Air conditioner refrigerant amount prediction method, air conditioner, and storage medium

Abstract

The application relates to the air conditioning technical field, and provides an air conditioner refrigerant quantity prediction method, an air conditioner and a storage medium, which comprise the following steps: acquiring all key parameter data corresponding to different refrigerant quantities of an air conditioner; the key parameter data comprises suction superheat data and discharge superheat data; all the key parameter data is segmented to form a plurality of segments; the ratio of the key parameter data corresponding to each segment to all the key parameter data is calculated; the ratio is input into a refrigerant quantity prediction value model to calculate a predicted refrigerant quantity; and the refrigerant quantity prediction value model is obtained by training based on historical key parameter data. The patent adopts a multi-dimensional linear method based on a large amount of random data, adopts the proportion of key parameters for analysis, and has higher accuracy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of air conditioners, and provides an air conditioner refrigerant quantity prediction method, an air conditioner and a storage medium. BACKGROUND

[0002] When the air conditioning system is running, the refrigerant quantity is an essential factor, and the amount of refrigerant directly affects the effect of the multi-split air conditioner. The prediction of the refrigerant quantity of the air conditioning system is a very necessary technology. The multi-split air conditioner is a kind of air conditioning system, and the running parts thereof are more than those of the conventional air conditioning system, the running mechanism is more complex, and the running parameters are more. The parameters involved include the running mode, the environmental temperature, the compressor frequency, the high pressure, the low pressure, the exhaust temperature, the exhaust superheat, the suction temperature, the suction superheat, the supercooling degree, the electronic expansion valve opening degree, the electromagnetic valve opening condition, the oil return temperature, etc.

[0003] In the prior art, there are studies that extract key parameters (such as high pressure and suction superheat), but when judging the refrigerant quantity, a specific data is used instead of big data, and the use of a specific data for refrigerant quantity judgment is prone to judgment errors.

[0004] For example, patent CN113654182A discloses a method for detecting refrigerant leakage, a computer readable storage medium and an air conditioner, but the calculated refrigerant quantity extracted thereby is a specific data, i.e., a key parameter pressure value, which is prone to errors. Patent CN113739348A discloses a refrigerant state detection method, an air conditioner and a storage medium, and the model established thereby is a neural network model, and no multi-dimensional linear model is provided. Patent CN110131838A provides an air conditioning unit control method, device, computer equipment and storage medium, and the specific data extracted thereby is not provided with an explicit linear formula calculation method. SUMMARY

[0005] In view of the above problems, the present application provides an air conditioner refrigerant quantity prediction method, an air conditioner and a storage medium. The present application establishes a multi-dimensional linear method based on a large amount of random data, analyzes the proportion of key parameters instead of simply judging the refrigerant quantity by a point of data, and provides an explicit linear formula calculation method, which has higher accuracy.

[0006] The technical scheme provided by the present application is as follows:

[0007] A refrigerant quantity prediction method of an air conditioner, comprising the following steps:

[0008] Obtaining all key parameter data corresponding to different refrigerant quantities of the air conditioner; the key parameter data includes suction superheat data and exhaust superheat data;

[0009] Segmenting all the key parameter data to form several segments;

[0010] Calculating the ratio of the key parameter data of each segment to the total key parameter data; establishing a multi-order mathematical model of the segmented key parameter and the refrigerant amount relationship, and solving through matrix to obtain the calculation coefficient of the key parameter data of different segments, and constructing a refrigerant amount prediction value model based on the calculation coefficient;

[0011] Inputting the ratio of the total key parameter data into the refrigerant amount prediction value model to calculate the predicted refrigerant amount.

[0012] Further, the maximum value and the minimum value of the key data are obtained, the difference between the maximum value and the minimum value is divided by the number of segments, and the key parameter data is evenly divided into multiple segments.

[0013] Further, the total key parameter data corresponding to different refrigerant amounts of the air conditioner is obtained, specifically: obtaining the total key parameter data corresponding to 10%-150% of the system standard refrigerant amount of the air conditioner.

[0014] Further, the calculation method of the predicted refrigerant amount is:

[0015] F(x)=A1*f(SH≤M1)+A2*f(M1<SH≤M2)+A3*f(M2<SH≤M3)+……

[0016] +A m-1 *f(M m-2 <SH≤M m-1 )+A m *f(M m-1 <SH)+B1*g(DIS≤N1)+B2*g(N1<DIS≤N2)+B3*g(N2<DIS≤N3)+……+B n-1 *g(N n-2 <DIS≤N n-1 )+B n *g(N n-1 <DIS);

[0017] F(x) is the proportion of the predicted refrigerant amount to the system standard refrigerant amount; A1, A2, … A m , B1, B2, … B n are calculation coefficients; SH is the suction superheat, and DIS is the discharge superheat; the suction superheat is divided into m equal parts, and the end point values are M1, M2, … M m-1 ; the discharge superheat is divided into n equal parts, and the end point values are N1, N2, … N n-1 ; f(M m-2 <SH≤M m-1 ) represents that in a certain group of data, (Mm-2 , M m-1 ] segment corresponding to all the values of the intake air superheat in the sum of all the values of the intake air superheat; g(N n-2 <DIS≤N n-1 ) represents the proportion of the sum of all the values of the exhaust air superheat corresponding to the segment [N n-2 , N n-1 ] in the sum of all the values of the exhaust air superheat.

[0018] Further, the expression of the multi-order matrix model is:

[0019] F(1) = A1*M(1,1) + A2*M(1,2) + … + A m *M(1,m)

[0020] +B1*N(1,1) + B2*N(1,2) + … + B n *N(1,n)

[0021] F(2) = A1*M(2,1) + A2*M(2,2) + … + A m *M(2,m)

[0022] +B1*N(2,1) + B2*N(2,2) + … + B n *N(2,n)

[0023] …

[0024] F(m) = A1*M(m,1) + A2*M(m,2) + … + A m *M(m,m)

[0025] +B1*N(m,1) + B2*N(m,2) + … + B n *N(m,n)

[0026] …

[0027] F(m+n) = A1*M(m+n,1) + A2*M(m+n,2) + … + A m *M(m+n,m)

[0028] +B1*N(m+n,1) + B2*N(m+n,2) + … + B n *N(m+n,n)

[0029] A1, A2, … A m , B1, B2, … B nare coefficients; F(1), F(2), …, F(m), …, F(m+n) represent the proportion of the historical predicted refrigerant amount of the 1st, 2nd, …, mth, …, m+nth group to the system standard refrigerant amount; the suction gas superheat degree is divided into m sections, and the end point values are M1, M2, …, M m-1 ; the discharge gas superheat degree is divided into n equal parts, and the end point values are N1, N2, …, N n-1 ; M(1,1), M(1,2), …, M(m,m), M(m+n,m) are the values of f(SH≤M1), f(M1 m-1 < SH), f(M m-1 < SH) respectively; N(1,1), N(1,2), …, N(n,n), M(m+n,n) are the values of f(DIS≤N1), f(N1 m-1 < DIS), f(N m-1 < DIS) respectively; f(M m-2 < SH≤M m-1 ) represents the proportion of the sum of all suction gas superheat degree values corresponding to the section (M m-2 , M m-1 ] in a certain group of data in the sum of all suction gas superheat degree values; g(N n-2 < DIS≤N n-1 ) represents the proportion of the sum of all discharge gas superheat degree values corresponding to the section (N n-2 , N n-1 ] in a certain group of data in the sum of all discharge gas superheat degree values.

[0030] The application also provides an air conditioner, which comprises a processor, a memory, and a refrigerant amount prediction program stored on the memory and executable on the processor, wherein the refrigerant amount prediction program is executed by the processor to adopt the above method.

[0031] The application also provides a computer readable storage medium, which stores a refrigerant amount prediction program, wherein the refrigerant amount prediction program is executed by a processor to adopt the above method.

[0032] Advantages

[0033] The application provides a multi-dimensional linear refrigerant amount prediction method for an air conditioner based on big data, extracts key parameters of suction gas superheat degree and discharge gas superheat degree from a plurality of groups of modeling basic data of different refrigerant amounts of the air conditioner, establishes a mathematical model, and determines a refrigerant amount calculation formula through matrix solution, instead of simply determining the refrigerant amount through data of a single point.

[0034] The patent proposes real-time judgment of system refrigerant amount, which is judged by cumulative data. When extracting key parameters, the patent uses the combined action of suction superheat and discharge superheat to avoid errors of a single data. The patent establishes a multi-dimensional linear model based on a large amount of random data, analyzes the proportion of key parameters, and does not simply judge the refrigerant amount by a single point of data. The patent proposes a clear linear formula calculation method, which can realize a thousandth refrigerant amount prediction and has higher accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 Flow chart for air conditioner refrigerant amount prediction method;

[0036] Figure 2 Flow chart for air conditioner refrigerant amount prediction model construction. DETAILED DESCRIPTION

[0037] Example 1

[0038] As shown in Figure 1 and Figure 2 , the embodiment provides an air conditioner refrigerant amount prediction method, including the following steps:

[0039] Obtain all key parameter data corresponding to different refrigerant amounts of the air conditioner; the key parameter data includes suction superheat data and discharge superheat data;

[0040] Segment all key parameter data to form several segments;

[0041] Calculate the ratio of the key parameter data of each segment to all key parameter data;

[0042] Input the ratio into a refrigerant amount prediction value model to calculate the predicted refrigerant amount; the refrigerant amount prediction value model is obtained based on historical key parameter data training.

[0043] The refrigerant amount prediction value model is obtained by the following steps:

[0044] Obtain key parameter data corresponding to different refrigerant amounts of the air conditioner; the key parameter data includes historical suction superheat data and historical discharge superheat data;

[0045] Segment the historical suction superheat data and historical discharge superheat data respectively to form several segmented historical suction superheat data and segmented historical discharge superheat data;

[0046] Calculate the historical suction superheat ratio of each segmented historical suction superheat data to the historical suction superheat data; calculate the historical discharge superheat ratio of each segmented historical discharge superheat data to the historical discharge superheat data;

[0047] Based on each historical suction superheat ratio and historical exhaust superheat ratio and corresponding refrigerant amount, a multi-order matrix model is established, and a calculation coefficient corresponding to each historical suction superheat ratio and historical exhaust superheat ratio is obtained by solving, and a refrigerant amount prediction value model is constructed based on the calculation coefficient.

[0048] Embodiment 2

[0049] The modeling basic data of the patent is a plurality of basic operation data of a multi-split air conditioner with different refrigerant amounts; the refrigerant amounts cover the refrigerant amounts that may occur during normal operation of the multi-split air conditioner, such as 10% to 150% refrigerant amounts; the number of groups of the basic operation data is greater than or equal to 2, and a plurality of groups are set between the minimum refrigerant amount and the maximum refrigerant amount in an average distribution, and the more the number of groups, the higher the accuracy of the established model.

[0050] The modeling basic data of the multi-split air conditioner of each group of refrigerant amounts is a group of big data, which should include the refrigeration and heating modes of the unit, the operating parameters when different outdoor units output, and usually includes the operating mode, the environment temperature, the compressor frequency, the high pressure, the low pressure, the exhaust temperature, the exhaust superheat, the suction temperature, the suction superheat, the supercooling degree, the electronic expansion valve opening degree, etc.

[0051] The key parameters extracted by the patent are the suction superheat and the exhaust superheat, and the extraction method is to calculate the proportion of the suction superheat or the exhaust superheat in the entire data within a certain value range.

[0052] Suction superheat = compressor suction temperature detected by the unit - low pressure side saturation temperature; the compressor suction temperature is measured by a compressor suction temperature bulb; the low pressure side saturation temperature is converted into the corresponding saturation temperature by measuring the low pressure value by a low pressure sensor;

[0053] Exhaust superheat = compressor exhaust temperature detected by the unit - high pressure saturation temperature; the compressor exhaust temperature is measured by a compressor exhaust temperature bulb; the high pressure side saturation temperature is converted into the corresponding saturation temperature by measuring the high pressure value by a high pressure sensor;

[0054] Through a plurality of modeling basic data of the multi-split air conditioner with different refrigerant amounts, a multi-order mathematical model based on the suction superheat and the exhaust superheat is established, and a calculation formula of the refrigerant amount is obtained by matrix solving;

[0055] As follows:

[0056] F(x) = A1*f(SH≤M1) + A2*f(M1

[0057] +A m-1 *f(M m-2 <SH≤M m-1 )+A m *f(Mm-1 <SH)+B1*g(DIS≤N1)+B2*g(N1<DIS≤N2)+B3*g(N2<DIS≤N3)+……+B n-1 *g(N n-2 <DIS≤N n-1 )+B n *g(N n-1 <DIS);

[0058] f(M m-2 <SH≤M m-1 ) indicates that in a certain set of data, the intake superheat is within (M m-2 M m-1 The proportion of g(N) n-2 <DIS≤N n-1 ) indicates that in a certain set of data, the exhaust superheat is within (N) n-2 N n-1 The percentage of superheat; SH represents intake superheat; DIS represents exhaust superheat; the intake superheat is divided into m equal parts, with endpoint values ​​M1, M2, ... M m-1 The exhaust superheat is divided into n equal parts, with endpoint values ​​of N1, N2...N n-1 ;

[0059] F(x) represents the proportion of the predicted refrigerant quantity relative to the standard refrigerant quantity of the system. If the standard refrigerant quantity of the system is 100%, then F(x) usually takes a value range of 10% to 150%, that is, the proportion of the refrigerant quantity is 10% to 150% of the standard refrigerant quantity.

[0060] A1, A2, ... A m B1, B2...B n The calculation coefficient for refrigerant quantity needs to be solved by establishing a mathematical model;

[0061] The data of m+n groups with different refrigerant quantities are used as the basic data for modeling;

[0062] F(1), F(2), ..., F(m), ..., F(m+n) represent the basic values ​​for refrigerant quantity modeling in the 1st, 2nd, ..., mth, ..., m+nth groups. These basic values ​​for refrigerant quantity cover the commonly used refrigerant quantities in multi-split systems.

[0063] M(1,1) represents the value of f(SH≤M1) in the first set of basic modeling data;

[0064] M(1,2) represents the value of f(M1<SH≤M2) in the first set of basic modeling data;

[0065] ...

[0066] M(1, m-1) represents the first set of basic modeling data, where f(M) = ...m-2 < SH ≤ M m-1 ) of the first group of modeling basis data;

[0067] M(1, m) represents the value of f(M m-1 < SH) of the first group of modeling basis data;

[0068] M(m, m) represents the value of f(M m-1 < SH) of the mth group of modeling basis data;

[0069] M(m+n, m) represents the value of f(M m-1 < SH) of the (m+n)th group of modeling basis data;

[0070] N(1, 1) represents the value of g(DIS≤N1) of the first group of modeling basis data;

[0071] N(1, 2) represents g(N1<DIS≤N2) of the first group of modeling basis data;

[0072] …

[0073] N(1, n-1) represents the value of g(N n-2 < DIS≤N n-1 ) of the first group of modeling basis data;

[0074] N(1, n) represents the value of g(N n-1 < DIS) of the first group of modeling basis data;

[0075] N(m+n, n) represents the value of g(N n-1 < DIS) of the (m+n)th group of modeling basis data;

[0076] Through m+n groups of data; a (m+n)×(m+n) order matrix can be established as follows:

[0077] F(1)=A1*M(1,1)+A2*M(1,2)+…+A m *M(1, m)

[0078] +B1*N(1,1)+B2*N(1,2)+…+B n *N(1, n)

[0079] F(2)=A1*M(2,1)+A2*M(2,2)+…+A m *M(2, m)

[0080] +B1*N(2,1)+B2*N(2,2)+…+B n *N(2, n)

[0081] …

[0082] F(m) = A1*M(m,1) + A2*M(m,2) +... + Am*M(m,m) + B1*N(m,1) + B2*N(m,2) +... + Bn*N(m,n) m *M(m,m)

[0083] +B1*N(m,1) + B2*N(m,2) +... + Bn*N(m,n) n *N(m,n)

[0084] ...

[0085] F(m+n) = A1*M(m+n,1) + A2*M(m+n,2) +... + Am*M(m+n,m) + B1*N(m+n,1) + B2*N(m+n,2) +... + Bn*N(m+n,n) m *M(m+n,m)

[0086] +B1*N(m+n,1) + B2*N(m+n,2) +... + Bn*N(m+n,n) n *N(m+n,n)

[0087] That is, can be expressed as: matrix F = matrix MN x matrix AB

[0088] F(1) = A1*M(1,1) + A2*M(1,2) +... + Am*M(1,m) + B1*N(1,1) + B2*N(1,2) +... + Bn*N(1,n)

[0089] F(2) = A1*M(2,1) + A2*M(2,2) +... + Am*M(2,m) + B1*N(2,1) + B2*N(2,2) +... + Bn*N(2,n)

[0090] ...

[0091] ...

[0092] F(m) = A1*M(m,1) + A2*M(m,2) +... + Am*M(m,m) + B1*N(m,1) + B2*N(m,2) +... + Bn*N(m,n)

[0093] F(m+1) = A1*M(m+1,1) + A2*M(m+1,2) +... + Am*M(m+1,m) + B1*N(m+1,1) + B2*N(m+1,2) +... + Bn*N(m+1,n)

[0094] ...

[0095] ...

[0096] F(m+n) = A1*M(m+n, 1) + A2*M(m+n, 2) +... + Am*M(m+n, m) + B1*N(m+n, 1) + B2*N(m+n, 2) +... + Bn*N(m+n, n)

[0097] The matrix F can be represented as:

[0098]

[0099] The matrix AB can be represented as:

[0100]

[0101] The matrix MN can be represented as:

[0102] M(1, 1), M(1, 2),..., M(1, m), N(1, 1), N(1, 2),..., N(1, n)

[0103] M(2, 1), M(2, 2),..., M(2, m), N(2, 1), N(2, 2),..., N(2, n)

[0104] ...

[0105] ...

[0106] M(m, 1), M(m, 2),..., M(m, m), N(m, 1), N(m, 2),..., N(m, n)

[0107] M(m+1, 1), M(m+1, 2),..., M(m+1, m), N(m+1, 1), N(m+1, 2),..., N(m+1, n)

[0108] ...

[0109] ...

[0110] M(m+n, 1), M(m+n, 2),..., M(m+n, m), N(m+n, 1), N(m+n, 2),..., N(m+n, n)

[0111] When the M+N groups of modeling base data are given, the matrix F is known, the matrix MN is known, and by solving the (m+n) x (m+n) order matrix as above, the solution of the matrix AB can be obtained, that is, the values of A1, A2,... Am, B1, B2,... Bn can be obtained. m n ;

[0112] ​For any set of multi-split air conditioning data, the refrigerant quantity F(x) can be calculated using the following formula: F(x) = A1*f(SH≤M1) + A2*f(M1<SH≤M2) + A3*f(M2<SH≤M3) + ... + A m-1 *f(M m-2 <SH≤M m-1 )+A m *f(M m-1 <SH)+B1*g(DIS≤N1)+B2*g(N1<DIS≤N2)+B3*g(N2<DIS≤N3)+……+B n-1 *g(N n-2 <DIS≤N n-1 )+B n *g(N n-1 <DIS)

[0113] In this patent, the number of basic modeling data sets, m+n, can be adjusted according to the actual amount of data available. m and n are integers, and the number of m and n does not need to be the same. The number of sets of m+n should be less than or equal to the number of basic modeling data sets with known refrigerant quantity. The larger the value of m+n, the more accurate the refrigerant quantity prediction formula will be. At the same time, the refrigerant quantity prediction method can be optimized at any time based on newly added modeling data.

[0114] Example 3

[0115] An air conditioner includes a processor, a memory, and a refrigerant quantity prediction program stored in the memory and executable on the processor, wherein the refrigerant quantity prediction program is executed by the processor using the method described in Example 1.

[0116] Example 4

[0117] A computer-readable storage medium storing a refrigerant quantity prediction program, which, when executed by a processor, employs the method described in Example 1.

[0118] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0119] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart Figure 1 one or more flows and / or blocks. Figure 1 one or more flows and / or blocks.

[0120] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart Figure 1 one or more flows and / or blocks. Figure 1 one or more flows and / or blocks.

[0121] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart Figure 1 one or more flows and / or blocks. Figure 1 one or more flows and / or blocks.

[0122] The above only is the preferred embodiment of the present application, it should be pointed out that, for those skilled in the technical field, without departing from the technical principles of the present application, can also make a number of improvements and variations, these improvements and variations should also be considered as the protection scope of the present application.

Claims

1. A method for predicting refrigerant quantity in an air conditioner, characterized in that, Includes the following steps: Acquire all key parameter data corresponding to different refrigerant quantities of the air conditioner; the key parameter data includes suction superheat data and exhaust superheat data; specifically, acquiring all key parameter data corresponding to different refrigerant quantities of the air conditioner involves acquiring all key parameter data corresponding to 10%-150% of the system standard refrigerant quantity of the air conditioner. All key parameter data are segmented into several segments; Calculate the ratio of the key parameter data corresponding to each segment to all data of that key parameter; establish a multi-order mathematical model of the relationship between the key parameters of each segment and the refrigerant quantity; and obtain the calculation coefficients of the key parameter data of different segments through matrix solving; and construct a refrigerant quantity prediction model based on the calculation coefficients. The ratio of all the key parameter data is input into the refrigerant quantity prediction model to calculate the predicted refrigerant quantity.

2. The refrigerant quantity prediction method for an air conditioner according to claim 1, characterized in that, Obtain the maximum and minimum values ​​of the key data, divide the difference between the maximum and minimum values ​​by the number of segments, and divide the key parameter data into multiple segments on an average basis.

3. The method for predicting refrigerant quantity in an air conditioner according to claim 1, characterized in that, The method for calculating the predicted refrigerant quantity is as follows: F(x)=A1* f(SH≤M1)+ A2*f(M1<SH≤M2)+ A3*f(M2<SH≤M3)+……+ A m-1 *f(M m-2 <SH≤M m-1 )+ A m *f(M m-1 <SH)+ B1* g(DIS≤N1)+ B2* g(N1<DIS≤N2)+ B3* g(N2<DIS≤N3)+……+ B n-1 * g(N n-2 <DIS≤N n-1 )+ B n * g(N n-1 <DIS); F(x) represents the proportion of the predicted refrigerant quantity relative to the system's standard refrigerant quantity; A1, A2, ..., A m B1, B2...B n The calculation coefficients are: SH represents the intake superheat, and DIS represents the exhaust superheat. The intake superheat is divided into m equal parts, with endpoint values ​​M1, M2, ..., M... m-1 The exhaust superheat is divided into n equal parts, with endpoint values ​​of N1, N2...N n-1 ;f(M m-2 <SH≤M m-1 ) indicates that in a certain set of data, (M m-2 M m-1 The proportion of the sum of all intake superheat values ​​corresponding to segment ] to the total sum of all intake superheat values; g (N n-2 <DIS≤N n-1 ) represents a set of data in which (N n-2 N n-1 The proportion of the sum of all exhaust superheat values ​​corresponding to this segment to the total sum of all exhaust superheat values.

4. The refrigerant quantity prediction method for an air conditioner according to claim 1, characterized in that, The expression for the multi-order matrix model is: F(1)= A1*M(1,1)+ A2* M(1,2)+……+ A m *M(1,m) + B1*N(1,1)+ B2*N(1,2)+……+ B n *N(1,n) F(2)= A1*M(2,1)+ A2* M(2,2)+……+ A m *M(2,m) + B1*N(2,1)+ B2*N(2,2)+……+ B n *N(2,n) …… F(m)= A1*M(m,1)+ A2* M(m,2)+……+ A m *M(m,m) + B1*N(m,1)+ B2*N(m,2)+……+ B n *N(m,n) …… F(m+n)= A1*M(m+n,1)+ A2* M(m+n,2)+……+ A m *M(m+n,m) + B1*N(m+n,1)+ B2*N(m+n,2)+……+ B n *N(m+n,n); A1, A2, ... A m B1, B2...B n For calculation coefficients; F(1), F(2), ..., F(m), ..., F(m+n) represent the proportion of the historical predicted refrigerant quantity in groups 1, 2, ..., m, ..., m+n relative to the standard refrigerant quantity in the system; the suction superheat is divided into m segments, with endpoint values ​​M1, M2, ..., M m-1 The exhaust superheat is divided into n equal parts, with endpoint values ​​of N1, N2...N n-1 M(1,1), M(1,2)...M(m,m), M(m+n,m) are respectively f(SH≤M1), f(M1<SH≤M2), f(M... m-1 <SH), f(M) m-1 The values ​​of <SH) are: N(1,1), N(1,2)...N(n,n), and M(m+n,n), respectively, f(DIS≤N1), f(N1<DIS≤N2), and f(N... m-1 <DIS), f(N) m-1 The value of <DIS); f(M) m-2 <SH≤M m-1 ) indicates that in a certain set of data, (M m-2 M m-1 The proportion of the sum of all intake superheat values ​​corresponding to segment ] to the total sum of all intake superheat values; g (N n-2 <DIS≤N n-1 ) represents a set of data in which (N n-2 N n-1 The proportion of the sum of all exhaust superheat values ​​corresponding to this segment to the total sum of all exhaust superheat values.

5. An air conditioner, characterized in that, The air conditioner includes a processor, a memory, and a refrigerant quantity prediction program stored in the memory and executable on the processor, wherein the refrigerant quantity prediction program is executed by the processor using the method described in any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a refrigerant quantity prediction program, which, when executed by a processor, employs the method described in any one of claims 1 to 4.

Images