A method for diagnosing faults in a liquid chromatograph, a liquid chromatograph, an apparatus and equipment

Abstract

This application provides a method, instrument, apparatus, and device for diagnosing liquid chromatograph (LC) faults, relating to the field of sample detection technology. The method includes: determining the parameter values ​​to be utilized for each specified pressure dimension based on the pressure values ​​of the infusion pump in the LC during the current pump cycle and a preset number of pump cycles prior to the current pump cycle; determining the membership value to be detected corresponding to the parameter value to be utilized for each specified pressure dimension based on the membership function corresponding to that specified pressure dimension; combining each membership value to be detected, using a pre-constructed fuzzy inference model, determining the probability that the operating state represented by the parameter value to be utilized belongs to each preset operating state; the fuzzy inference model is constructed based on the reference membership values ​​corresponding to the parameter values ​​of each specified pressure dimension under each preset operating state; and determining the preset operating state with the highest probability as the fault diagnosis result for the current pump cycle. This allows for timely detection of LC faults.

Description

Technical Field

[0001] This application relates to the field of sample testing technology, and in particular to a method for diagnosing faults in a liquid chromatograph, a liquid chromatograph, an apparatus, and equipment. Background Technology

[0002] Liquid chromatographs (LCs) are sophisticated analytical instruments widely used in sample testing. However, in practical applications, LCs may malfunction, affecting operational efficiency and the reliability of sample testing.

[0003] In the event of certain malfunctions, such as worn seals, failure of check valves, or depletion of mobile phase, the liquid chromatograph will not automatically report an error. Instead, it will continue to operate according to the set program, which may lead to damage to other components in the liquid chromatograph.

[0004] Therefore, there is an urgent need for a method that can detect malfunctions in liquid chromatographs in a timely manner. Summary of the Invention

[0005] The purpose of this application is to provide a method, instrument, apparatus, and device for diagnosing faults in a liquid chromatograph, so as to achieve timely detection of faults in the liquid chromatograph. The specific technical solution is as follows:

[0006] A first aspect of this application provides a method for diagnosing faults in a liquid chromatograph, the method comprising:

[0007] Based on the pressure values ​​of the infusion pump in the liquid chromatograph during the current pump cycle and a preset number of pump cycles prior to the current pump cycle, the parameter values ​​to be used for each specified pressure dimension are determined; wherein, each specified pressure dimension includes: a dimension characterizing the magnitude of the pressure value and / or a dimension characterizing the fluctuation state of the pressure value.

[0008] Based on the membership function corresponding to each specified pressure dimension that is set in advance, the membership value corresponding to the parameter value to be used in the specified pressure dimension is determined and used as the membership value to be detected.

[0009] Based on the determined membership values ​​to be detected, a pre-constructed fuzzy inference model is used to determine the probability that the operating state represented by the parameter value to be used belongs to each preset operating state; wherein, the preset operating states include: normal operating state and multiple fault operating states; the fuzzy inference model is constructed based on the reference membership values ​​corresponding to the parameter values ​​of each specified pressure dimension under each preset operating state; the reference membership value corresponding to the parameter value of a specified pressure dimension under a preset operating state is determined according to the membership function corresponding to the specified pressure dimension and the parameter value of the specified pressure dimension under the preset operating state;

[0010] The preset operating state with the highest probability is determined as the fault diagnosis result of the liquid chromatograph in the current pump cycle.

[0011] Optionally, the fuzzy inference model is constructed in the following manner:

[0012] For each preset operating state, based on the pressure value of the infusion pump in that preset operating state during a reference pump cycle and a preset number of pump cycles before the reference pump cycle, a reference parameter value for each specified pressure dimension is determined.

[0013] Based on the membership function corresponding to each specified pressure dimension, the membership value corresponding to the reference parameter value of the specified pressure dimension is determined and used as the reference membership value; wherein, the membership value corresponding to the parameter value of any specified pressure dimension is greater than -1 and less than 1;

[0014] Using the probability of belonging to the preset operating state as the dependent variable and the membership degree corresponding to the parameter values ​​of each specified pressure dimension as the independent variable, a fuzzy inference model corresponding to the preset operating state is constructed. In the fuzzy inference model corresponding to the preset operating state, the probability of belonging to the preset operating state is positively correlated with the membership degree represented by the reference membership degree value not less than the first membership degree threshold, positively correlated with the membership degree represented by the reference membership degree value less than 0 and greater than the second membership degree threshold, negatively correlated with the membership degree represented by the reference membership degree value greater than 0 and less than the first membership degree threshold, and negatively correlated with the membership degree represented by the reference membership degree value not greater than the second membership degree threshold.

[0015] The step of combining the determined membership values ​​to be detected and using a pre-constructed fuzzy inference model to determine the probability that the operating state represented by the parameter value to be used belongs to each preset operating state includes:

[0016] For each preset operating state, the determined membership values ​​to be detected are substituted into the fuzzy inference model corresponding to the preset operating state to obtain the probability that the operating state represented by the pressure parameter to be utilized belongs to the preset operating state.

[0017] Optionally, the infusion pump includes a primary pump and a accumulator pump connected in series; each pump cycle includes: a single-pump infusion phase and a dual-pump synergistic phase.

[0018] Each specified stress dimension includes any of the following:

[0019] The pressure dimension represents the average pressure level of the accumulator pump during the single-pump delivery phase of each pump cycle; the pressure fluctuation dimension represents whether the pressure value of the accumulator pump fluctuates during the single-pump delivery phase of each pump cycle; the first pressure change dimension represents the trend of the pressure fluctuation amplitude of the accumulator pump during the single-pump delivery phase of each pump cycle; the maximum pressure dimension represents the maximum pressure value of the primary pump during the dual-pump synergy phase of each pump cycle; and the second pressure change dimension represents the trend of the pressure value of the primary pump during the dual-pump synergy phase of adjacent pump cycles.

[0020] Optionally, the membership degree corresponding to the parameter value of the average pressure dimension represents the probability that the pressure value of the accumulator pump in the single pump infusion phase of each pump cycle belongs to no pressure.

[0021] The membership degree corresponding to the parameter value of the pressure fluctuation dimension represents the probability that the pressure value of the accumulator pump fluctuates during the single-pump infusion phase of each pump cycle.

[0022] The membership degree corresponding to the parameter value of the first pressure change dimension indicates that the probability of the pressure fluctuation amplitude of the accumulator pump decreasing during the single pump delivery phase of each pump cycle is:

[0023] The membership degree corresponding to the parameter value of the maximum pressure dimension represents the probability that the pressure value of the primary pump in the dual-pump collaborative phase of each pump cycle is pressureless.

[0024] The membership degree corresponding to the parameter value of the second pressure change dimension indicates the probability that the pressure value of the primary pump in the dual-pump collaborative phase of adjacent pump cycles is decreasing.

[0025] Optionally, the step of determining the membership value corresponding to the parameter value to be used in the specified pressure dimension based on the membership function corresponding to each specified pressure dimension in a pre-set manner, and using it as the membership value to be detected, includes:

[0026] If the value of the parameter to be used in the average pressure dimension is not less than the first pressure value, then the membership value to be detected corresponding to the average pressure dimension is determined to be a first value; the first pressure value is the product of a first coefficient and the system pressure value; the first coefficient is less than 1; the system pressure value is the pressure value of the accumulator pump in the current pump cycle under the most recent normal operating condition; if the value of the parameter to be used in the average pressure dimension is less than the first pressure value but greater than the second pressure value, then the membership value to be detected corresponding to the average pressure dimension is determined based on the second value; the second pressure value is the product of a second coefficient and the system pressure value; if the value of the parameter to be used in the average pressure dimension is less than the second pressure value, then the membership value to be detected corresponding to the average pressure dimension is determined to be a third value.

[0027] And / or,

[0028] If the parameter value to be used in the pressure fluctuation dimension indicates that the pressure value of the accumulator pump fluctuates in the single-pump infusion stage of each pump cycle, then the membership value to be detected corresponding to the pressure fluctuation dimension is determined based on the fourth value; if the parameter value to be used in the pressure fluctuation dimension indicates that the pressure value of the accumulator pump does not fluctuate in the single-pump infusion stage of each pump cycle, then the membership value to be detected corresponding to the pressure fluctuation dimension is determined to be the fifth value.

[0029] And / or,

[0030] If the proportion of the first pump cycle in each pump cycle is not less than a preset proportion, then the membership value to be detected corresponding to the first pressure change dimension is determined to be the sixth value; the degree to which the pressure fluctuation amplitude of the accumulator pump decreases in the single-pump infusion stage of the first pump cycle relative to the previous pump cycle is greater than the second pressure value; if the proportion of the second pump cycle in each pump cycle is not less than a preset proportion, then the membership value to be detected corresponding to the first pressure change dimension is determined to be the seventh value; the degree to which the pressure fluctuation amplitude of the accumulator pump increases in the single-pump infusion stage of the second pump cycle relative to the previous pump cycle is greater than the second pressure value; if the proportions of both the first pump cycle and the second pump cycle in each pump cycle are less than a preset proportion, then the membership value to be detected corresponding to the first pressure change dimension is determined to be the eighth value.

[0031] And / or,

[0032] If the value of the parameter to be used in the maximum pressure dimension is not less than the system pressure value, then the membership value to be detected corresponding to the maximum pressure dimension is determined to be the ninth value; if the value of the parameter to be used in the maximum pressure dimension is less than the system pressure value but greater than the second pressure value, then the membership value to be detected corresponding to the maximum pressure dimension is determined based on the tenth value; if the value of the parameter to be used in the maximum pressure dimension is less than the second pressure value, then the membership value to be detected corresponding to the maximum pressure dimension is determined to be the eleventh value.

[0033] And / or,

[0034] If the proportion of the third pump cycle in each pump cycle is not less than a preset proportion, then the membership degree value to be detected corresponding to the second pressure change dimension is determined to be the twelfth value; the pressure value of the primary pump in the dual-pump collaborative stage of the third pump cycle relative to the previous pump cycle decreases by a greater degree than the second pressure value; if the proportion of the fourth pump cycle in each pump cycle is not less than a preset proportion, then the membership degree value to be detected corresponding to the second pressure change dimension is determined to be the thirteenth value; the pressure value of the primary pump in the dual-pump collaborative stage of the third pump cycle relative to the previous pump cycle increases by a greater degree than the second pressure value; if the proportions of both the third and fourth pump cycles in each pump cycle are less than a preset proportion, then the membership degree value to be detected corresponding to the second pressure change dimension is determined to be the fourteenth value.

[0035] A second aspect of this application also provides a liquid chromatograph, the liquid chromatograph comprising: an infusion pump, a pressure sensor, and a status detection terminal;

[0036] The pressure sensor is used to measure the pressure of the infusion pump and send the measured pressure to the status detection terminal;

[0037] The status detection terminal is used to receive the pressure sent by the pressure sensor and execute any of the liquid chromatograph fault diagnosis methods described above.

[0038] Optionally, when the infusion pump includes a primary pump and a accumulator pump connected in series, the pressure sensor includes: a first pressure sensor deployed at the outlet of the primary pump for measuring the pressure value of the primary pump, and a second pressure sensor deployed at the outlet of the accumulator pump for measuring the pressure value of the accumulator pump.

[0039] A third aspect of this application also provides a liquid chromatograph fault diagnosis device, the device comprising:

[0040] The parameter value determination module is used to determine the parameter values ​​to be used for each specified pressure dimension based on the pressure values ​​of the infusion pump in the liquid chromatograph during the current pump cycle and a preset number of pump cycles prior to the current pump cycle; wherein, each specified pressure dimension includes: a dimension characterizing the magnitude of the pressure value and / or a dimension characterizing the fluctuation state of the pressure value;

[0041] The membership value determination module is used to determine the membership value corresponding to the parameter value to be used in the specified pressure dimension according to the membership function corresponding to each specified pressure dimension that is preset, and use it as the membership value to be detected.

[0042] The probability determination module is used to combine the determined membership values ​​to be detected with a pre-built fuzzy inference model to determine the probability that the operating state represented by the parameter value to be used belongs to each preset operating state; wherein, the preset operating states include: normal operating state and multiple fault operating states; the fuzzy inference model is constructed based on the reference membership values ​​corresponding to the parameter values ​​of each specified pressure dimension in each preset operating state; the reference membership value corresponding to the parameter value of a specified pressure dimension in a preset operating state is determined according to the membership function corresponding to the specified pressure dimension and the parameter value of the specified pressure dimension in the preset operating state;

[0043] The diagnostic result determination module is used to determine the preset operating state with the highest probability as the fault diagnosis result of the liquid chromatograph in the current pump cycle.

[0044] Optionally, the fuzzy inference model is constructed in the following manner:

[0045] For each preset operating state, based on the pressure value of the infusion pump in that preset operating state during a reference pump cycle and a preset number of pump cycles before the reference pump cycle, a reference parameter value for each specified pressure dimension is determined.

[0046] Based on the membership function corresponding to each specified pressure dimension, the membership value corresponding to the reference parameter value of the specified pressure dimension is determined and used as the reference membership value; wherein, the membership value corresponding to the parameter value of any specified pressure dimension is greater than -1 and less than 1;

[0047] Using the probability of belonging to the preset operating state as the dependent variable and the membership degree corresponding to the parameter values ​​of each specified pressure dimension as the independent variable, a fuzzy inference model corresponding to the preset operating state is constructed. In the fuzzy inference model corresponding to the preset operating state, the probability of belonging to the preset operating state is positively correlated with the membership degree represented by the reference membership degree value not less than the first membership degree threshold, positively correlated with the membership degree represented by the reference membership degree value less than 0 and greater than the second membership degree threshold, negatively correlated with the membership degree represented by the reference membership degree value greater than 0 and less than the first membership degree threshold, and negatively correlated with the membership degree represented by the reference membership degree value not greater than the second membership degree threshold.

[0048] The probability determination module is specifically used to, for each preset operating state, substitute the determined membership values ​​to be detected into the fuzzy inference model corresponding to the preset operating state to obtain the probability that the operating state represented by the pressure parameter to be utilized belongs to the preset operating state.

[0049] Optionally, the infusion pump includes a primary pump and a accumulator pump connected in series; each pump cycle includes: a single-pump infusion phase and a dual-pump synergistic phase.

[0050] Each specified stress dimension includes any of the following:

[0051] The pressure dimension represents the average pressure level of the accumulator pump during the single-pump delivery phase of each pump cycle; the pressure fluctuation dimension represents whether the pressure value of the accumulator pump fluctuates during the single-pump delivery phase of each pump cycle; the first pressure change dimension represents the trend of the pressure fluctuation amplitude of the accumulator pump during the single-pump delivery phase of each pump cycle; the maximum pressure dimension represents the maximum pressure value of the primary pump during the dual-pump synergy phase of each pump cycle; and the second pressure change dimension represents the trend of the pressure value of the primary pump during the dual-pump synergy phase of adjacent pump cycles.

