910053039.2A CN109781345B (en) | 2019-01-21 | 2019-01-21 | Refrigerating system refrigerant leakage detection method based on data driving and model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109781345A CN109781345A (en) | 2019-05-21 |
| CN109781345B true CN109781345B (en) | 2021-05-07 |
Family
ID=66501615
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910053039.2A Active CN109781345B (en) | 2019-01-21 | 2019-01-21 | Refrigerating system refrigerant leakage detection method based on data driving and model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109781345B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110779249B (en) * | 2019-11-06 | 2021-03-26 | 大连理工大学 | On-line diagnosis method for sensor concurrent faults in solar heat pump system |
| CN111964846A (en) * | 2020-09-28 | 2020-11-20 | 郑州斯倍思机电有限公司 | By using CO2System for detecting tightness of workpiece |
| CN113361158B (en) * | 2021-05-31 | 2022-04-05 | 江西理工大学 | Method and system for measuring metal temperature in inner cavity of continuous casting crystallizer |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000258039A (en) * | 1999-03-08 | 2000-09-22 | Daikin Ind Ltd | Refrigerating device for show case |
| US20110112814A1 (en) * | 2009-11-11 | 2011-05-12 | Emerson Retail Services, Inc. | Refrigerant leak detection system and method |
| CN105387665B (en) * | 2015-11-25 | 2018-08-10 | 东南大学 | A kind of defrosting control method with air source heat pump comprehensive performance most preferably target |
| CN105424150B (en) * | 2015-12-23 | 2018-03-09 | 珠海格力电器股份有限公司 | A kind of refrigerant quality test device and method of testing |
| CN106568249A (en) * | 2016-11-01 | 2017-04-19 | 南京师范大学 | Non-azeotropic refrigerant charge determining method |
| CN106524551A (en) * | 2016-11-07 | 2017-03-22 | 清华大学 | Mass and flow measuring method and device of refrigerant in refrigerating system and measuring instrument |
| CN106766440A (en) * | 2016-12-31 | 2017-05-31 | 广州市粤联水产制冷工程有限公司 | A kind of refrigerant leakage early warning system and method for early warning based on internet |
| CN106766439A (en) * | 2016-12-31 | 2017-05-31 | 广州市粤联水产制冷工程有限公司 | A kind of refrigerant leakage detecting device and detection method |
| CN108763721B (en) * | 2018-05-23 | 2022-09-30 | 特灵空调系统(中国)有限公司 | Simulation method for air conditioning system charging amount |
-
2019
- 2019-01-21 CN CN201910053039.2A patent/CN109781345B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109781345A (en) | 2019-05-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109781345B (en) | Refrigerating system refrigerant leakage detection method based on data driving and model | |
| JP5183609B2 (en) | Refrigeration air conditioner | |
| JP5525965B2 (en) | Refrigeration cycle equipment | |
| Li et al. | Decoupling features and virtual sensors for diagnosis of faults in vapor compression air conditioners | |
| CN105091215B (en) | Multi-gang air-conditioner lacks the start-up detection method and device of fluorine | |
| Grace et al. | Sensitivity of refrigeration system performance to charge levels and parameters for on-line leak detection | |
| EP2546588B1 (en) | Refrigeration and air conditioning device | |
| Kim et al. | Fault detection and diagnostics analysis of air conditioners using virtual sensors | |
| CN104390803B (en) | Refrigerant supply device for refrigerator heat exchanger performance test system and test method | |
| CN102313471B (en) | The function of cooling system monitors and/or control method and corresponding cooling system | |
| CN106288566A (en) | Refrigeration system is intersected map parts | |
| Navarro-Esbri et al. | A vapour compression chiller fault detection technique based on adaptative algorithms. Application to on-line refrigerant leakage detection | |
| CN107632042A (en) | The single-phase local heat transfer test platform of shell-and-tube heat exchanger and method of testing | |
| Tran et al. | Refrigerant-based measurement method of heat pump seasonal performances | |
| CN108061411A (en) | The coolant injection method and air-conditioning of air-conditioning | |
| JP4301085B2 (en) | Deterioration diagnosis system for heat source equipment | |
| Hu et al. | Multiple simultaneous faults’ impacts on air-conditioner behavior and performance of a charge diagnostic method | |
| CN108458888B (en) | Performance testing device for low-temperature heat exchanger with temperature range below K | |
| CN105865661B (en) | Positive displacement enthalpy-increasing compressor refrigeration capacity test device and test method | |
| Wichman et al. | Fault detection and diagnostics for commercial coolers and freezers | |
| CN105865574A (en) | Thermal type liquid level meter | |
| CN102520010A (en) | Condenser dirt detection method for vapor compression cycle cold water unit | |
| Hu | Development and demonstration of a method to determine residential air-conditioning system fault prevalence | |
| Tran et al. | In-situ method for air-to-air heat pump seasonal performance determination including steady-state and dynamic operations | |
| Lee et al. | Refrigerant charge prediction based on start-up characteristics of an air conditioner using a gray box model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |