Fault diagnosis method and device and computer readable storage medium

Abstract

The application discloses a kind of fault diagnosis method, device and computer readable storage medium, the fault diagnosis method includes the following steps: obtaining the diagnostic parameter of machine to be diagnosed;The sample parameter of each sample machine in sample database is matched with the diagnostic parameter, and the matching result of each sample machine with the machine to be diagnosed is obtained;According to the matching result, the sample machine matched with the machine to be diagnosed is determined;The fault information of the sample machine is obtained, and the diagnostic result of the machine to be diagnosed is determined according to the fault information of the sample machine, wherein the diagnostic result includes the problem component of the machine to be diagnosed, problem influencing factor and / or the influence weight of the problem influencing factor.The present application can solve the problem of low energy efficiency stability of machine equipment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fault diagnosis, in particular to a fault diagnosis method, device and computer readable storage medium. BACKGROUND

[0002] Many machine devices, such as air conditioners, in the product design and development process, considering the factors of incoming materials, processes, parts, etc., will reserve a certain design margin to prevent the problem of inconsistent product quality in the same batch caused by the bearing range error in the production process. However, in the product trial production, trial production or even mass production stage, there are still problems such as insufficient running margin, large fluctuation of energy efficiency test value, which affect the stability of product performance. In this case, since the problem parts of the machine device are not known, it is not possible to provide theoretical guidance and optimization direction to improve the energy efficiency stability of the machine device, resulting in low energy efficiency stability of the machine device. SUMMARY

[0003] The main purpose of the present application is to provide a fault diagnosis method, device and computer readable storage medium, which aims to solve the problem of low energy efficiency stability of machine devices.

[0004] To achieve the above purpose, the present application provides a fault diagnosis method, which comprises:

[0005] obtaining the diagnosis parameters of the machine to be diagnosed;

[0006] matching the diagnosis parameters with the sample parameters of each sample machine in the sample database to obtain the matching results of the machine to be diagnosed and each sample machine;

[0007] determining the sample machine matched with the machine to be diagnosed according to the matching results;

[0008] obtaining the fault information of the sample machine, and determining the diagnosis result of the machine to be diagnosed according to the fault information, wherein the diagnosis result includes the problem parts of the machine to be diagnosed, the problem influencing factors and / or the influence weight of the problem influencing factors.

[0009] Optionally, the step of matching the diagnosis parameters with the sample parameters of each sample machine in the sample database to obtain the matching results of the machine to be diagnosed and each sample machine comprises:

[0010] obtaining the parameter level of the diagnosis parameters, wherein the parameter level is the relative level of the actual value of the diagnosis parameters relative to the reference value of the diagnosis parameters;

[0011] determining whether the parameter level of the diagnosis parameter is same as the parameter level of the sample parameter of each of the sample machines, to determine the matching result of the machine to be diagnosed and each of the sample machines.

[0012] Optionally, the step of obtaining the parameter level of the diagnosis parameter comprises:

[0013] obtaining a reference value of the diagnosis parameter;

[0014] obtaining a ratio of an actual value of the diagnosis parameter to the reference value;

[0015] determining the parameter level of the diagnosis parameter according to the ratio.

[0016] Optionally, the diagnosis parameter comprises an input parameter and an output parameter of the machine to be diagnosed, and the sample parameter comprises an input parameter and an output parameter of the sample machine, and the step of determining whether the parameter level of the diagnosis parameter is same as the parameter level of the sample parameter of each of the sample machines, to determine the matching result of the machine to be diagnosed and each of the sample machines comprises:

[0017] determining whether the parameter level of the input parameter of the machine to be diagnosed is same as the parameter level of the same input parameter of each of the sample machines, to determine a first matching result of the machine to be diagnosed and each of the sample machines;

[0018] determining whether the parameter level of the output parameter of the machine to be diagnosed is same as the parameter level of the same output parameter of each of the sample machines, to determine a second matching result of the machine to be diagnosed and each of the sample machines;

[0019] determining the matching result of the machine to be diagnosed and each of the sample machines according to the first matching result and the second matching result.

[0020] Optionally, the step of obtaining the diagnosis parameter of the machine to be diagnosed comprises:

[0021] obtaining an input parameter of the machine to be diagnosed;

[0022] inputting the input parameter of the machine to be diagnosed into a target neural network model for prediction, to obtain an output parameter of the machine to be diagnosed;

[0023] determining the input parameter and the output parameter as the diagnosis parameter of the machine to be diagnosed.

[0024] Optionally, the step of obtaining the input parameter of the machine to be diagnosed further comprises:

[0025] obtaining a preset neural network model;

[0026] inputting an input parameter of the training machine in the training database into the preset neural network model to obtain an actual output value of an output parameter of the training machine;

[0027] obtaining an expected output value of the output parameter of the training machine;

[0028] when the actual output value is equal to the expected output value, determining the preset neural network model as the target neural network model.

[0029] Optionally, the step of obtaining the expected output value of the output parameter of the training machine further comprises:

[0030] when the actual output value is not equal to the expected output value, obtaining an error value between the expected output value and the actual output value;

[0031] correcting a weight of a network node bias in the preset neural network model according to the error value;

[0032] updating the preset neural network model into the corrected neural network model, and returning to execute the step of obtaining the preset neural network model.

[0033] Optionally, the step of obtaining the fault information of the sample machine and determining the diagnosis result of the machine to be diagnosed according to the fault information of the sample machine, wherein the diagnosis result comprises a problem component of the machine to be diagnosed, a problem influencing factor and / or an influence weight of the problem influencing factor further comprises:

[0034] associating the diagnosis parameter of the machine to be diagnosed with the diagnosis result;

[0035] storing the associated diagnosis parameter of the machine to be diagnosed and the diagnosis result into the machine sample database.

[0036] In addition, to achieve the above object, the present application further provides a fault diagnosis device, which comprises a memory, a processor and a fault diagnosis program stored in the memory and executable on the processor, and the fault diagnosis program realizes the steps of the fault diagnosis method as described above when executed by the processor.

[0037] In addition, to achieve the above object, the present application further provides a computer readable storage medium, which stores a fault diagnosis program, and the fault diagnosis program realizes the steps of the fault diagnosis method as described above when executed by a processor.

[0038] The application provides a fault diagnosis method, device and computer readable storage medium, wherein the diagnosis parameter of a machine to be diagnosed is acquired, the diagnosis parameter is matched with the sample parameter of each sample machine in a sample database, the matching result of the machine to be diagnosed and each sample machine is obtained, then the sample machine matched with the machine to be diagnosed is determined according to the matching result, the fault information of the sample machine is acquired, and the diagnosis result of the machine to be diagnosed is determined according to the fault information of the sample machine, wherein the diagnosis result comprises the problem component of the machine to be diagnosed, the problem influencing factor and / or the influence weight of the problem influencing factor. In this way, the problem component of the machine to be diagnosed with unknown problems can be traced, theoretical guidance and optimization direction for improving the energy efficiency stability of the machine to be diagnosed are provided, and the problem of low energy efficiency stability of the machine is effectively solved. BRIEF DESCRIPTION OF DRAWINGS

[0039] The implementation, functional features and advantages of the application will be further described with reference to the embodiments and the accompanying drawings.

[0040] Figure 1 The hardware architecture of the fault diagnosis device involved in the embodiment of the application is shown in the figure.

[0041] Figure 2 The flowchart of the first embodiment of the fault diagnosis method of the application is shown in the figure.

[0042] Figure 3 The flowchart of the second embodiment of the fault diagnosis method of the application is shown in the figure.

[0043] Figure 4 The flowchart of the third embodiment of the fault diagnosis method of the application is shown in the figure.

[0044] Figure 5 The flowchart of the fourth embodiment of the fault diagnosis method of the application is shown in the figure.

[0045] Figure 6 The flowchart of the fifth embodiment of the fault diagnosis method of the application is shown in the figure.

[0046] Figure 7 The flowchart of the sixth embodiment of the fault diagnosis method of the application is shown in the figure.

[0047] Figure 8 The input parameter table of the air conditioner X / Y to be diagnosed involved in the embodiment of the application is shown in the figure.

[0048] Figure 9 The rule diagram of the matching degree sorting involved in the embodiment of the application is shown in the figure.

[0049] Figure 10 The diagnosis result table of the air conditioner X / Y to be diagnosed involved in the embodiment of the application is shown in the figure.

[0050] Figure 11 The energy efficiency data table before and after X / Y diagnosis of the air conditioner to be diagnosed involved in the embodiment of the present application;

[0051] Figure 12 The BP neural network model schematic diagram involved in the embodiment of the present application. DETAILED DESCRIPTION

[0052] It should be understood that the specific embodiments described herein merely exemplify the present application and do not limit the present application.

[0053] In order to better understand the above technical solutions, the exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0054] In order to better understand the above technical solutions, the above technical solutions will be described in detail below in combination with the drawings and specific embodiments. Referring to Figure 1 , Figure 1 is a hardware architecture schematic diagram of the fault diagnosis device involved in the embodiment of the present application. As Figure 1 shown, the fault diagnosis device can include a processor 1001, which can be a CPU, a network interface 1004, a user interface 1003, a memory 1005, and a communication bus 1002. The communication bus 1002 is used to realize the connection and communication between these components. The network interface 1004 can optionally include a standard wired interface, a wireless interface (such as non-volatile memory), for example, a disk memory. The memory 1005 can also be a storage device independent of the aforementioned processor 1001.

[0055] Those skilled in the art can understand that Figure 1 the structure of the fault diagnosis device shown in the above

[0056] As Figure 1 shown, the memory 1005, as a computer storage medium, can include an operating system and a fault diagnosis program.

[0057] In Figure 1 the fault diagnosis device shown, the processor 1001 can be used to call the fault diagnosis program stored in the memory 1005 and perform the following operations:

[0058] obtaining a diagnosis parameter of a machine to be diagnosed;

[0059] matching the diagnosis parameter with sample parameters of each sample machine in a sample database to obtain a matching result of the machine to be diagnosed and each sample machine;

[0060] determining a sample machine matched with the machine to be diagnosed according to the matching result;

[0061] obtaining fault information of the sample machine, and determining a diagnosis result of the machine to be diagnosed according to the fault information of the sample machine, wherein the diagnosis result includes a problem component of the machine to be diagnosed, a problem influencing factor and / or an influence weight of the problem influencing factor.

[0062] Referring to Figure 2 , Figure 2 is a flowchart of a first embodiment of the fault diagnosis method of the present application, and the fault diagnosis method comprises the following steps:

[0063] Step S10, obtaining a diagnosis parameter of a machine to be diagnosed;

[0064] In this embodiment, the energy efficiency stability of the machine device refers to the performance stability of the machine device in the running process under all-year and all-working conditions, and is usually taken as an index for evaluating the performance of the machine device. The higher the energy efficiency stability is, the better the performance of the machine device is, and vice versa. Generally, the machine device has an energy efficiency standard evaluation index, and the energy efficiency standard evaluation index is different in different fields and different devices. The energy efficiency standard evaluation index usually has a qualified standard, i.e. a qualified numerical range of the energy efficiency test value of the machine device. When the energy efficiency test value of the machine device fluctuates within the qualified standard, it indicates that the energy efficiency stability of the machine device is high, and the performance of the machine device is good. Conversely, when the energy efficiency test value of the machine device fluctuates greatly, especially when the energy efficiency test value of the machine device is not within the qualified standard, it indicates that the energy efficiency stability of the machine device is low, and the performance of the machine device is poor.

[0065] In the present embodiment, taking an air conditioner as an example, in the field of air conditioners, the new standard GB21455-2019 stipulates that the energy efficiency label value of a room air conditioner product should be within the value range corresponding to its rated energy efficiency level, the actual measured value of its energy efficiency should be not less than 95% of the label value, and the rated cooling capacity labeled by the product and its actual measured value should be within the rated cooling capacity range corresponding to its rated energy efficiency level. Generally, the energy efficiency standard evaluation index of an air conditioner is APF (Annual Performance Factor, annual energy consumption efficiency). The energy efficiency standard evaluation index of an air conditioner, APF, takes into account both the refrigeration capacity of the air conditioner and the heating factor, which changes the previous energy efficiency index of variable frequency air conditioners, which only evaluates the energy consumption of air conditioners in the refrigeration season. The energy efficiency standard evaluation index of an air conditioner, APF, evaluates the energy consumption level of the whole year, and the evaluation of air conditioner performance is more comprehensive. The qualified standard of the energy efficiency standard evaluation index of an air conditioner, APF, is 3.822-3.978. According to the qualified standard, when the energy efficiency test value of the air conditioner fluctuates within the range of 3.822-3.978, i.e. only within the range of the qualified standard of APF, it indicates that the energy efficiency stability of the air conditioner is high, and the performance of the air conditioner is good. On the contrary, when the energy efficiency test value of the air conditioner fluctuates greatly, especially when the energy efficiency test value of the air conditioner deviates from the range of 3.822-3.978, i.e. deviates from the range of the qualified standard of APF, it indicates that the energy efficiency stability of the air conditioner is low, and the performance of the air conditioner is poor.

[0066] In the present embodiment, in order to ensure the energy efficiency stability of machine equipment products, the design and development personnel will consider various possible factors in the design and development process of the product, including the incoming materials, processes and equipment parts of the machine equipment product, in order to reserve a certain design allowance. This can prevent the problem of inconsistent products in the same batch and low energy efficiency stability of machine equipment products caused by load range error in the production process of machine equipment products to some extent. However, in the actual production process of machine equipment products, there will be more or less problems of large fluctuation of energy efficiency test value of machine equipment products and low energy efficiency stability of machine equipment products. When such problems occur, it indicates that the parts of the machine equipment may have failed, and the fault of the machine equipment parts needs to be checked to determine the problem parts and the causes of the low energy efficiency stability of the machine equipment products, and then the problem parts are replaced or optimized to improve the energy efficiency stability of the machine equipment products and the performance of the machine equipment. However, due to the lack of effective ways to trace the problem parts, it is difficult to comprehensively analyze the factors affecting the energy efficiency stability of the machine equipment, so it is impossible to provide theoretical guidance and optimization direction to improve the energy efficiency stability of the machine equipment products, which makes the energy efficiency stability of the produced machine equipment products still low.

[0067] To solve the above problems, the fault diagnosis method provided by the application establishes a sample database of the problem machine, wherein the sample database is composed of a certain number of sample machines of the same type of machine equipment as the machine to be diagnosed, but the energy efficiency test value is not within the qualified standard and the parts of the equipment have faults, the sample database includes the sample parameters and the fault information of each sample machine whose energy efficiency test value is not within the qualified standard and the parts of the equipment have faults, and the sample parameters of each sample machine whose parts have faults are associated with the fault information thereof.

[0068] In this embodiment, the fault diagnosis method provided by the application first acquires the diagnosis parameters of the machine to be diagnosed whose energy efficiency test value is not within the qualified standard, compares and matches the diagnosis parameters of the machine to be diagnosed with the sample data stored in the pre-established sample database, determines the sample machines with better matching degrees in the sample database by comparison and matching, wherein the sample machines with better matching degrees can be one or more, and the specific number can be determined according to actual needs, then acquires the fault information of the sample machines with better matching degrees, the fault information at least includes the parts of the sample machines that have problems, i.e. the problem parts of the sample machines, and finally determines the parts of the machine to be diagnosed that have problems according to the parts of the sample machines with better matching degrees that have problems, thereby realizing the traceability of the problem parts of the machine to be diagnosed whose energy efficiency test value is not within the qualified standard, providing an improvement strategy for the research and development designers to improve the energy efficiency stability of the machine to be diagnosed, optimizing and updating the parts that have problems, and improving the energy efficiency stability of the machine equipment products in production. The application realizes the accurate connection from the numerical end of the machine to the problem end of the parts, improves the efficiency and accuracy of fault diagnosis, and saves the additional labor and experimental costs required by the traditional diagnosis method.

[0069] In this embodiment, the fault diagnosis method provided by the application can be applied to a fault diagnosis device, which is a device or apparatus that can diagnose the fault of the machine to be diagnosed, and the fault diagnosis device can be selected as a terminal device, wherein the terminal device is a device that inputs programs and data to a computer or receives the processing results output by the computer via a communication facility, and the terminal device can be selected as a computer, of course, in other embodiments, the fault diagnosis device can also be other devices that can perform fault diagnosis, and in the field of industrial control, the fault diagnosis device can be selected as a PLC industrial control device, and this embodiment does not limit it.

[0070] In this embodiment, the machine to be diagnosed is a machine equipment with low energy efficiency stability, whose energy efficiency test value is not within the qualified standard and the parts of the equipment have faults.

[0071] In the embodiment, the machine to be diagnosed is taken as an air conditioner for example, the air conditioner to be diagnosed is an air conditioner with low energy efficiency stability, the energy efficiency test value of which is not within the qualified standard (3.822-3.978), and the air conditioner to be diagnosed is an air conditioner with a fault in a component. Of course, in other embodiments, the machine to be diagnosed can be determined according to actual conditions, and the embodiment is not limited in this regard.

[0072] In the embodiment, the diagnosis parameter is relevant working condition data used for fault diagnosis of the machine to be diagnosed, and generally includes an operating parameter of the machine to be diagnosed under a relevant working condition, the operating parameter at least including an input parameter and an output parameter. It should be noted that in specific embodiments, the diagnosis parameter can be determined according to specific conditions.

[0073] In the embodiment, the machine to be diagnosed is taken as an air conditioner for example, the diagnosis parameter of the air conditioner to be diagnosed is an operating parameter of the air conditioner to be diagnosed under a relevant working condition, including an input parameter and an output parameter of the air conditioner to be diagnosed, wherein the input parameter at least includes five parameters of working condition capacity, power, indoor unit air volume, indoor outlet dry-bulb temperature and indoor outlet wet-bulb temperature of the air conditioner to be diagnosed; and the output parameter at least includes seven parameters of return air temperature, discharge air temperature, indoor heat exchanger inlet temperature, indoor heat exchanger middle temperature, indoor heat exchanger outlet temperature, outdoor heat exchanger middle temperature and outdoor heat exchanger outlet temperature of the air conditioner. Of course, in other embodiments, according to different machines to be diagnosed, the diagnosis parameter will also be different according to actual conditions, and the embodiment is not limited in this regard.

[0074] In the embodiment, the fault diagnosis device can automatically acquire the diagnosis parameter of the machine to be diagnosed used for fault diagnosis.

[0075] Optionally, the fault diagnosis device obtains diagnosis parameters of the machine to be diagnosed for fault diagnosis through sensors, the machine to be diagnosed is provided with sensors corresponding to the input parameters, the sensors are in communication connection with the fault diagnosis device, the input parameters are collected through the sensors, and the collected input parameters are sent to the fault diagnosis device. The fault diagnosis device receives the input parameters sent by the sensors and inputs the received input parameters into the neural network model that is pre-established and trained to obtain output parameters. The input parameters and the output parameters of the machine to be diagnosed are taken as diagnosis parameters of the machine to be diagnosed for fault diagnosis. The neural network model can be a BP (back propagation) neural network model. The BP (back propagation) neural network model is a multi-layer feedforward network trained by error back propagation (error back propagation for short). The algorithm is called BP algorithm. The basic idea is gradient descent method. Gradient search technology is used to minimize the mean square error of the actual output value and the expected output value of the network. The BP (back propagation) network increases a plurality of layers (one layer or more) of neurons between the input layer and the output layer. These neurons are called hidden units. They have no direct contact with the outside world, but their state changes can affect the relationship between the input and the output. Each layer can have several nodes. The calculation process of the BP (back propagation) neural network consists of a forward calculation process and a backward calculation process. In the forward propagation process, the input mode is processed layer by layer from the input layer to the hidden unit layer, and is transferred to the output layer. The state of each layer of neurons only affects the state of the next layer of neurons. If the expected output cannot be obtained in the output layer, the error signal is returned along the original connection path, and the weights of each neuron are modified to minimize the error signal. Of course, in other embodiments, the neural network model can be determined according to the actual situation, and this embodiment does not limit it. It should be noted that the input parameters are collected through the sensors, and then the output parameters are obtained based on the collected input parameters and the pre-established and trained neural network model. On the one hand, the number of sensors in the machine to be diagnosed can be reduced, saving hardware resources. On the other hand, the efficiency and accuracy of obtaining diagnosis parameters of the machine to be diagnosed can be improved. It can be understood that diagnosis parameters that cannot be collected by sensors can also be calculated through manual experiments.

[0076] Optionally, the fault diagnosis device obtains all diagnosis parameters (including input parameters and output parameters) of the machine to be diagnosed through sensors. The machine to be diagnosed is provided with sensors corresponding to the input parameters and the output parameters. The sensors are connected with the fault diagnosis device. The input parameters and the output parameters are collected through the sensors, and the collected input parameters and output parameters are sent to the fault diagnosis device. It can be understood that diagnosis parameters that cannot be collected by sensors can also be calculated through manual experiments.