[0052] Optionally, the membership degree corresponding to the parameter value of the average pressure dimension represents the probability that the pressure value of the accumulator pump in the single pump infusion phase of each pump cycle belongs to no pressure.

[0053] The membership degree corresponding to the parameter value of the pressure fluctuation dimension represents the probability that the pressure value of the accumulator pump fluctuates during the single-pump infusion phase of each pump cycle.

[0054] The membership degree corresponding to the parameter value of the first pressure change dimension indicates that the probability of the pressure fluctuation amplitude of the accumulator pump decreasing during the single pump delivery phase of each pump cycle is:

[0055] The membership degree corresponding to the parameter value of the maximum pressure dimension represents the probability that the pressure value of the primary pump in the dual-pump collaborative phase of each pump cycle is pressureless.

[0056] The membership degree corresponding to the parameter value of the second pressure change dimension indicates the probability that the pressure value of the primary pump in the dual-pump collaborative phase of adjacent pump cycles is decreasing.

[0057] Optionally, the membership value determination module is specifically used to determine the membership value to be detected for the average pressure dimension as a first value if the usable parameter value of the average pressure dimension is not less than a first pressure value; the first pressure value is the product of a first coefficient and a system pressure value; the first coefficient is less than 1; the system pressure value is the pressure value of the accumulator pump in the current pump cycle under the most recent normal operating condition; if the usable parameter value of the average pressure dimension is less than the first pressure value and greater than a second pressure value, the membership value to be detected for the average pressure dimension is determined based on a second value; the second pressure value is the product of a second coefficient and a system pressure value; if the usable parameter value of the average pressure dimension is less than the second pressure value, the membership value to be detected for the average pressure dimension is determined as a third value.

[0058] And / or,

[0059] If the parameter value to be used in the pressure fluctuation dimension indicates that the pressure value of the accumulator pump fluctuates in the single-pump infusion stage of each pump cycle, then the membership value to be detected corresponding to the pressure fluctuation dimension is determined based on the fourth value; if the parameter value to be used in the pressure fluctuation dimension indicates that the pressure value of the accumulator pump does not fluctuate in the single-pump infusion stage of each pump cycle, then the membership value to be detected corresponding to the pressure fluctuation dimension is determined to be the fifth value.

[0060] And / or,

[0061] If the proportion of the first pump cycle in each pump cycle is not less than a preset proportion, then the membership value to be detected corresponding to the first pressure change dimension is determined to be the sixth value; the degree to which the pressure fluctuation amplitude of the accumulator pump decreases in the single-pump infusion stage of the first pump cycle relative to the previous pump cycle is greater than the second pressure value; if the proportion of the second pump cycle in each pump cycle is not less than a preset proportion, then the membership value to be detected corresponding to the first pressure change dimension is determined to be the seventh value; the degree to which the pressure fluctuation amplitude of the accumulator pump increases in the single-pump infusion stage of the second pump cycle relative to the previous pump cycle is greater than the second pressure value; if the proportions of both the first pump cycle and the second pump cycle in each pump cycle are less than a preset proportion, then the membership value to be detected corresponding to the first pressure change dimension is determined to be the eighth value.

[0062] And / or,

[0063] If the value of the parameter to be used in the maximum pressure dimension is not less than the system pressure value, then the membership value to be detected corresponding to the maximum pressure dimension is determined to be the ninth value; if the value of the parameter to be used in the maximum pressure dimension is less than the system pressure value but greater than the second pressure value, then the membership value to be detected corresponding to the maximum pressure dimension is determined based on the tenth value; if the value of the parameter to be used in the maximum pressure dimension is less than the second pressure value, then the membership value to be detected corresponding to the maximum pressure dimension is determined to be the eleventh value.

[0064] And / or,

[0065] If the proportion of the third pump cycle in each pump cycle is not less than a preset proportion, then the membership degree value to be detected corresponding to the second pressure change dimension is determined to be the twelfth value; the pressure value of the primary pump in the dual-pump collaborative stage of the third pump cycle relative to the previous pump cycle decreases by a greater degree than the second pressure value; if the proportion of the fourth pump cycle in each pump cycle is not less than a preset proportion, then the membership degree value to be detected corresponding to the second pressure change dimension is determined to be the thirteenth value; the pressure value of the primary pump in the dual-pump collaborative stage of the third pump cycle relative to the previous pump cycle increases by a greater degree than the second pressure value; if the proportions of both the third and fourth pump cycles in each pump cycle are less than a preset proportion, then the membership degree value to be detected corresponding to the second pressure change dimension is determined to be the fourteenth value.

[0066] A fourth aspect of this application also provides an electronic device, comprising:

[0067] Memory, used to store computer programs;

[0068] The processor, when executing a program stored in memory, implements any of the above-described liquid chromatograph fault diagnosis methods.

[0069] A fifth aspect of this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements any of the above-described liquid chromatograph fault diagnosis methods.

[0070] This application also provides a computer program product containing instructions that, when run on a computer, cause the computer to execute any of the liquid chromatograph fault diagnosis methods described above.

[0071] Beneficial effects of the embodiments in this application:

[0072] This application provides a method for diagnosing faults in a liquid chromatograph. Based on the pressure values ​​of the infusion pump in the liquid chromatograph during the current pump cycle and a preset number of pump cycles prior to the current pump cycle, the method determines the parameter values ​​to be utilized for each specified pressure dimension. Each specified pressure dimension includes: a dimension representing the magnitude of the pressure value and / or a dimension representing the fluctuation state of the pressure value. According to a pre-set membership function corresponding to each specified pressure dimension, the method determines the membership degree value corresponding to the parameter value to be utilized for that specified pressure dimension, which is then used as the membership degree value to be detected. Combining the determined membership degree values ​​to be detected, a pre-constructed fuzzy inference model is used... The model determines the probability that the operating state represented by the parameter value to be used belongs to each preset operating state; wherein, the preset operating states include: normal operating state and multiple fault operating states; the fuzzy inference model is constructed based on the reference membership values ​​corresponding to the parameter values ​​of each specified pressure dimension under each preset operating state; the reference membership value corresponding to the parameter value of a specified pressure dimension under a preset operating state is determined according to the membership function corresponding to the specified pressure dimension and the parameter value of the specified pressure dimension under the preset operating state; the preset operating state with the highest probability is determined as the fault diagnosis result of the liquid chromatograph in the current pump cycle.

[0073] Based on the above processing, the parameter values ​​(i.e., the parameters to be used) for each specified pressure dimension can be determined according to the pump cycle in the liquid chromatograph that needs to be tested (i.e., the current pump cycle) and the pressure values ​​of a preset number of pump cycles prior to the current pump cycle. Each specified pressure dimension includes: a dimension characterizing the magnitude of the pressure value and / or a dimension characterizing the fluctuation state of the pressure value. Accordingly, each parameter value to be used can characterize the recent pressure state of the pump in the liquid chromatograph from different dimensions. Furthermore, a membership function corresponding to each specified pressure dimension is pre-set. By combining the corresponding membership function, the membership value (i.e., the membership value to be detected) corresponding to each specified pressure dimension can be determined. Accordingly, each determined membership value to be detected can represent the recent pressure state of the pump in the liquid chromatograph.

[0074] For each preset operating state, a reference membership value corresponding to the parameter value of a specified pressure dimension in that preset operating state can be determined in advance based on the membership function corresponding to each specified pressure dimension and the parameter value of that specified pressure dimension in that preset operating state. Then, based on the reference membership values ​​corresponding to the parameter values ​​of each specified pressure dimension in each preset operating state, a fuzzy inference model can be constructed. Correspondingly, the fuzzy inference model can represent the fuzzy relationship between the membership values ​​corresponding to the parameter values ​​of each specified pressure dimension and the operating state. Combining each membership value to be detected and the fuzzy inference model, the probability that the operating state represented by the parameter value to be used corresponding to each membership value to be detected belongs to each preset operating state can be determined based on the similarity between each membership value to be detected and the reference membership value in each preset operating state in the fuzzy inference model.

[0075] The preset operating state with the highest probability can represent the operating state of the liquid chromatograph in the current pump cycle. Therefore, the preset operating state with the highest probability can be determined as the fault diagnosis result of the liquid chromatograph in the current pump cycle. The preset operating states include: normal operating state and various fault operating states. Correspondingly, the determined fault diagnosis result can indicate whether the liquid chromatograph is in a faulty operating state. In this way, by combining the recent pressure status of the infusion pump in the liquid chromatograph and using fuzzy reasoning, the fault diagnosis result of the liquid chromatograph in the current pump cycle can be determined in a timely manner. That is, it is possible to detect whether the liquid chromatograph is in a faulty operating state in a timely manner, and thus, to detect the liquid chromatograph malfunction in a timely manner.

[0076] Of course, implementing any product or method of this application does not necessarily require achieving all of the advantages described above at the same time. Attached Figure Description

[0077] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other embodiments can be obtained based on these drawings.

[0078] Figure 1 A schematic flowchart of a first method for diagnosing faults in a liquid chromatograph provided in an embodiment of this application;

[0079] Figure 2 This is a schematic diagram of the structure of an infusion pump provided in an embodiment of this application;

[0080] Figure 3a A pressure curve of an infusion pump under normal operating conditions is provided in an embodiment of this application;

[0081] Figure 3bA pressure curve of an infusion pump under a first fault condition is provided in an embodiment of this application;

[0082] Figure 3c A pressure curve of an infusion pump under a second fault condition is provided in an embodiment of this application;

[0083] Figure 3d A pressure curve of an infusion pump under a third fault condition is provided in an embodiment of this application;

[0084] Figure 3e A pressure curve of an infusion pump under a fourth fault condition is provided in an embodiment of this application;

[0085] Figure 3f A pressure curve of an infusion pump under a fifth fault condition is provided in an embodiment of this application;

[0086] Figure 4 A flowchart illustrating the construction of a fuzzy inference model is provided for an embodiment of this application;

[0087] Figure 5 A schematic diagram of a second process for a liquid chromatograph fault diagnosis method provided in an embodiment of this application;

[0088] Figure 6 This is a schematic diagram of the structure of a liquid chromatograph fault diagnosis device provided in an embodiment of this application;

[0089] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0090] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art based on this application are within the scope of protection of this application.

[0091] Liquid chromatographs (LCs) are sophisticated analytical instruments widely used in sample testing. However, in practical applications, LCs may malfunction, affecting operational efficiency and the reliability of sample testing.

[0092] In the event of certain malfunctions, such as overpressure, motor step loss, or leakage, the liquid chromatograph (LC) can automatically report an error and shut down to prevent continued operation and potential damage to other components. However, in cases of other malfunctions, such as worn seals, malfunctioning check valves, or depleted mobile phase, the LC will not automatically report an error and will continue operating according to the set program, potentially leading to damage to other components.

[0093] To enable timely detection of HPLC malfunctions, this application provides a HPLC malfunction diagnosis method, see [link to relevant documentation]. Figure 1 , Figure 1 This is a schematic flowchart of a first method for diagnosing a fault in a liquid chromatograph provided in this application. The method for diagnosing a fault in a liquid chromatograph may include:

[0094] Step S101: Based on the pressure values ​​of the infusion pump in the liquid chromatograph during the current pump cycle and a preset number of pump cycles prior to the current pump cycle, determine the parameter values ​​to be used for each specified pressure dimension.

[0095] Each specified pressure dimension includes: a dimension that characterizes the magnitude of the pressure value and / or a dimension that characterizes the fluctuation state of the pressure value.

[0096] Step S102: Based on the membership function corresponding to each specified pressure dimension that is preset, determine the membership value corresponding to the parameter value to be used in the specified pressure dimension, and use it as the membership value to be detected.

[0097] Step S103: Combining the determined membership values ​​to be detected, use the pre-built fuzzy inference model to determine the probability that the operating state represented by the parameter value to be used belongs to each preset operating state.

[0098] The preset operating states include: normal operating state and multiple fault operating states; the fuzzy inference model is constructed based on the reference membership values ​​corresponding to the parameter values ​​of each specified pressure dimension in each preset operating state; the reference membership value corresponding to the parameter value of a specified pressure dimension in a preset operating state is determined according to the membership function corresponding to the specified pressure dimension and the parameter value of the specified pressure dimension in the preset operating state.

[0099] Step S104: The preset operating state with the highest probability is determined as the fault diagnosis result of the liquid chromatograph in the current pump cycle.

[0100] Based on the above processing, the parameter values ​​(i.e., the parameters to be used) for each specified pressure dimension can be determined according to the pump cycle in the liquid chromatograph that needs to be tested (i.e., the current pump cycle) and the pressure values ​​of a preset number of pump cycles prior to the current pump cycle. Each specified pressure dimension includes: a dimension characterizing the magnitude of the pressure value and / or a dimension characterizing the fluctuation state of the pressure value. Accordingly, each parameter value to be used can characterize the recent pressure state of the pump in the liquid chromatograph from different dimensions. Furthermore, a membership function corresponding to each specified pressure dimension is pre-set. By combining the corresponding membership function, the membership value (i.e., the membership value to be detected) corresponding to each specified pressure dimension can be determined. Accordingly, each determined membership value to be detected can represent the recent pressure state of the pump in the liquid chromatograph.

[0101] For each preset operating state, a reference membership value corresponding to the parameter value of a specified pressure dimension in that preset operating state can be determined in advance based on the membership function corresponding to each specified pressure dimension and the parameter value of that specified pressure dimension in that preset operating state. Then, based on the reference membership values ​​corresponding to the parameter values ​​of each specified pressure dimension in each preset operating state, a fuzzy inference model can be constructed. Correspondingly, the fuzzy inference model can represent the fuzzy relationship between the membership values ​​corresponding to the parameter values ​​of each specified pressure dimension and the operating state. Combining each membership value to be detected and the fuzzy inference model, the probability that the operating state represented by the parameter value to be used corresponding to each membership value to be detected belongs to each preset operating state can be determined based on the similarity between each membership value to be detected and the reference membership value in each preset operating state in the fuzzy inference model.

[0102] The preset operating state with the highest probability can represent the operating state of the liquid chromatograph in the current pump cycle. Therefore, the preset operating state with the highest probability can be determined as the fault diagnosis result of the liquid chromatograph in the current pump cycle. The preset operating states include: normal operating state and various fault operating states. Correspondingly, the determined fault diagnosis result can indicate whether the liquid chromatograph is in a faulty operating state. In this way, by combining the recent pressure status of the infusion pump in the liquid chromatograph and using fuzzy reasoning, the fault diagnosis result of the liquid chromatograph in the current pump cycle can be determined in a timely manner. That is, it is possible to detect whether the liquid chromatograph is in a faulty operating state in a timely manner, and thus, to detect the liquid chromatograph malfunction in a timely manner.

[0103] Regarding step S101, during the operation of the liquid chromatograph, the infusion pump's infusion process consists of several pump cycles. The duration of one pump cycle is the time required for the infusion pump to complete one complete liquid delivery operation. There can be only one or more infusion pumps, and the type of infusion pump is not specifically limited. A pressure sensor for measuring the pressure value of the infusion pump can be deployed at the outlet of the infusion pump.

[0104] Taking an infusion pump consisting of a primary pump and a accumulator pump connected in series as an example, the primary pump and the accumulator pump can work together to deliver a stable flow of liquid (also known as a solvent flow). Each pump cycle includes a single-pump infusion phase and a dual-pump collaborative phase. In the single-pump infusion phase, the accumulator pump delivers liquid at a constant rate according to the user-set flow rate, while the primary pump simultaneously completes liquid aspiration. In the dual-pump collaborative phase, the accumulator pump begins aspiration, and while the primary pump replenishes the accumulator pump, the accumulator pump also delivers liquid. A pressure sensor (i.e., the first pressure sensor in subsequent embodiments) is deployed at the outlet of the primary pump to measure the pressure value of the primary pump, and a pressure sensor (i.e., the second pressure sensor in subsequent embodiments) is deployed at the outlet of the accumulator pump to measure the pressure value of the accumulator pump.

[0105] like Figure 2 As shown, Figure 2 This is a schematic diagram of an infusion pump provided in an embodiment of this application. The infusion pump 21 includes a primary pump 211 and a accumulator pump 212 connected in series. The primary pump 211 has a primary pump check valve 213 at its inlet and a first pressure sensor 214 at its outlet. The accumulator pump 212 has an accumulator pump check valve 215 at its inlet and a second pressure sensor 216 at its outlet. The primary pump 211 and the accumulator pump 212 are driven by motors 217 and 218, respectively. The primary pump 211 also has a proportional valve 219 at its inlet for adjusting the input solvents (also known as the mobile phase, i.e.,...). Figure 2 The proportions of A, B, C, and D in the liquid chromatograph. The infusion pump 21 delivers liquid to the load 22 (i.e., the chromatographic column in the liquid chromatograph), and the liquid output from the load 22 can enter the waste liquid tank 23.

[0106] The current pump cycle can be the pump cycle for which testing is currently required. For example, the current pump cycle can be the latest pump cycle, allowing for timely fault diagnosis of the liquid chromatograph. Since a liquid chromatograph needs to run for a period of time after startup to reach a stable state, the pressure of the infusion pump is often unstable during this period, which can interfere with subsequent fault diagnosis based on the infusion pump pressure status, leading to lower accuracy. Therefore, to ensure accurate fault diagnosis, testing should usually begin after the liquid chromatograph has been running stably. In other words, the current pump cycle is the pump cycle after the liquid chromatograph has stabilized. For example, by observing the waveform of the pressure change of the liquid chromatograph's infusion pump, it can be determined whether the pressure change of the liquid chromatograph's infusion pump has reached a stable and normal state (as in subsequent...). Figure 3a The diagram shows the pressure curve of the infusion pump under normal operating conditions.

[0107] This function can obtain the pressure values ​​of the infusion pump in the liquid chromatograph for the current pump cycle and a preset number of pump cycles prior to the current pump cycle. The preset number can be set as needed and is not specifically limited. For example, the preset number can be 4 or 5. Accordingly, the pressure values ​​obtained for each pump cycle can represent the recent pressure state of the infusion pump. For ease of description, the current pump cycle and the preset number of pump cycles prior to the current pump cycle can be referred to as the pump cycles to be utilized. Taking a preset number of 5 as an example, the current pump cycle can be the 11th pump cycle, and the corresponding pump cycles to be utilized can be the 11th, 10th, 9th, 8th, and 7th pump cycles; the current pump cycle can be the 12th pump cycle, and the corresponding pump cycles to be utilized can be the 12th, 11th, 10th, 9th, and 8th pump cycles.

[0108] By using the pressure values ​​obtained from each pump cycle to be utilized, the parameter values ​​to be utilized for each specified pressure dimension can be determined. Each specified pressure dimension can include: a dimension characterizing the magnitude of the pressure value and / or a dimension characterizing the fluctuation state of the pressure value. The magnitude of the pressure value can be represented by: the average level of the pressure values ​​for each pump cycle to be utilized or the maximum value of the pressure values ​​for each pump cycle to be utilized. The fluctuation state of the pressure value can be represented by: whether the pressure value fluctuates, the trend of the pressure value's amplitude, and the trend of the pressure value's change. Both the magnitude of the pressure value and the fluctuation state of the pressure value can reflect the pressure state; correspondingly, the parameter values ​​to be utilized for each specified pressure dimension can characterize the recent pressure state of the infusion pump in the liquid chromatograph from different dimensions.

[0109] In one implementation, where the infusion pump comprises a primary pump and a accumulator pump connected in series, and each pump cycle includes a single-pump infusion phase and a dual-pump synergistic phase, each specified pressure dimension may include any of the following:

[0110] The pressure dimension represents the average pressure level of the accumulator pump during the single-pump delivery phase of each pump cycle; the pressure fluctuation dimension represents whether the pressure value of the accumulator pump fluctuates during the single-pump delivery phase of each pump cycle; the first pressure change dimension represents the trend of the pressure fluctuation amplitude of the accumulator pump during the single-pump delivery phase of each pump cycle; the maximum pressure dimension represents the maximum pressure value of the primary pump during the dual-pump synergy phase of each pump cycle; and the second pressure change dimension represents the trend of the pressure value of the primary pump during the dual-pump synergy phase of adjacent pump cycles.

[0111] In this implementation, the specified pressure dimension can include any number of the following: average pressure dimension, pressure fluctuation dimension, first pressure change dimension, maximum pressure dimension, and second pressure change dimension.

[0112] For the average pressure dimension, the average pressure value of the accumulator pump during the single-pump delivery phase of each pump cycle to be utilized can be calculated to obtain the utilization parameter value of the average pressure dimension, which can be represented as P1.

[0113] Regarding the pressure fluctuation dimension, the pressure change of the accumulator pump within a unit time interval during the single-pump delivery phase of each pump cycle can be calculated, and it can be determined whether the obtained pressure change values ​​simultaneously contain both positive and negative values. If the pressure change values ​​obtained in a pump cycle simultaneously contain both positive and negative values, it indicates that the pressure value of that pump cycle fluctuates. Accordingly, it can be determined whether the pressure value of the accumulator pump fluctuates in the single-pump delivery phase of each pump cycle, obtaining the utilization parameter value in the pressure fluctuation dimension, which can be expressed as: P1. Furthermore, the magnitude of the pressure fluctuation can be determined based on the absolute value of the obtained pressure change. For example, the larger the absolute value of the obtained pressure change, the greater the pressure fluctuation.

[0114] For the first pressure change dimension, for each pump cycle to be utilized, the pressure fluctuation amplitude of the accumulator pump during the single-pump infusion phase of the previous pump cycle (which can be called the first amplitude) and the pressure fluctuation amplitude of the accumulator pump during the single-pump infusion phase of the current pump cycle to be utilized (which can be called the second amplitude) can be obtained. The difference between the first amplitude and the second amplitude can be calculated to obtain the utilization parameter value for the first pressure change dimension, which can be expressed as follows: P1'. Correspondingly, the difference obtained for each pump cycle can characterize the changing trend of the pressure fluctuation amplitude of the accumulator pump during the single-pump delivery phase of each pump cycle.

[0115] For the maximum pressure dimension, the maximum pressure value can be determined from the pressure values ​​of the primary pump during the dual-pump coordination phase of each pump cycle to be utilized. This maximum pressure value can be used as the parameter value to be utilized, and can be represented as P2.

[0116] Regarding the second pressure change dimension, for each pump cycle to be utilized, the maximum pressure value of the primary pump during the dual-pump infusion phase of the previous pump cycle (which can be called the first pressure value) and the maximum pressure value of the primary pump during the dual-pump infusion phase of the current pump cycle (which can be called the second pressure value) can be obtained. The difference between the first and second pressure values ​​can be calculated to obtain the parameter value to be utilized in the second pressure change dimension, which can be expressed as follows: P2. Accordingly, the difference obtained for each pump cycle to be utilized can characterize the pressure change trend of the primary pump during the dual-pump synergy phase in adjacent pump cycles.

[0117] This allows us to further ensure that the recent pressure state of the infusion pump in the liquid chromatograph can be characterized by parameter values ​​from multiple pressure dimensions. It also ensures that subsequent determination of the operating state represented by the pressure state, based on fuzzy reasoning, is possible. This, in turn, enables timely detection of whether the liquid chromatograph is in a faulty operating state, and thus timely detection of liquid chromatograph malfunctions.

[0118] Regarding step S102, a membership function corresponding to each specified pressure dimension can be pre-set. The membership function corresponding to each specified pressure dimension can represent the mapping relationship between the parameter values ​​and membership degrees of that specified pressure dimension. Accordingly, for each specified pressure dimension, the membership function corresponding to that specified pressure dimension can be used to determine the membership value corresponding to the parameter value to be used in that specified pressure dimension, which is then used as the membership value to be detected.

[0119] The membership degree corresponding to the parameter value of each specified pressure dimension can represent the probability that the pressure state represented by the parameter value of the specified pressure dimension belongs to the preset pressure state of the specified pressure dimension. For example, for a dimension representing the magnitude of the pressure value, the preset pressure state can be "normal", "depressurized", or "no pressure"; for a dimension representing the fluctuation state of the pressure value, such as a dimension representing whether the pressure value fluctuates, the preset pressure state can be "fluctuates" or "no fluctuation"; for a dimension representing the trend of the change in the magnitude of the pressure value or the trend of the pressure value change, the preset pressure state can be "gradually decreasing", "remaining unchanged", or "gradually increasing".

[0120] In one implementation, when the infusion pump includes a primary pump and an accumulator pump connected in series, the membership degree corresponding to the parameter value of the average pressure dimension can be represented as the probability that the pressure value of the accumulator pump in the single-pump infusion phase of each pump cycle belongs to the no-pressure category.

[0121] The membership degree corresponding to the parameter value of the pressure fluctuation dimension can represent the probability that the pressure value of the accumulator pump fluctuates during the single-pump infusion phase of each pump cycle.

[0122] The membership degree corresponding to the parameter value of the first pressure change dimension can represent the probability that the pressure fluctuation amplitude of the accumulator pump in the single pump delivery phase of each pump cycle is decreasing.

[0123] The membership degree corresponding to the parameter value of the maximum pressure dimension can represent the probability that the pressure value of the primary pump in the dual-pump synergy phase of each pump cycle belongs to no pressure.

[0124] The membership degree corresponding to the parameter value of the second pressure change dimension can represent the probability that the pressure value of the primary pump in the dual-pump collaborative phase of adjacent pump cycles is decreasing.

[0125] In this implementation, the membership degree corresponding to the parameter value of each specified pressure dimension can characterize the pressure change pattern of the infusion pump in each pump cycle from different pressure dimensions. Correspondingly, the membership degree value corresponding to the parameter value to be used in each specified pressure dimension (i.e., the membership degree value to be detected) can also effectively characterize the pressure change pattern of the infusion pump in each pump cycle to be used, that is, it can accurately characterize the recent pressure state of the infusion pump in the liquid chromatograph.

[0126] This allows for the timely determination of fault diagnosis results for the liquid chromatograph in the current pump cycle by combining the recent pressure status of the pump in the liquid chromatograph with fuzzy reasoning. In other words, it enables timely detection of whether the liquid chromatograph is operating in a faulty state, and thus timely detection of liquid chromatograph malfunctions.

[0127] For steps S103 and S104, preset operating states can be determined in advance. These preset operating states can include normal operating states and various fault operating states. A fault operating state can represent the operating state under a specific fault cause; correspondingly, the fault operating state can characterize the fault cause. The fault operating states in the preset operating states can include operating states under various common fault causes. For each preset operating state, the liquid chromatograph can be preset to that preset operating state, and the parameter values ​​(which can be called reference parameter values) of each specified pressure dimension under that preset operating state can be obtained. Based on the membership function corresponding to each specified pressure dimension and the reference parameter value of that specified pressure dimension under that preset operating state, the membership value (i.e., the reference membership value) corresponding to the reference parameter value of a specified pressure dimension under that preset operating state can be determined. For details, please refer to the relevant description of determining the corresponding membership value to be detected based on the parameter value to be utilized for each specified pressure dimension.

[0128] Furthermore, based on the reference membership values ​​corresponding to the reference parameter values ​​of each specified pressure dimension under each preset operating state, the mapping relationship between the reference membership values ​​of each specified pressure dimension and each preset operating state can be determined, thus constructing a fuzzy inference model. Correspondingly, the fuzzy inference model can characterize the fuzzy relationship between the membership values ​​corresponding to the parameter values ​​of each specified pressure dimension and the operating state.

[0129] By utilizing the constructed fuzzy inference model and combining each membership value to be detected, the probability that the operating state represented by the parameter value to be utilized corresponding to each membership value to be detected belongs to each preset operating state can be determined based on the similarity between each membership value to be detected and the reference membership value under each preset operating state in the fuzzy inference model. The higher the similarity between each membership value to be detected and the reference membership value under a preset operating state, the closer the pressure state of the infusion pump under the operating state represented by the parameter value to be utilized corresponding to each membership value to is to the pressure state of the infusion pump under that preset operating state. Correspondingly, the probability that the operating state represented by the parameter value to be utilized corresponding to each membership value to be detected belongs to that preset operating state is also greater.

[0130] Therefore, the preset operating state with the highest probability can represent the operating state of the liquid chromatograph in the current pump cycle. Thus, the preset operating state with the highest probability can be determined as the fault diagnosis result of the liquid chromatograph in the current pump cycle. If the fault diagnosis result of the liquid chromatograph in the current pump cycle is a normal operating state, it means that the liquid chromatograph is operating normally and no fault has occurred, and the liquid chromatograph can continue to operate. If the fault diagnosis result of the liquid chromatograph in the current pump cycle is a fault operating state, it means that the liquid chromatograph is abnormal and a fault has occurred, and an alarm can be issued to prompt the user to handle the fault in a timely manner.

[0131] Furthermore, the causes of the faults represented by the fault operating states in the preset operating states are all predetermined. That is, the causes of the faults represented by the fault operating states are all known. If it is determined that the liquid chromatograph is in a fault operating state, the cause of the liquid chromatograph fault can be determined at the same time. This allows for timely detection of whether the liquid chromatograph has a fault and timely determination of the cause of the fault, which facilitates subsequent fault handling by users and can further improve the efficiency of users and after-sales personnel in troubleshooting instrument faults.

[0132] Furthermore, the liquid chromatograph fault diagnosis method provided in this application embodiment can achieve online real-time diagnosis of various fault states solely through the pressure data of the infusion pump combined with a pre-built fuzzy inference model during the detection process. It only requires the existing pressure sensor on the infusion pump, without the need for additional sensors, resulting in low application costs. Moreover, the liquid chromatograph fault diagnosis method provided in this application embodiment does not consume enormous computing resources or require pre-training, nor does it require additional offline fault diagnosis programs. It can be directly deployed on the liquid chromatograph, i.e., it can be directly deployed at the instrument end.

[0133] In one embodiment, where the infusion pump includes a primary pump and a accumulator pump connected in series, the cause of the failure characterized by the faulty operating state may include: using a single mobile phase (e.g., Figure 2When using only one solvent, the mobile phase runs out, the filter head becomes clogged, the primary pump check valve cannot be opened, or when using a multi-channel online mixed mobile phase (such as...). Figure 2 The fault conditions are as follows: (1) When using two or more solvents mixed together, the mobile phase in one stream runs out, the primary pump check valve cannot close, a small amount of air bubbles enter the mobile phase, or the primary pump seal or accumulator pump seal is worn. When using a single mobile phase, the fault condition where the mobile phase runs out, the filter head is clogged, and the primary pump check valve cannot open (this can be called fault cause 1) can be called the first fault condition. When using multiple online mixed mobile phases, the fault condition where one mobile phase runs out and the primary pump check valve cannot close (this can be called fault cause 2) can be called the second fault condition. The fault condition where a small amount of air bubbles enter the mobile phase (this can be called fault cause 3) can be called the third fault condition. The fault condition where the primary pump seal is worn (this can be called fault cause 4) can be called the fourth fault condition. The fault condition where the accumulator pump seal is worn (this can be called fault cause 5) can be called the fifth fault condition.

[0134] Figures 3a-3f This is a graph showing the pressure curves of the infusion pump under different operating conditions of the liquid chromatograph, i.e., how the pressure value of the infusion pump changes over time. The solid line represents the pressure value of the accumulator pump (also known as the accumulator pump pressure), and the dashed line represents the pressure value of the primary pump (also known as the primary pump pressure).

[0135] When the liquid chromatograph is in normal operating condition, such as Figure 3a As shown, the pressure of the accumulator pump is stable and without fluctuation in each pump cycle, while the pressure of the primary pump is 0 during the single-pump infusion stage and pulsed during the dual-pump synergy stage.

[0136] When the liquid chromatograph is in its first fault state, during primary pump aspiration, due to factors such as the mobile phase running out when using a single mobile phase, filter blockage, or the primary pump's check valve failing to open, almost no liquid is drawn into the primary pump. Therefore, during the dual-pump coordination phase, the primary pump pressure is 0. In the initial stage of the first fault state, a small amount of liquid may remain in the primary pump, enough to replenish the accumulator pump. At this time, the accumulator pump pressure decreases but still maintains some pressure. As the pump cycle continues, almost no liquid is replenished to the accumulator pump, and the accumulator pump pressure decreases further until it reaches zero. Figure 3b As shown, the first half of the liquid chromatograph is in normal operation, while the second half is in the first fault state. When the liquid chromatograph is in the first fault state, the accumulator pump pressure drops, and the accumulator pump pressure decreases further. The primary pump has no pressure, and after several cycles, both the accumulator pump pressure and the primary pump pressure approach 0.

[0137] When the liquid chromatograph is in its second fault state, and multiple mobile phases are mixed online, one mobile phase may run out, resulting in the primary pump only drawing in a portion of the liquid. When the primary pump's check valve cannot close, the primary pump can draw in liquid normally, but during the dual-pump co-operation phase, due to the primary pump's check valve not closing, some liquid flows back from the primary pump, leaving only a portion to replenish the accumulator pump. Both of these situations result in only a portion of liquid in the primary pump during the dual-pump co-operation phase, preventing it from pressurizing (i.e., the primary pump pressure is almost zero) and failing to adequately replenish the accumulator pump, causing its pressure to drop to almost zero. In the subsequent single-pump delivery phase, because there is some liquid in the accumulator pump, its pressure gradually begins to rise during delivery. Therefore, the accumulator pump pressure fluctuates significantly with the pump cycle, but the fluctuation range remains relatively constant. Figure 3c As shown, when the liquid chromatograph is in the second fault state, the pressure of the accumulator pump fluctuates significantly, and the fluctuation range remains basically unchanged, while the primary pump has no pressure.

[0138] When the liquid chromatograph is in the third fault state, a small amount of air bubbles are drawn into the primary pump from the mobile phase line during primary pump aspiration. At the start of the dual-pump synergy phase, due to the presence of air bubbles in the mobile phase within the primary pump, the primary pump pressure cannot increase to the accumulator pump pressure of the previous pump cycle. Therefore, both the primary pump pressure and the accumulator pump pressure decrease during the dual-pump synergy phase. Simultaneously, the air bubbles in the primary pump chamber are compressed into the liquid and discharged with it. Therefore, as the pump cycle increases, the air bubbles in the primary pump chamber become smaller, and the pressure drop during the dual-pump synergy phase also decreases. Figure 3d As shown, the accumulator pump pressure is normal during the single-pump infusion stage, but fluctuates during the dual-pump synergy stage, and the amplitude of this fluctuation gradually decreases. Furthermore, the primary pump pressure decreases during the dual-pump synergy stage, but gradually recovers as the pump cycle increases.

[0139] When the liquid chromatograph is in its fourth fault state, the accumulator pump pressure remains unaffected during the single-pump delivery phase. However, during the dual-pump synergy phase, the primary pump begins delivery. Due to wear on the primary pump's seals, the primary pump experiences pressure loss. Since the degree of wear does not change significantly within a certain timeframe, the amplitude of the pressure fluctuation remains essentially constant. Figure 3e As shown, the accumulator pump pressure is stable during the single-pump infusion stage, while the accumulator pump pressure fluctuates during the dual-pump synergy stage, and the amplitude of this pressure fluctuation remains basically constant. During the dual-pump synergy stage, the primary pump experiences pressure loss, but the maximum pressure of the primary pump remains unchanged.

[0140] When the liquid chromatograph is in the fifth fault state, the accumulator pump pressure is unstable and fluctuates during the single-pump dispensing phase due to some wear on the accumulator pump seal. The degree of seal wear is generally small, therefore the pressure fluctuation is usually small. Furthermore, since the degree of seal wear does not change significantly within a certain period, the amplitude of the pressure fluctuation remains relatively constant. However, during the dual-pump synergy phase, when the primary pump starts dispensing, the pressure is stable due to the good condition of the primary pump seal. Figure 3f As shown, the pressure of the accumulator pump is unstable and fluctuates during the single-pump infusion stage, while the pressure of the accumulator pump and the primary pump are normal during the dual-pump synergy stage.

[0141] It is evident that the infusion pump of the liquid chromatograph exhibits different pressure characteristics, i.e., different pressure states, under normal operation or different fault conditions. Based on the pressure states of the primary pump and accumulator pump under the aforementioned preset operating conditions, and with specified pressure dimensions including average pressure, pressure fluctuation, first pressure change, maximum pressure, and second pressure change, the correspondence between the pressure states represented by the parameter values ​​of each specified pressure dimension and the causes of the fault can be obtained, as shown in Table 1.

[0142] Table 1

[0143]

[0144] Taking the first row of Table 1 as an example, under normal operating conditions, without any fault cause, the parameter value of the average pressure dimension (i.e., P1) indicates that the pressure of the accumulator pump is normal during the single-pump infusion stage; the parameter value of the pressure fluctuation dimension (i.e., P1) indicates that the accumulator pump pressure is stable and without fluctuation during the single-pump infusion phase; since the accumulator pump pressure is stable during the single-pump infusion phase, the parameter value of the first pressure change dimension (i.e. P1') is not relevant; this item is "-" in the table. The parameter value of the maximum pressure dimension (i.e., P2) indicates that the primary pump pressure is normal during the dual-pump synergy phase. Normal primary pump pressure during the dual-pump synergy phase means that the maximum value of the primary pump pressure remains stable. Therefore, the parameter value of the second pressure change dimension (i.e., P2) is not relevant. P2) No need to pay attention to this; this item in the table is marked as "-".

[0145] In one embodiment, for each specified pressure dimension, the parameter values ​​of that specified pressure dimension can be divided into fuzzy sets, and the corresponding universe of discourse and membership function can be constructed.

[0146] Step S102 includes:

[0147] If the value of the parameter to be used in the average pressure dimension is not less than the first pressure value, then the membership value to be detected corresponding to the average pressure dimension is determined to be the first value; the first pressure value is the product of the first coefficient and the system pressure value; the first coefficient is less than 1; the system pressure value is the pressure value of the accumulator pump in the current pump cycle under the most recent normal operating condition; if the value of the parameter to be used in the average pressure dimension is less than the first pressure value but greater than the second pressure value, then the membership value to be detected corresponding to the average pressure dimension is determined based on the second value; the second pressure value is the product of the second coefficient and the system pressure value; if the value of the parameter to be used in the average pressure dimension is less than the second pressure value, then the membership value to be detected corresponding to the average pressure dimension is determined to be the third value.

[0148] And / or,

[0149] If the parameter value to be used in the pressure fluctuation dimension indicates that the pressure value of the accumulator pump fluctuates in the single-pump infusion stage of each pump cycle, then the membership value to be detected for the pressure fluctuation dimension is determined based on the fourth value; if the parameter value to be used in the pressure fluctuation dimension indicates that the pressure value of the accumulator pump does not fluctuate in the single-pump infusion stage of each pump cycle, then the membership value to be detected for the pressure fluctuation dimension is determined to be the fifth value.

[0150] And / or,

[0151] If the proportion of the first pump cycle in each pump cycle is not less than a preset proportion, then the membership value to be detected for the first pressure change dimension is determined to be the sixth value; if the pressure fluctuation amplitude of the accumulator pump in the single-pump infusion stage of the first pump cycle relative to the previous pump cycle decreases by a greater degree than the second pressure value; if the proportion of the second pump cycle in each pump cycle is not less than a preset proportion, then the membership value to be detected for the first pressure change dimension is determined to be the seventh value; if the pressure fluctuation amplitude of the accumulator pump in the single-pump infusion stage of the second pump cycle relative to the previous pump cycle increases by a greater degree than the second pressure value; if the proportions of both the first pump cycle and the second pump cycle in each pump cycle are less than a preset proportion, then the membership value to be detected for the first pressure change dimension is determined to be the eighth value.

[0152] And / or,

[0153] If the value of the parameter to be used in the maximum pressure dimension is not less than the system pressure value, then the membership value to be detected for the maximum pressure dimension is determined to be the ninth value; if the value of the parameter to be used in the maximum pressure dimension is less than the system pressure value but greater than the second pressure value, then the membership value to be detected for the maximum pressure dimension is determined based on the tenth value; if the value of the parameter to be used in the maximum pressure dimension is less than the second pressure value, then the membership value to be detected for the maximum pressure dimension is determined to be the eleventh value.

[0154] And / or,

[0155] If the proportion of the third pump cycle in each pump cycle is not less than the preset proportion, then the membership value to be detected for the second pressure change dimension is determined to be the twelfth value; the pressure value of the primary pump in the third pump cycle during the dual-pump synergy phase relative to the previous pump cycle decreases by a greater degree than the second pressure value; if the proportion of the fourth pump cycle in each pump cycle is not less than the preset proportion, then the membership value to be detected for the second pressure change dimension is determined to be the thirteenth value; the pressure value of the primary pump in the third pump cycle during the dual-pump synergy phase relative to the previous pump cycle increases by a greater degree than the second pressure value; if the proportions of both the third and fourth pump cycles in each pump cycle are less than the preset proportion, then the membership value to be detected for the second pressure change dimension is determined to be the fourteenth value.

[0156] In this embodiment of the application, during the fault diagnosis process of the liquid chromatograph, when the fault diagnosis result of the current pump cycle is normal operation, the pressure value of the accumulator pump in the current pump cycle can be recorded, that is, the pressure value of the accumulator pump in the latest pump cycle where the fault diagnosis result is normal operation can be recorded as the system pressure value. The first pressure value can be obtained by multiplying a first coefficient by the system pressure value. The first coefficient is less than 1; for example, the first coefficient can be 0.8 or 0.85 to ensure that the first pressure value is slightly less than the system pressure value. The second pressure value can be obtained by multiplying a second coefficient by the system pressure value. The second coefficient is less than the first coefficient; for example, the second coefficient can be 0.1 or 0.15 to ensure that the second pressure value is close to 0. The membership value to be detected corresponding to the parameter value to be utilized for each specified pressure dimension can also be simply referred to as the membership value to be detected corresponding to that specified pressure dimension.

[0157] For the average pressure dimension, a mapping relationship between parameter values ​​and membership values ​​can be constructed. The range of membership values ​​corresponding to the parameter values ​​of the average pressure dimension can be preset; for example, the range can be (0, 1), with the first endpoint being either 0 or 1, and the second endpoint being the other endpoint besides the first. Correspondingly, one endpoint of the usable parameter value of the average pressure dimension can be mapped to the first endpoint near the value range, the other endpoint to the second endpoint near the value range, and the median value of the usable parameter value to the midpoint near the value range (i.e., 0.5). This distinguishes the membership values ​​corresponding to different parameter values ​​of the average pressure dimension, further ensuring that the probability of the operating state represented by each usable parameter value belonging to each preset operating state can be accurately determined based on the similarity between each detected membership value and the reference membership value in each preset operating state in the fuzzy inference model. That is, it further ensures the accuracy of the fault diagnosis results.

[0158] If the value of the parameter to be utilized in the average pressure dimension is not less than the first pressure value, it indicates that the pressure of the accumulator pump is normal during the single-pump infusion phase of each pump cycle. In this case, the membership degree value to be detected corresponding to the average pressure dimension can be determined as the first value. For example, the first value can be close to 0, such as 0.1 or 0.2. If the value of the parameter to be utilized in the average pressure dimension is less than the first pressure value but greater than the second pressure value, it indicates that the pressure of the accumulator pump is lost during the single-pump infusion phase of each pump cycle. In this case, the membership degree value to be detected corresponding to the average pressure dimension can be determined based on the second value. For example, the second value can be directly used as the membership degree value to be detected corresponding to the average pressure dimension. Alternatively, the membership degree value to be detected can have a linear relationship with the parameter value of the average pressure dimension, such as a negative correlation. That is, the larger the value of the parameter to be utilized in the average pressure dimension, the smaller the membership degree value to be detected corresponding to the average pressure dimension, and the maximum membership degree value to be detected corresponding to the average pressure dimension is the second value. For example, the second value can be close to 0.5, such as 0.6 or 0.7. If the value of the parameter to be utilized in the average pressure dimension is less than the second pressure value, it indicates that the pressure of the accumulator pump in the single-pump infusion phase of each pump cycle is close to 0, i.e., there is no pressure. In this case, the membership value to be detected corresponding to the average pressure dimension is determined to be the third value. For example, the third value can be close to 1, such as 0.9 or 0.8.

[0159] The larger the membership value of the average pressure dimension, the greater the degree of pressure loss of the accumulator pump in the single-pump infusion stage of each pump cycle, and the greater the probability that the pressure value of the accumulator pump in the single-pump infusion stage of each pump cycle is pressureless.

[0160] For the pressure fluctuation dimension, a mapping relationship can be constructed between the parameter values ​​and membership values. The range of membership values ​​corresponding to the parameter values ​​of the pressure fluctuation dimension can be preset; for example, the range can be (0, 1), with the first endpoint being either 0 or 1, and the second endpoint being the other endpoint besides the first. Correspondingly, one endpoint of the parameter value to be used in the pressure fluctuation dimension can be mapped to the first endpoint closer to the value range, and the other endpoint can be mapped to the second endpoint closer to the value range.

[0161] If the parameter value to be utilized in the pressure fluctuation dimension represents the pressure value of the accumulator pump fluctuating during the single-pump infusion phase of each pump cycle—that is, the pressure change value of the accumulator pump within a unit time interval during the single-pump infusion phase of each pump cycle has both positive and negative values—indicating pressure fluctuation in the accumulator pump during the single-pump infusion phase of each pump cycle, then the membership value to be detected for the pressure fluctuation dimension is determined based on the fourth value. For example, the fourth value can be directly determined as the membership value to be detected for the pressure fluctuation dimension. Alternatively, the membership value corresponding to the pressure fluctuation dimension can have a linear relationship with the absolute value of the pressure change represented by the parameter value of the pressure fluctuation dimension, such as a positive correlation. The larger the absolute value of the pressure change represented by the parameter value to be utilized in the pressure fluctuation dimension, the larger the membership value to be detected for the pressure fluctuation dimension, and the maximum membership value to be detected for the pressure fluctuation dimension is the fourth value. For example, the fourth value can be close to 1, such as 0.9 or 0.8. The fourth value can be the same as the third value. If the parameter value to be utilized in the pressure fluctuation dimension indicates that the pressure value of the accumulator pump does not fluctuate uniformly during the single-pump infusion phase of each pump cycle, that is, the pressure change value of the accumulator pump within a unit time interval during the single-pump infusion phase of each pump cycle does not simultaneously have positive or negative values, it means that the pressure of the accumulator pump does not fluctuate during the single-pump infusion phase of each pump cycle to be utilized. Then, the membership value to be detected corresponding to the pressure fluctuation dimension is determined to be the fifth value. For example, the fifth value can be close to 0, such as 0.1 or 0.2. The fifth value can be the same as the first value.

[0162] The larger the membership value of the pressure fluctuation dimension, the greater the degree of pressure fluctuation of the accumulator pump in the single-pump infusion stage of each pump cycle, and the greater the probability that the pressure value of the accumulator pump fluctuates in the single-pump infusion stage of each pump cycle.

[0163] For each pump cycle to be utilized, if the decrease in pressure fluctuation amplitude of the accumulator pump in the single-pump delivery phase of the current pump cycle relative to the previous pump cycle is greater than the second pressure value (i.e., the difference between the pressure fluctuation amplitude of the accumulator pump in the single-pump delivery phase of the previous pump cycle and the pressure fluctuation amplitude in the single-pump delivery phase of the current pump cycle is greater than the second pressure value), it indicates that the pressure fluctuation amplitude of the accumulator pump in the single-pump delivery phase of the current pump cycle has decreased, and this pump cycle to be utilized is the first pump cycle. If the change in pressure fluctuation amplitude of the accumulator pump in the current pump cycle relative to the previous pump cycle is not greater than the second pressure value (i.e., the absolute value of the difference between the pressure fluctuation amplitude of the accumulator pump in the single-pump delivery phase of the previous pump cycle and the pressure fluctuation amplitude in the single-pump delivery phase of the current pump cycle is not greater than the second pressure value), it indicates that the pressure fluctuation amplitude of the accumulator pump in the single-pump delivery phase of the current pump cycle remains unchanged. If the pressure fluctuation amplitude of the accumulator pump in the current pump cycle increases more than the second pressure value in the single-pump delivery phase of the previous pump cycle, that is, the difference between the pressure fluctuation amplitude of the accumulator pump in the single-pump delivery phase of the previous pump cycle and the pressure fluctuation amplitude in the single-pump delivery phase of the current pump cycle is less than the negative of the second pressure value, it indicates that the pressure fluctuation amplitude of the accumulator pump in the single-pump delivery phase of the current pump cycle has increased, and then the current pump cycle is the second pump cycle.

[0164] For the first pressure change dimension, a mapping relationship can be constructed between the parameter values ​​and membership values ​​of the first pressure change dimension. The range of membership values ​​corresponding to the parameter values ​​of the first pressure change dimension can be preset; for example, the range can be (-1, 1), the first endpoint of the range can be -1 or 1, and the second endpoint can be the other endpoint besides the first endpoint. Correspondingly, one endpoint of the parameter value to be used in the first pressure change dimension can be mapped to the first endpoint closer to the value range, the other endpoint can be mapped to the second endpoint closer to the value range, and the median value of the parameter value to be used in the first pressure change dimension can be mapped to the midpoint (i.e., 0) closer to the value range.

[0165] If the proportion of the first pump cycle in each pump cycle to be utilized is not less than a preset proportion, it indicates that the pressure fluctuation amplitude of the accumulator pump in the single-pump infusion stage of each pump cycle to be utilized is gradually decreasing. Therefore, the membership value to be detected for the first pressure change dimension is determined to be the sixth value. For example, the sixth value can be close to 1, such as 0.9 or 0.8. The sixth value can be the same as the third value. For example, the preset proportion can be 40% or 60%. Taking 5 pump cycles to be utilized and a preset proportion of 40% as an example, at least 2 of the 5 pump cycles to be utilized are the first pump cycles. If the proportion of the second pump cycle in each pump cycle to be utilized is not less than a preset proportion, it indicates that the pressure fluctuation amplitude of the accumulator pump in the single-pump infusion stage of each pump cycle to be utilized is gradually increasing. Therefore, the membership value to be detected for the first pressure change dimension is determined to be the seventh value. For example, the seventh value can be close to -1, such as -0.9 or -0.8. The seventh value can be the opposite of the sixth value. For example, if the sixth value is 0.9, then the seventh value can be -0.9. If the proportions of the first and second pump cycles in each pump cycle to be utilized are both less than the preset proportions, it indicates that the pressure fluctuation amplitude of the accumulator pump remains constant during the single-pump infusion phase of each pump cycle to be utilized. In this case, the membership value to be detected corresponding to the first pressure change dimension is determined to be the eighth value. For example, the eighth value can be close to 0, such as 0.1 or 0.2. The eighth value can be the same as the first value.

[0166] The larger the membership value of the first pressure change dimension, the greater the degree to which the pressure fluctuation amplitude of the accumulator pump decreases in the single-pump infusion stage of each pump cycle, and the greater the probability that the pressure fluctuation amplitude of the accumulator pump in the single-pump infusion stage of each pump cycle is decreasing.

[0167] For the maximum pressure dimension, a mapping relationship can be constructed between the parameter values ​​and membership values. The range of membership values ​​corresponding to the parameter values ​​of the maximum pressure dimension can be preset; for example, the range can be (0, 1), with the first endpoint being either 0 or 1, and the second endpoint being the other endpoint besides the first. Correspondingly, one endpoint of the available parameter value of the maximum pressure dimension can be mapped to the first endpoint near the value range, the other endpoint can be mapped to the second endpoint near the value range, and the median value of the available parameter value of the maximum pressure dimension can be mapped to the midpoint near the value range (i.e., 0.5).

[0168] If the value of the parameter to be utilized in the maximum pressure dimension is not less than the system pressure value, it indicates that the pressure value of the primary pump is normal during the dual-pump coordination phase of each pump cycle. In this case, the membership degree value to be detected for the maximum pressure dimension can be determined as the ninth value. For example, the ninth value can be close to 0, such as 0.1 or 0.2. The ninth value can be the same as the first value. If the value of the parameter to be utilized in the maximum pressure dimension is less than the system pressure value but greater than the second pressure value, it indicates that the pressure of the primary pump is lost during the dual-pump coordination phase of each pump cycle. In this case, the membership degree value to be detected for the maximum pressure dimension is determined based on the tenth value. For example, the tenth value can be directly used as the membership degree value to be detected for the maximum pressure dimension. Alternatively, the membership degree value to be detected for the maximum pressure dimension can have a linear relationship with the parameter value of the maximum pressure dimension, such as a negative correlation. That is, the larger the value of the parameter to be utilized in the maximum pressure dimension, the smaller the membership degree value to be detected for the maximum pressure dimension, and the maximum membership degree value to be detected for the maximum pressure dimension is the tenth value. For example, the tenth value can be close to 0.5, such as 0.6 or 0.7. The tenth value can be the same as the second value. If the value of the parameter to be utilized in the maximum pressure dimension is less than the second pressure value, it indicates that the pressure of the primary pump in the dual-pump collaborative phase of each pump cycle is close to 0, i.e., there is no pressure. In this case, the membership degree value to be detected corresponding to the maximum pressure dimension is determined to be the eleventh value. For example, the eleventh value can be close to 1, such as 0.9 or 0.8. The eleventh value can be the same as the third value.

[0169] The larger the membership value of the maximum pressure dimension, the greater the degree of pressure loss of the primary pump in the dual-pump coordination stage of each pump cycle, and the greater the probability that the pressure value of the primary pump in the dual-pump coordination stage of each pump cycle is pressureless.

[0170] For each pump cycle to be utilized, if the pressure value of the primary pump in the dual-pump coordination phase of the current pump cycle decreases by a greater degree than the second pressure value (i.e., the difference between the pressure value of the primary pump in the dual-pump coordination phase of the previous pump cycle and the pressure value in the dual-pump coordination phase of the current pump cycle is greater than the second pressure value), it indicates that the pressure value of the primary pump in the dual-pump coordination phase of the current pump cycle has decreased, and then the current pump cycle to be utilized is the third pump cycle. If the pressure value of the primary pump in the current pump cycle changes by a lesser degree than the second pressure value (i.e., the absolute value of the difference between the pressure value of the primary pump in the dual-pump coordination phase of the previous pump cycle and the pressure value in the dual-pump coordination phase of the current pump cycle is not greater than the second pressure value), it indicates that the pressure value of the primary pump in the dual-pump coordination phase of the current pump cycle remains unchanged. If the pressure value of the primary pump increases more than the second pressure value in the dual-pump coordination phase of the current pump cycle relative to the previous pump cycle, that is, the difference between the pressure values ​​of the primary pump in the previous pump cycle and the dual-pump coordination phase of the current pump cycle is less than the negative of the second pressure value, it indicates that the pressure value of the primary pump increases in the dual-pump coordination phase of the current pump cycle, and the current pump cycle is the fourth pump cycle.

[0171] For the second pressure change dimension, a mapping relationship can be constructed between the parameter values ​​and membership values. The range of membership values ​​corresponding to the parameter values ​​of the second pressure change dimension can be preset; for example, the range can be (-1, 1), with the first endpoint being -1 or 1, and the second endpoint being the other endpoint besides the first. Correspondingly, one endpoint of the parameter value to be used in the second pressure change dimension can be mapped to the first endpoint near the value range, the other endpoint can be mapped to the second endpoint near the value range, and the median value of the parameter value to be used in the second pressure change dimension can be mapped to the midpoint (i.e., 0) near the value range.

[0172] If the proportion of the third pump cycle in each pump cycle is not less than a preset proportion, it indicates that the pressure value of the primary pump in the dual-pump collaborative phase of the pump cycle to be utilized is gradually decreasing. Therefore, the membership value to be detected for the second pressure change dimension is determined to be the twelfth value. For example, the twelfth value can be close to 1, such as 0.9 or 0.8. The twelfth value can be the same as the third value. If the proportion of the fourth pump cycle in each pump cycle is not less than a preset proportion, it indicates that the pressure value of the primary pump in the dual-pump collaborative phase of the pump cycle to be utilized is gradually increasing. Therefore, the membership value to be detected for the second pressure change dimension is determined to be the thirteenth value. For example, the thirteenth value can be close to -1, such as -0.9 or -0.8. The thirteenth value can be the opposite of the twelfth value. For example, if the twelfth value is 0.9, then the thirteenth value can be -0.9. If the proportions of the third and fourth pump cycles in each pump cycle are both less than the preset proportions, it indicates that the pressure value of the primary pump remains unchanged during the dual-pump synergy phase of the pump cycle to be utilized. In this case, the membership value to be detected corresponding to the second pressure change dimension is determined to be the fourteenth value. For example, the fourteenth value can be close to 0, such as 0.1 or 0.2. The fourteenth value can be the same as the first value.

[0173] The larger the membership value of the second pressure change dimension, the greater the degree to which the pressure value of the primary pump decreases during the dual-pump coordination phase of each pump cycle to be utilized, and the greater the probability that the pressure value of the primary pump decreases during the dual-pump coordination phase of adjacent pump cycles.

[0174] Based on the above processing, for each specified pressure dimension, the parameter values ​​to be utilized in that specified pressure dimension can be substituted into a pre-set membership function to determine the membership value to be detected for each specified pressure dimension. The membership values ​​corresponding to the parameter values ​​of each specified pressure dimension can characterize the pressure change pattern of the infusion pump in each pump cycle from different pressure dimensions. Correspondingly, the membership values ​​to be detected for each specified pressure dimension can also effectively characterize the pressure change pattern of the infusion pump in each pump cycle to be utilized, that is, can accurately characterize the recent pressure state of the infusion pump in the liquid chromatograph. In this way, it can be further ensured that the recent pressure state of the infusion pump in the liquid chromatograph can be combined with fuzzy inference to promptly determine the fault diagnosis result of the liquid chromatograph in the current pump cycle. That is, it is possible to detect whether the liquid chromatograph is in a faulty operating state in a timely manner, and thus detect the fault of the liquid chromatograph in a timely manner.

[0175] In the subsequent construction of the fuzzy inference model, the method for determining the reference membership values ​​corresponding to the reference parameter values ​​of each specified pressure dimension can refer to the relevant description above for determining the membership values ​​to be detected.

[0176] In one embodiment, a fuzzy inference model can be constructed as follows:

[0177] Step 1: For each preset operating state, based on the pressure value of the infusion pump in the preset operating state during the reference pump cycle and the preset number of pump cycles before the reference pump cycle, determine the reference parameter value for each specified pressure dimension.

[0178] Step 2: Based on the membership function corresponding to each specified pressure dimension that is set in advance, determine the membership value corresponding to the reference parameter value of the specified pressure dimension, and use it as the reference membership value.

[0179] Among them, the membership value corresponding to any parameter value of a specified pressure dimension is greater than -1 and less than 1.

[0180] Step 3: Using the probability of belonging to the preset operating state as the dependent variable and the membership degree corresponding to the parameter values ​​of each specified pressure dimension as the independent variable, construct the fuzzy inference model corresponding to the preset operating state.

[0181] In the fuzzy inference model corresponding to the preset operating state, the probability of belonging to the preset operating state is positively correlated with the membership degree represented by the reference membership degree value that is not less than the first membership degree threshold, positively correlated with the membership degree represented by the reference membership degree value that is less than 0 and greater than the second membership degree threshold, negatively correlated with the membership degree represented by the reference membership degree value that is greater than 0 and less than the first membership degree threshold, and negatively correlated with the membership degree represented by the reference membership degree value that is not greater than the second membership degree threshold.

[0182] Step S103 includes: for each preset operating state, substituting the determined membership values ​​to be detected into the fuzzy inference model corresponding to the preset operating state to obtain the probability that the operating state represented by the pressure parameter to be used belongs to the preset operating state.

[0183] In this embodiment, for each preset operating state, the pressure values ​​of the infusion pump in that preset operating state during a reference pump cycle and a preset number of pump cycles prior to the reference pump cycle can be obtained. The obtained pressure values ​​are used to determine reference parameter values ​​for each specified pressure dimension. For each specified pressure dimension, the membership function corresponding to that specified pressure dimension can be used to determine the membership value corresponding to the reference parameter value of that specified pressure dimension, which is then used as the reference membership value. Step 1 can refer to the relevant description of step S101 in the above embodiment, and step 2 can refer to the relevant description of step S102 in the above embodiment.

[0184] Furthermore, the probability of belonging to the preset operating state can be used as the dependent variable, and the membership degree corresponding to the parameter values ​​of each specified pressure dimension can be used as the independent variable to construct the fuzzy inference model corresponding to the preset operating state. In this way, the mapping relationship between the membership degree corresponding to the parameter value of each preset operating state and the probability of belonging to that preset operating state can be obtained. In the fuzzy inference model corresponding to each preset operating state, the closer the parameter value of each specified pressure dimension is to the reference parameter value of each specified pressure dimension in the preset operating state, the closer the membership degree value corresponding to the parameter value of each specified pressure dimension is to the membership degree value corresponding to the reference parameter value of each specified pressure dimension in the preset operating state. Correspondingly, the probability that the operating state represented by the parameter value of each specified pressure dimension belongs to the preset operating state is greater.

[0185] For example, the probability of belonging to this preset operating state is positively correlated with the membership degree represented by a reference membership degree value not less than the first membership degree threshold, positively correlated with the membership degree represented by a reference membership degree value less than 0 and greater than the second membership degree threshold, negatively correlated with the membership degree represented by a reference membership degree value greater than 0 and less than the first membership degree threshold, and negatively correlated with the membership degree represented by a reference membership degree value not greater than the second membership degree threshold. The first membership degree threshold and the second membership degree threshold can be opposite numbers, such as the first membership degree threshold being 0.5 and the second membership degree threshold being -0.5; or the first membership degree threshold being 0.6 and the second membership degree threshold being -0.6.

[0186] The membership degree corresponding to the parameter value of the average pressure dimension can be represented as u1, the membership degree corresponding to the parameter value of the pressure fluctuation dimension can be represented as u2, the membership degree corresponding to the parameter value of the first pressure change dimension can be represented as u3, the membership degree corresponding to the parameter value of the maximum pressure dimension can be represented as u4, and the membership degree corresponding to the parameter value of the second pressure change dimension can be represented as u5. Accordingly, when the specified pressure dimensions include the average pressure dimension, the pressure fluctuation dimension, the first pressure change dimension, the maximum pressure dimension, and the second pressure change dimension, the set of membership degrees corresponding to the parameter values ​​of each specified pressure dimension can be represented as A = [u1, u2, u3, u4, u5].

[0187] With the first, fifth, eighth, ninth, and fourteenth values ​​being 0.1, the second and tenth values ​​being 0.7, the third, fourth, sixth, eleventh, and twelfth values ​​being 0.9, and the seventh and thirteenth values ​​being -0.9, and the first membership threshold being 0.5 and the second membership threshold being -0.5, the membership values ​​corresponding to the parameter values ​​of each specified pressure dimension are determined according to the method described in the above embodiment. Taking normal operation as an example, the parameter value of the average pressure dimension indicates that the pressure of the accumulator pump is normal during the single-pump infusion stage, and the corresponding reference membership value is 0.1, that is, u1=0.1; the parameter value of the pressure fluctuation dimension indicates that the pressure of the accumulator pump is normal during the single-pump infusion stage. The pressure is stable and without fluctuation, with a corresponding reference membership value of 0.1, i.e., u2=0.1. During the single-pump infusion stage, the accumulator pump pressure is stable; therefore, the parameter value for the first pressure change dimension does not need to be considered, and the corresponding reference membership value can be set to any value, represented as x, i.e., u3=x. The parameter value for the maximum pressure dimension indicates that the primary pump pressure is normal during the dual-pump synergy stage, with a corresponding reference membership value of 0.1, i.e., u4=0.1. During the dual-pump synergy stage, the primary pump pressure is normal, meaning the maximum value of the primary pump pressure remains stable. Therefore, the parameter value for the second pressure change dimension does not need to be considered, and the corresponding reference membership value can be set to any value, represented as y, i.e., u5=y. Accordingly, the set of reference membership values ​​corresponding to the reference parameter values ​​for each specified pressure dimension under normal operating conditions can be represented as A1=[0.1, 0.1, x, 0.1, y]. If u1, u2, and u4 are all greater than 0 and less than the first membership threshold, then a fuzzy inference model R1(A) corresponding to the preset operating state can be constructed. R1(A) is negatively correlated with u1, u2, and u4, but unrelated to u3 and u5. For example, R1(A) = 1 / u1 + 1 / u2 + 10 + 1 / u4 + 10 can be constructed. This R1(A) is just an example and is not limited, as long as it satisfies the condition that R1(A) is negatively correlated with u1, u2, and u4, but unrelated to u3 and u5.

[0188] Taking the first fault state as an example, the parameter value of the average pressure dimension indicates that the accumulator pump has no pressure during the single-pump infusion stage, and the corresponding reference membership degree is 0.9, i.e., u1=0.9; the parameter value of the pressure fluctuation dimension indicates that the accumulator pump pressure fluctuates during the single-pump infusion stage, and the fluctuation degree is moderate, and the corresponding reference membership degree is 0.5, i.e., u2=0.5; the pressure fluctuation amplitude of the accumulator pump gradually decreases during the single-pump infusion stage, and the corresponding reference membership degree is 0.1, i.e., u3=0.1; the parameter value of the maximum pressure dimension indicates that the primary pump has no pressure during the dual-pump synergy stage, and the corresponding reference membership degree is 0.9, i.e., u4=0.9; the primary pump has no pressure during the dual-pump synergy stage, i.e., the maximum value of the primary pump pressure is stable at 0. Therefore, the parameter value of the second pressure change dimension does not need to be considered, and the corresponding reference membership degree value can be set to any value, which can be represented as y, i.e., u5=y. If u1, u2, and u4 are all not less than the first membership threshold, and u3 is greater than 0 and less than the first membership threshold, then a fuzzy inference model R2(A) corresponding to this preset operating state can be constructed. R2(A) is positively correlated with u1, u2, and u4, negatively correlated with u3, and unrelated to u5. For example, R2(A) = 10 × u1 + 10 × u2 + 1 / u3 + 10 × u4 + 10 can be constructed. This R2(A) is just an example and is not limited, as long as it satisfies the condition that R2(A) is positively correlated with u1, u2, and u4, negatively correlated with u3, and unrelated to u5.

[0189] Following this logic, we can obtain the fuzzy inference models R3(A), R4(A), R5(A), and R6(A) corresponding to the second, third, fourth, and fifth fault states, respectively. Furthermore, the overall fuzzy inference model can be represented as R = [R1, R2, R3, R4, R5, R6]. Subsequently, adding other preset operating states can continue to construct new fuzzy inference models (also known as fuzzy rules) according to the above rules.

[0190] When diagnosing faults in the current pump cycle, for each preset operating state, the determined membership values ​​to be detected can be substituted into the fuzzy inference model corresponding to that preset operating state to obtain the probability that the operating state represented by the pressure parameter to be utilized belongs to that preset operating state. With 6 preset operating states and 5 specified pressure dimensions, the probability that the operating state represented by the pressure parameter to be utilized belongs to each preset operating state can be calculated using the fuzzy inference model based on the determined membership values ​​to be detected, and can be expressed as follows: .in, =[P1, P2, P3, P4, P5, P6] represents the probability that the operating state characterized by the pressure parameter to be used belongs to each preset operating state. This represents the membership value to be detected corresponding to the parameter value to be utilized for each specified pressure dimension. This represents the reference membership value corresponding to the reference parameter value for each specified pressure dimension under each preset operating state. For example, 1 represents the number of rows in the matrix, and 6 represents the number of columns in the matrix.

[0191] Based on the above processing, it is further ensured that the fault diagnosis result of the liquid chromatograph in the current pump cycle can be determined in a timely manner by combining the recent pressure status of the pump in the liquid chromatograph and using fuzzy inference rules. That is, it is possible to detect in a timely manner whether the liquid chromatograph is in a faulty operating state, and thus to detect the fault of the liquid chromatograph in a timely manner.

[0192] In one embodiment, see Figure 4 , Figure 4 This is a flowchart illustrating the construction of a fuzzy inference model, provided as an embodiment of this application. The fuzzy inference model can also be referred to as a fault diagnosis and diagnostic model based on fuzzy inference. The fuzzy inference model can be constructed through the following steps:

[0193] Step S401: Select appropriate infusion pump pressure parameters. That is, determine the specified pressure dimension.

[0194] Step S402: Construct a fuzzy set of pressure parameters. That is, steps 1 and 2 in the above embodiments.

[0195] Step S403: Construct a fault detection and diagnosis model based on fuzzy reasoning. That is, step 3 in the above embodiment.

[0196] Based on the above processing, it is further ensured that appropriate pressure parameters can be extracted during the pump cycle, and online fault detection and diagnosis of the liquid chromatograph can be achieved through fuzzy reasoning. That is, it is possible to detect in a timely manner whether the liquid chromatograph is in a faulty operating state, and thus detect the fault of the liquid chromatograph in a timely manner.

[0197] In one embodiment, see Figure 5 , Figure 5 This is a schematic diagram of a second process for a liquid chromatograph fault diagnosis method provided in an embodiment of this application. The liquid chromatograph fault diagnosis method may include:

[0198] Step S501: Start infusion. That is, the liquid chromatograph starts running and the infusion pump starts infusion.

[0199] Step S502: Obtain the pressure parameters for the most recent n pump cycles. That is, step S101 in the above embodiment.

[0200] Step S503: Calculate the membership value of each pressure parameter according to the membership function. That is, step S102 in the above embodiment.

[0201] Step S504: Calculate the current status of the infusion pump using a fault detection and diagnosis model based on fuzzy reasoning. That is, steps S103 and S104 in the above embodiment.

[0202] Step S505: Is there an abnormality? That is, determine whether the fault diagnosis result of the liquid chromatograph in the current pump cycle indicates a faulty operating state. If yes, proceed to step S506; if no, proceed to step S502, and the liquid chromatograph continues to run to continue fault diagnosis for the latest pump cycle.

[0203] Step S506: Execute appropriate abnormal response measures. For example, issue an abnormal warning message to inform the user that the liquid chromatograph has malfunctioned and to indicate the corresponding cause of the malfunction. Subsequently, the user can handle the malfunction according to the indicated cause.

[0204] Step S507: End.

[0205] Based on the above processing, by combining the recent pressure status of the infusion pump in the liquid chromatograph and using fuzzy reasoning, the fault diagnosis result of the liquid chromatograph in the current pump cycle can be determined in a timely manner. That is, it is possible to detect in a timely manner whether the liquid chromatograph is in a faulty operating state, and thus to detect the fault in the liquid chromatograph promptly. Furthermore, it is possible to determine the cause of the liquid chromatograph fault in a timely manner, facilitating subsequent fault handling by the user and further improving the efficiency of instrument fault diagnosis for users and after-sales personnel.

[0206] Based on the same inventive concept, this application also provides a liquid chromatograph, which includes: an infusion pump, a pressure sensor, and a status detection terminal;

[0207] The pressure sensor is used to measure the pressure of the infusion pump and send the measured pressure to the status detection terminal;

[0208] The status detection terminal is used to receive the pressure sent by the pressure sensor and execute any of the liquid chromatograph fault diagnosis methods in the above embodiments.

[0209] Based on the liquid chromatograph provided in this application embodiment, parameter values ​​(i.e., parameter values ​​to be used) for each specified pressure dimension can be determined according to the pump cycle in the liquid chromatograph that needs to be detected (i.e., the current pump cycle) and the pressure values ​​of a preset number of pump cycles prior to the current pump cycle. Each specified pressure dimension includes: a dimension characterizing the magnitude of the pressure value and / or a dimension characterizing the fluctuation state of the pressure value. Accordingly, each parameter value to be used can characterize the recent pressure state of the pump in the liquid chromatograph from different dimensions. Furthermore, a membership function corresponding to each specified pressure dimension is preset. By combining the corresponding membership function, the membership value corresponding to each specified pressure dimension (i.e., the membership value to be detected) can be determined. Accordingly, each determined membership value to be detected can represent the recent pressure state of the pump in the liquid chromatograph.

[0210] For each preset operating state, a reference membership value corresponding to the parameter value of a specified pressure dimension in that preset operating state can be determined in advance based on the membership function corresponding to each specified pressure dimension and the parameter value of that specified pressure dimension in that preset operating state. Then, based on the reference membership values ​​corresponding to the parameter values ​​of each specified pressure dimension in each preset operating state, a fuzzy inference model can be constructed. Correspondingly, the fuzzy inference model can represent the fuzzy relationship between the membership values ​​corresponding to the parameter values ​​of each specified pressure dimension and the operating state. Combining each membership value to be detected and the fuzzy inference model, the probability that the operating state represented by the parameter value to be used corresponding to each membership value to be detected belongs to each preset operating state can be determined based on the similarity between each membership value to be detected and the reference membership value in each preset operating state in the fuzzy inference model.

[0211] The preset operating state with the highest probability can represent the operating state of the liquid chromatograph in the current pump cycle. Therefore, the preset operating state with the highest probability can be determined as the fault diagnosis result of the liquid chromatograph in the current pump cycle. The preset operating states include: normal operating state and various fault operating states. Correspondingly, the determined fault diagnosis result can indicate whether the liquid chromatograph is in a faulty operating state. In this way, by combining the recent pressure status of the infusion pump in the liquid chromatograph and using fuzzy reasoning, the fault diagnosis result of the liquid chromatograph in the current pump cycle can be determined in a timely manner. That is, it is possible to detect whether the liquid chromatograph is in a faulty operating state in a timely manner, and thus, to detect the liquid chromatograph malfunction in a timely manner.

[0212] In one embodiment, where the infusion pump includes a primary pump and a accumulator pump connected in series, the pressure sensor comprises: a first pressure sensor deployed at the outlet of the primary pump for measuring the pressure value of the primary pump, and a second pressure sensor deployed at the outlet of the accumulator pump for measuring the pressure value of the accumulator pump.

[0213] In this embodiment, the structure of the infusion pump can be referred to Figure 2 The infusion pump shown.

[0214] Based on the same inventive concept, this application also provides a liquid chromatograph fault diagnosis device, see [link to relevant documentation]. Figure 6 , Figure 6 This is a schematic diagram of a liquid chromatograph fault diagnosis device provided in an embodiment of this application. The device includes:

[0215] The parameter value determination module 601 is used to determine the parameter values ​​to be used for each specified pressure dimension based on the pressure values ​​of the infusion pump in the liquid chromatograph during the current pump cycle and a preset number of pump cycles prior to the current pump cycle; wherein, each specified pressure dimension includes: a dimension characterizing the magnitude of the pressure value and / or a dimension characterizing the fluctuation state of the pressure value;

[0216] The membership value determination module 602 is used to determine the membership value corresponding to the parameter value to be used in the specified pressure dimension according to the membership function corresponding to each specified pressure dimension that is preset, and use it as the membership value to be detected.

[0217] The probability determination module 603 is used to combine the determined membership values ​​to be detected and, using a pre-constructed fuzzy inference model, determine the probability that the operating state represented by the parameter value to be used belongs to each preset operating state; wherein, the preset operating states include: normal operating state and multiple fault operating states; the fuzzy inference model is constructed based on the reference membership values ​​corresponding to the parameter values ​​of each specified pressure dimension in each preset operating state; the reference membership value corresponding to the parameter value of a specified pressure dimension in a preset operating state is determined according to the membership function corresponding to the specified pressure dimension and the parameter value of the specified pressure dimension in the preset operating state;

[0218] The diagnostic result determination module 604 is used to determine the preset operating state with the highest probability as the fault diagnosis result of the liquid chromatograph in the current pump cycle.

[0219] Based on the liquid chromatograph fault diagnosis device provided in this application embodiment, the parameter values ​​(i.e., the parameter values ​​to be used) for each specified pressure dimension can be determined according to the pump cycle (i.e., the current pump cycle) of the infusion pump in the liquid chromatograph that needs to be tested, and the pressure values ​​of a preset number of pump cycles prior to the current pump cycle. Each specified pressure dimension includes: a dimension characterizing the magnitude of the pressure value and / or a dimension characterizing the fluctuation state of the pressure value. Accordingly, each parameter value to be used can characterize the recent pressure state of the infusion pump in the liquid chromatograph from different dimensions. Furthermore, a membership function corresponding to each specified pressure dimension is preset. By combining the corresponding membership function, the membership value (i.e., the membership value to be detected) corresponding to each specified pressure dimension can be determined. Accordingly, each determined membership value to be detected can represent the recent pressure state of the infusion pump in the liquid chromatograph.

[0220] For each preset operating state, a reference membership value corresponding to the parameter value of a specified pressure dimension in that preset operating state can be determined in advance based on the membership function corresponding to each specified pressure dimension and the parameter value of that specified pressure dimension in that preset operating state. Then, based on the reference membership values ​​corresponding to the parameter values ​​of each specified pressure dimension in each preset operating state, a fuzzy inference model can be constructed. Correspondingly, the fuzzy inference model can represent the fuzzy relationship between the membership values ​​corresponding to the parameter values ​​of each specified pressure dimension and the operating state. Combining each membership value to be detected and the fuzzy inference model, the probability that the operating state represented by the parameter value to be used corresponding to each membership value to be detected belongs to each preset operating state can be determined based on the similarity between each membership value to be detected and the reference membership value in each preset operating state in the fuzzy inference model.

[0221] The preset operating state with the highest probability can represent the operating state of the liquid chromatograph in the current pump cycle. Therefore, the preset operating state with the highest probability can be determined as the fault diagnosis result of the liquid chromatograph in the current pump cycle. The preset operating states include: normal operating state and various fault operating states. Correspondingly, the determined fault diagnosis result can indicate whether the liquid chromatograph is in a faulty operating state. In this way, by combining the recent pressure status of the infusion pump in the liquid chromatograph and using fuzzy reasoning, the fault diagnosis result of the liquid chromatograph in the current pump cycle can be determined in a timely manner. That is, it is possible to detect whether the liquid chromatograph is in a faulty operating state in a timely manner, and thus, to detect the liquid chromatograph malfunction in a timely manner.

[0222] In one embodiment, the fuzzy inference model is constructed in the following manner:

[0223] For each preset operating state, based on the pressure value of the infusion pump in that preset operating state during a reference pump cycle and a preset number of pump cycles before the reference pump cycle, a reference parameter value for each specified pressure dimension is determined.

[0224] Based on the membership function corresponding to each specified pressure dimension, the membership value corresponding to the reference parameter value of the specified pressure dimension is determined and used as the reference membership value; wherein, the membership value corresponding to the parameter value of any specified pressure dimension is greater than -1 and less than 1;

[0225] Using the probability of belonging to the preset operating state as the dependent variable and the membership degree corresponding to the parameter values ​​of each specified pressure dimension as the independent variable, a fuzzy inference model corresponding to the preset operating state is constructed. In the fuzzy inference model corresponding to the preset operating state, the probability of belonging to the preset operating state is positively correlated with the membership degree represented by the reference membership degree value not less than the first membership degree threshold, positively correlated with the membership degree represented by the reference membership degree value less than 0 and greater than the second membership degree threshold, negatively correlated with the membership degree represented by the reference membership degree value greater than 0 and less than the first membership degree threshold, and negatively correlated with the membership degree represented by the reference membership degree value not greater than the second membership degree threshold.

[0226] The probability determination module 603 is specifically used to, for each preset operating state, substitute the determined membership values ​​to be detected into the fuzzy inference model corresponding to the preset operating state to obtain the probability that the operating state represented by the pressure parameter to be utilized belongs to the preset operating state.

[0227] In one embodiment, the infusion pump includes a primary pump and a accumulator pump connected in series; each pump cycle includes: a single-pump infusion phase and a dual-pump synergistic phase.

[0228] Each specified stress dimension includes any of the following:

[0229] The pressure dimension represents the average pressure level of the accumulator pump during the single-pump delivery phase of each pump cycle; the pressure fluctuation dimension represents whether the pressure value of the accumulator pump fluctuates during the single-pump delivery phase of each pump cycle; the first pressure change dimension represents the trend of the pressure fluctuation amplitude of the accumulator pump during the single-pump delivery phase of each pump cycle; the maximum pressure dimension represents the maximum pressure value of the primary pump during the dual-pump synergy phase of each pump cycle; and the second pressure change dimension represents the trend of the pressure value of the primary pump during the dual-pump synergy phase of adjacent pump cycles.

[0230] In one embodiment, the membership degree corresponding to the parameter value of the average pressure dimension represents the probability that the pressure value of the accumulator pump in the single-pump infusion phase of each pump cycle belongs to the state of no pressure.

[0231] The membership degree corresponding to the parameter value of the pressure fluctuation dimension represents the probability that the pressure value of the accumulator pump fluctuates during the single-pump infusion phase of each pump cycle.

[0232] The membership degree corresponding to the parameter value of the first pressure change dimension indicates that the probability of the pressure fluctuation amplitude of the accumulator pump decreasing during the single pump delivery phase of each pump cycle is:

[0233] The membership degree corresponding to the parameter value of the maximum pressure dimension represents the probability that the pressure value of the primary pump in the dual-pump collaborative phase of each pump cycle is pressureless.

[0234] The membership degree corresponding to the parameter value of the second pressure change dimension indicates the probability that the pressure value of the primary pump in the dual-pump collaborative phase of adjacent pump cycles is decreasing.

[0235] In one embodiment, the membership value determination module 602 is specifically configured to: if the usable parameter value of the average pressure dimension is not less than a first pressure value, determine the membership value to be detected for the average pressure dimension as a first value; the first pressure value is the product of a first coefficient and a system pressure value; the first coefficient is less than 1; the system pressure value is the pressure value of the accumulator pump in the current pump cycle under the most recent normal operating condition; if the usable parameter value of the average pressure dimension is less than the first pressure value but greater than a second pressure value, determine the membership value to be detected for the average pressure dimension based on a second value; the second pressure value is the product of a second coefficient and a system pressure value; if the usable parameter value of the average pressure dimension is less than the second pressure value, determine the membership value to be detected for the average pressure dimension as a third value.

[0236] And / or,

[0237] If the parameter value to be used in the pressure fluctuation dimension indicates that the pressure value of the accumulator pump fluctuates in the single-pump infusion stage of each pump cycle, then the membership value to be detected corresponding to the pressure fluctuation dimension is determined based on the fourth value; if the parameter value to be used in the pressure fluctuation dimension indicates that the pressure value of the accumulator pump does not fluctuate in the single-pump infusion stage of each pump cycle, then the membership value to be detected corresponding to the pressure fluctuation dimension is determined to be the fifth value.

[0238] And / or,

[0239] If the proportion of the first pump cycle in each pump cycle is not less than a preset proportion, then the membership value to be detected corresponding to the first pressure change dimension is determined to be the sixth value; the degree to which the pressure fluctuation amplitude of the accumulator pump decreases in the single-pump infusion stage of the first pump cycle relative to the previous pump cycle is greater than the second pressure value; if the proportion of the second pump cycle in each pump cycle is not less than a preset proportion, then the membership value to be detected corresponding to the first pressure change dimension is determined to be the seventh value; the degree to which the pressure fluctuation amplitude of the accumulator pump increases in the single-pump infusion stage of the second pump cycle relative to the previous pump cycle is greater than the second pressure value; if the proportions of both the first pump cycle and the second pump cycle in each pump cycle are less than a preset proportion, then the membership value to be detected corresponding to the first pressure change dimension is determined to be the eighth value.

[0240] And / or,

[0241] If the value of the parameter to be used in the maximum pressure dimension is not less than the system pressure value, then the membership value to be detected corresponding to the maximum pressure dimension is determined to be the ninth value; if the value of the parameter to be used in the maximum pressure dimension is less than the system pressure value but greater than the second pressure value, then the membership value to be detected corresponding to the maximum pressure dimension is determined based on the tenth value; if the value of the parameter to be used in the maximum pressure dimension is less than the second pressure value, then the membership value to be detected corresponding to the maximum pressure dimension is determined to be the eleventh value.

[0242] And / or,

[0243] If the proportion of the third pump cycle in each pump cycle is not less than a preset proportion, then the membership degree value to be detected corresponding to the second pressure change dimension is determined to be the twelfth value; the pressure value of the primary pump in the dual-pump collaborative stage of the third pump cycle relative to the previous pump cycle decreases by a greater degree than the second pressure value; if the proportion of the fourth pump cycle in each pump cycle is not less than a preset proportion, then the membership degree value to be detected corresponding to the second pressure change dimension is determined to be the thirteenth value; the pressure value of the primary pump in the dual-pump collaborative stage of the third pump cycle relative to the previous pump cycle increases by a greater degree than the second pressure value; if the proportions of both the third and fourth pump cycles in each pump cycle are less than a preset proportion, then the membership degree value to be detected corresponding to the second pressure change dimension is determined to be the fourteenth value.

[0244] This application also provides an electronic device, such as... Figure 7 As shown, it includes:

[0245] Memory 701 is used to store computer programs;

[0246] When processor 702 executes a program stored in memory 701, it performs the following steps:

[0247] Based on the pressure values ​​of the infusion pump in the liquid chromatograph during the current pump cycle and a preset number of pump cycles prior to the current pump cycle, the parameter values ​​to be used for each specified pressure dimension are determined; wherein, each specified pressure dimension includes: a dimension characterizing the magnitude of the pressure value and / or a dimension characterizing the fluctuation state of the pressure value.

[0248] Based on the membership function corresponding to each specified pressure dimension that is set in advance, the membership value corresponding to the parameter value to be used in the specified pressure dimension is determined and used as the membership value to be detected.

[0249] Based on the determined membership values ​​to be detected, a pre-constructed fuzzy inference model is used to determine the probability that the operating state represented by the parameter value to be used belongs to each preset operating state; wherein, the preset operating states include: normal operating state and multiple fault operating states; the fuzzy inference model is constructed based on the reference membership values ​​corresponding to the parameter values ​​of each specified pressure dimension under each preset operating state; the reference membership value corresponding to the parameter value of a specified pressure dimension under a preset operating state is determined according to the membership function corresponding to the specified pressure dimension and the parameter value of the specified pressure dimension under the preset operating state;

[0250] The preset operating state with the highest probability is determined as the fault diagnosis result of the liquid chromatograph in the current pump cycle.

[0251] Furthermore, the aforementioned electronic device may also include a communication bus and / or a communication interface, with the processor 702, the communication interface, and the memory 701 communicating with each other via the communication bus.

[0252] The communication bus mentioned in the above electronic devices can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0253] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0254] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0255] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0256] In another embodiment provided in this application, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor, implements the steps of any of the above-described liquid chromatograph fault diagnosis methods.

[0257] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the liquid chromatograph fault diagnosis methods described in the above embodiments.

[0258] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The 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 integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a solid-state drive (SSD), etc.

[0259] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0260] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of liquid chromatographs, apparatuses, electronic devices, storage media, and program products are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0261] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application are included within the scope of protection of this application.

Claims

1. A liquid chromatograph failure diagnosis method characterized by, The method includes: Based on the pressure values ​​of the infusion pump in the liquid chromatograph during the current pump cycle and a preset number of pump cycles prior to the current pump cycle, determine the parameter values ​​to be used for each specified pressure dimension. Based on the membership function corresponding to each specified pressure dimension that is set in advance, the membership value corresponding to the parameter value to be used in the specified pressure dimension is determined and used as the membership value to be detected. For each preset operating state, the determined membership values ​​to be detected are substituted into the fuzzy inference model corresponding to the preset operating state to obtain the probability that the operating state represented by the parameter value to be used belongs to the preset operating state; wherein, the preset operating states include: normal operating state and multiple fault operating states; The fuzzy inference model is constructed as follows: For each preset operating state, based on the pressure values ​​of the infusion pump in that preset operating state during a reference pump cycle and a preset number of pump cycles prior to the reference pump cycle, reference parameter values ​​for each specified pressure dimension are determined; according to the membership function corresponding to each specified pressure dimension (pre-set), the membership value corresponding to the reference parameter value of that specified pressure dimension is determined as the reference membership value; wherein, the membership value corresponding to the parameter value of any specified pressure dimension is greater than -1 and less than 1; the probability of belonging to that preset operating state is used as the dependent variable, and each specified pressure... The membership degree corresponding to the parameter value of the dimension is used as the independent variable to construct the fuzzy inference model corresponding to the preset operating state. In the fuzzy inference model corresponding to the preset operating state, the probability of belonging to the preset operating state is positively correlated with the membership degree represented by the reference membership degree value not less than the first membership degree threshold, positively correlated with the membership degree represented by the reference membership degree value less than 0 and greater than the second membership degree threshold, negatively correlated with the membership degree represented by the reference membership degree value greater than 0 and less than the first membership degree threshold, and negatively correlated with the membership degree represented by the reference membership degree value not greater than the second membership degree threshold. The preset operating state with the highest probability is determined as the fault diagnosis result of the liquid chromatograph in the current pump cycle; The infusion pump includes a primary pump and a accumulator pump connected in series; each pump cycle includes a single-pump infusion phase and a dual-pump synergy phase; each specified pressure dimension includes any of the following: an average pressure dimension characterizing the average pressure level of the accumulator pump during the single-pump infusion phase of each pump cycle, a pressure fluctuation dimension characterizing whether the pressure value of the accumulator pump fluctuates during the single-pump infusion phase of each pump cycle, a first pressure change dimension characterizing the trend of the pressure fluctuation amplitude of the accumulator pump during the single-pump infusion phase of each pump cycle, a maximum pressure dimension characterizing the maximum pressure value of the primary pump during the dual-pump synergy phase of each pump cycle, and a second pressure change dimension characterizing the trend of the pressure value of the primary pump during the dual-pump synergy phase of adjacent pump cycles.

2. The method of claim 1, wherein, The membership degree corresponding to the parameter value of the average pressure dimension represents the probability that the pressure value of the accumulator pump in the single pump infusion phase of each pump cycle belongs to no pressure. The membership degree corresponding to the parameter value of the pressure fluctuation dimension represents the probability that the pressure value of the accumulator pump fluctuates during the single-pump infusion phase of each pump cycle. The membership degree corresponding to the parameter value of the first pressure change dimension indicates that the probability of the pressure fluctuation amplitude of the accumulator pump decreasing during the single pump delivery phase of each pump cycle is: The membership degree corresponding to the parameter value of the maximum pressure dimension represents the probability that the pressure value of the primary pump in the dual-pump collaborative phase of each pump cycle is pressureless. The membership degree corresponding to the parameter value of the second pressure change dimension indicates the probability that the pressure value of the primary pump in the dual-pump collaborative phase of adjacent pump cycles is decreasing.

3. The method of claim 1, wherein, The step of determining the membership value corresponding to the parameter value to be used in each specified pressure dimension according to the pre-set membership function for each specified pressure dimension, and using it as the membership value to be detected, includes: If the value of the parameter to be used in the average pressure dimension is not less than the first pressure value, then the membership value to be detected corresponding to the average pressure dimension is determined to be a first value; the first pressure value is the product of a first coefficient and the system pressure value; the first coefficient is less than 1; the system pressure value is the pressure value of the accumulator pump in the current pump cycle under the most recent normal operating condition; if the value of the parameter to be used in the average pressure dimension is less than the first pressure value but greater than the second pressure value, then the membership value to be detected corresponding to the average pressure dimension is determined based on the second value; the second pressure value is the product of a second coefficient and the system pressure value; if the value of the parameter to be used in the average pressure dimension is less than the second pressure value, then the membership value to be detected corresponding to the average pressure dimension is determined to be a third value. If the parameter value to be used in the pressure fluctuation dimension indicates that the pressure value of the accumulator pump fluctuates in the single-pump infusion stage of each pump cycle, then the membership value to be detected corresponding to the pressure fluctuation dimension is determined based on the fourth value; if the parameter value to be used in the pressure fluctuation dimension indicates that the pressure value of the accumulator pump does not fluctuate in the single-pump infusion stage of each pump cycle, then the membership value to be detected corresponding to the pressure fluctuation dimension is determined to be the fifth value. If the proportion of the first pump cycle in each pump cycle is not less than a preset proportion, then the membership value to be detected corresponding to the first pressure change dimension is determined to be the sixth value; the degree to which the pressure fluctuation amplitude of the accumulator pump decreases in the single-pump infusion stage of the first pump cycle relative to the previous pump cycle is greater than the second pressure value; if the proportion of the second pump cycle in each pump cycle is not less than a preset proportion, then the membership value to be detected corresponding to the first pressure change dimension is determined to be the seventh value; the degree to which the pressure fluctuation amplitude of the accumulator pump increases in the single-pump infusion stage of the second pump cycle relative to the previous pump cycle is greater than the second pressure value; if the proportions of both the first pump cycle and the second pump cycle in each pump cycle are less than a preset proportion, then the membership value to be detected corresponding to the first pressure change dimension is determined to be the eighth value. If the value of the parameter to be used in the maximum pressure dimension is not less than the system pressure value, then the membership value to be detected corresponding to the maximum pressure dimension is determined to be the ninth value; if the value of the parameter to be used in the maximum pressure dimension is less than the system pressure value but greater than the second pressure value, then the membership value to be detected corresponding to the maximum pressure dimension is determined based on the tenth value; if the value of the parameter to be used in the maximum pressure dimension is less than the second pressure value, then the membership value to be detected corresponding to the maximum pressure dimension is determined to be the eleventh value. If the proportion of the third pump cycle in each pump cycle is not less than a preset proportion, then the membership degree value to be detected corresponding to the second pressure change dimension is determined to be the twelfth value; the pressure value of the primary pump in the dual-pump collaborative stage of the third pump cycle relative to the previous pump cycle decreases by a greater degree than the second pressure value; if the proportion of the fourth pump cycle in each pump cycle is not less than a preset proportion, then the membership degree value to be detected corresponding to the second pressure change dimension is determined to be the thirteenth value; the pressure value of the primary pump in the dual-pump collaborative stage of the third pump cycle relative to the previous pump cycle increases by a greater degree than the second pressure value; if the proportions of both the third and fourth pump cycles in each pump cycle are less than a preset proportion, then the membership degree value to be detected corresponding to the second pressure change dimension is determined to be the fourteenth value.

4. A liquid chromatograph characterized by, The liquid chromatograph includes: an infusion pump, a pressure sensor, and a status detection terminal; The pressure sensor is used to measure the pressure of the infusion pump and send the measured pressure to the status detection terminal; The status detection terminal is used to receive the pressure sent by the pressure sensor and execute the method described in any one of claims 1-3.

5. The liquid chromatograph of claim 4, wherein, In the case where the infusion pump includes a primary pump and a accumulator pump connected in series, the pressure sensor comprises: a first pressure sensor deployed at the outlet of the primary pump for measuring the pressure value of the primary pump, and a second pressure sensor deployed at the outlet of the accumulator pump for measuring the pressure value of the accumulator pump.

6. A liquid chromatograph failure diagnosis device characterized by comprising: The device includes: The parameter value determination module is used to determine the parameter values ​​to be used for each specified pressure dimension based on the pressure values ​​of the infusion pump in the liquid chromatograph during the current pump cycle and a preset number of pump cycles before the current pump cycle. The membership value determination module is used to determine the membership value corresponding to the parameter value to be used in the specified pressure dimension according to the membership function corresponding to each specified pressure dimension that is preset, and use it as the membership value to be detected. The probability determination module is used to, for each preset operating state, substitute the determined membership values ​​to be detected into the fuzzy inference model corresponding to the preset operating state to obtain the probability that the operating state represented by the parameter value to be used belongs to the preset operating state; wherein, the preset operating states include: normal operating state and multiple fault operating states; The fuzzy inference model is constructed as follows: For each preset operating state, based on the pressure values ​​of the infusion pump in that preset operating state during a reference pump cycle and a preset number of pump cycles prior to the reference pump cycle, reference parameter values ​​for each specified pressure dimension are determined; according to the membership function corresponding to each specified pressure dimension (pre-set), the membership value corresponding to the reference parameter value of that specified pressure dimension is determined as the reference membership value; wherein, the membership value corresponding to the parameter value of any specified pressure dimension is greater than -1 and less than 1; the probability of belonging to that preset operating state is used as the dependent variable, and each specified pressure... The membership degree corresponding to the parameter value of the dimension is used as the independent variable to construct the fuzzy inference model corresponding to the preset operating state. In the fuzzy inference model corresponding to the preset operating state, the probability of belonging to the preset operating state is positively correlated with the membership degree represented by the reference membership degree value not less than the first membership degree threshold, positively correlated with the membership degree represented by the reference membership degree value less than 0 and greater than the second membership degree threshold, negatively correlated with the membership degree represented by the reference membership degree value greater than 0 and less than the first membership degree threshold, and negatively correlated with the membership degree represented by the reference membership degree value not greater than the second membership degree threshold. The diagnostic result determination module is used to determine the preset operating state with the highest probability as the fault diagnosis result of the liquid chromatograph in the current pump cycle; The infusion pump includes a primary pump and a accumulator pump connected in series; each pump cycle includes a single-pump infusion phase and a dual-pump synergy phase; each specified pressure dimension includes any of the following: an average pressure dimension characterizing the average pressure level of the accumulator pump during the single-pump infusion phase of each pump cycle, a pressure fluctuation dimension characterizing whether the pressure value of the accumulator pump fluctuates during the single-pump infusion phase of each pump cycle, a first pressure change dimension characterizing the trend of the pressure fluctuation amplitude of the accumulator pump during the single-pump infusion phase of each pump cycle, a maximum pressure dimension characterizing the maximum pressure value of the primary pump during the dual-pump synergy phase of each pump cycle, and a second pressure change dimension characterizing the trend of the pressure value of the primary pump during the dual-pump synergy phase of adjacent pump cycles.

7. The apparatus of claim 6, wherein, The membership degree corresponding to the parameter value of the average pressure dimension represents the probability that the pressure value of the accumulator pump in the single pump infusion phase of each pump cycle belongs to no pressure. The membership degree corresponding to the parameter value of the pressure fluctuation dimension represents the probability that the pressure value of the accumulator pump fluctuates during the single-pump infusion phase of each pump cycle. The membership degree corresponding to the parameter value of the first pressure change dimension indicates that the probability of the pressure fluctuation amplitude of the accumulator pump decreasing during the single pump delivery phase of each pump cycle is: The membership degree corresponding to the parameter value of the maximum pressure dimension represents the probability that the pressure value of the primary pump in the dual-pump collaborative phase of each pump cycle is pressureless. The membership degree corresponding to the parameter value of the second pressure change dimension indicates the probability that the pressure value of the primary pump in the dual-pump collaborative phase of adjacent pump cycles is decreasing. The membership value determination module is specifically used to determine the membership value to be detected corresponding to the average pressure dimension as a first value if the value of the parameter to be utilized in the average pressure dimension is not less than the first pressure value; the first pressure value is the product of the first coefficient and the system pressure value. The first coefficient is less than 1; the system pressure value is the pressure value of the accumulator pump in the current pump cycle under the most recent normal operating condition; if the value of the parameter to be used in the average pressure dimension is less than the first pressure value and greater than the second pressure value, then the membership value to be detected corresponding to the average pressure dimension is determined based on the second value. The second pressure value is the product of the second coefficient and the system pressure value; if the value of the parameter to be used in the average pressure dimension is less than the second pressure value, then the membership degree value to be detected corresponding to the average pressure dimension is determined to be the third value. If the parameter value to be used in the pressure fluctuation dimension indicates that the pressure value of the accumulator pump fluctuates in the single-pump infusion stage of each pump cycle, then the membership value to be detected corresponding to the pressure fluctuation dimension is determined based on the fourth value; if the parameter value to be used in the pressure fluctuation dimension indicates that the pressure value of the accumulator pump does not fluctuate in the single-pump infusion stage of each pump cycle, then the membership value to be detected corresponding to the pressure fluctuation dimension is determined to be the fifth value. If the proportion of the first pump cycle in each pump cycle is not less than a preset proportion, then the membership value to be detected corresponding to the first pressure change dimension is determined to be the sixth value; the degree to which the pressure fluctuation amplitude of the accumulator pump decreases in the single-pump infusion stage of the first pump cycle relative to the previous pump cycle is greater than the second pressure value; if the proportion of the second pump cycle in each pump cycle is not less than a preset proportion, then the membership value to be detected corresponding to the first pressure change dimension is determined to be the seventh value; the degree to which the pressure fluctuation amplitude of the accumulator pump increases in the single-pump infusion stage of the second pump cycle relative to the previous pump cycle is greater than the second pressure value; if the proportions of both the first pump cycle and the second pump cycle in each pump cycle are less than a preset proportion, then the membership value to be detected corresponding to the first pressure change dimension is determined to be the eighth value. If the value of the parameter to be used in the maximum pressure dimension is not less than the system pressure value, then the membership value to be detected corresponding to the maximum pressure dimension is determined to be the ninth value; if the value of the parameter to be used in the maximum pressure dimension is less than the system pressure value but greater than the second pressure value, then the membership value to be detected corresponding to the maximum pressure dimension is determined based on the tenth value; if the value of the parameter to be used in the maximum pressure dimension is less than the second pressure value, then the membership value to be detected corresponding to the maximum pressure dimension is determined to be the eleventh value. If the proportion of the third pump cycle in each pump cycle is not less than a preset proportion, then the membership degree value to be detected corresponding to the second pressure change dimension is determined to be the twelfth value; the pressure value of the primary pump in the dual-pump collaborative stage of the third pump cycle relative to the previous pump cycle decreases by a greater degree than the second pressure value; if the proportion of the fourth pump cycle in each pump cycle is not less than a preset proportion, then the membership degree value to be detected corresponding to the second pressure change dimension is determined to be the thirteenth value; the pressure value of the primary pump in the dual-pump collaborative stage of the third pump cycle relative to the previous pump cycle increases by a greater degree than the second pressure value; if the proportions of both the third and fourth pump cycles in each pump cycle are less than a preset proportion, then the membership degree value to be detected corresponding to the second pressure change dimension is determined to be the fourteenth value.

8. An electronic device, comprising: include: Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the method of any one of claims 1-3.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method described in any one of claims 1-3.

10. A computer program product, characterised in that, When the computer program product is run on a computer, it causes the computer to perform the method according to any one of claims 1-3.

Images