[0077] In the embodiment, taking the air conditioner as an example, the input parameters of the air conditioner to be diagnosed are obtained by the fault diagnosis device through sensors and manual experimental calculation. Specifically, the working condition capacity and power of the air conditioner to be diagnosed are obtained through manual experimental calculation. At the same time, the indoor unit air volume sensor, the indoor outlet air dry bulb temperature sensor and the indoor outlet air wet bulb temperature sensor are arranged on the air conditioner to be diagnosed. The indoor unit air volume, the indoor outlet air dry bulb temperature and the indoor outlet air wet bulb temperature of the air conditioner are collected by the above sensors, and the collected indoor unit air volume, the indoor outlet air dry bulb temperature and the indoor outlet air wet bulb temperature of the air conditioner to be diagnosed are sent to the fault diagnosis device. The fault diagnosis device receives the indoor unit air volume, the indoor outlet air dry bulb temperature and the indoor outlet air wet bulb temperature of the air conditioner to be diagnosed collected by the sensors. At the same time, the fault diagnosis device receives the working condition capacity and power of the air conditioner to be diagnosed obtained by manual experimental calculation and input. Refer to Figure 8 , Figure 8 The input parameter table of the air conditioner X / Y to be diagnosed involved in the embodiment of the present application. After obtaining the input parameters of the air conditioner, the fault diagnosis device inputs the input parameters of the air conditioner into the BP (back propagation) neural network model established and trained in advance. Since the BP (back propagation) neural network model establishes the neural network correlation between the input parameters and the output parameters of the air conditioner to be diagnosed, after the input parameters of the air conditioner to be diagnosed are input into the BP (back propagation) neural network model established and trained in advance, the output parameters of the air conditioner to be diagnosed can be obtained, that is, the return air temperature, the exhaust air temperature, the indoor heat exchanger inlet temperature, the indoor heat exchanger middle temperature, the indoor heat exchanger outlet temperature, the outdoor heat exchanger middle temperature and the outdoor heat exchanger outlet temperature of the air conditioner to be diagnosed are obtained. Then, the input parameters of the air conditioner to be diagnosed and the output parameters obtained according to the input parameters and the BP (back propagation) neural network model are determined as the diagnosis parameters of the air conditioner to be diagnosed.

[0078] In the embodiment, in the process of acquiring the diagnosis parameters of the air conditioner to be diagnosed for fault diagnosis, the diagnosis parameters can also be acquired completely by manual test calculation and by setting sensors. Specifically, the working condition capacity and power of the air conditioner to be diagnosed are obtained by manual test calculation, indoor unit air volume sensors, indoor outlet air dry-bulb temperature sensors, indoor outlet air wet-bulb temperature sensors, return air temperature sensors, exhaust air temperature sensors, indoor heat exchanger inlet temperature sensors, indoor heat exchanger middle temperature sensors, indoor heat exchanger outlet temperature sensors, outdoor heat exchanger middle temperature sensors and outdoor heat exchanger outlet temperature sensors are set on the air conditioner to be diagnosed, and the indoor unit air volume, indoor outlet air dry-bulb temperature, indoor outlet air wet-bulb temperature, return air temperature, exhaust air temperature, indoor heat exchanger inlet temperature, indoor heat exchanger middle temperature, indoor heat exchanger outlet temperature, outdoor heat exchanger middle temperature and outdoor heat exchanger outlet temperature of the air conditioner are collected by the above sensors, and then the related parameters of the air conditioner to be diagnosed acquired by the sensors and the related parameters obtained by manual test calculation are determined as the diagnosis parameters of the air conditioner to be diagnosed.

[0079] In step S20, the diagnosis parameters are matched with the sample parameters of each sample machine in the sample database, to obtain the matching results of the machine to be diagnosed and each sample machine.

[0080] In the embodiment, the sample machine refers to a machine device of the same type as the machine to be diagnosed, but the energy efficiency test value of which is not within the qualified standard of the machine device, and the components of which have faults. The sample parameters refer to the working condition data of the sample machine under the relevant working conditions, that is, the running data under the relevant working conditions. The sample parameters of each sample machine are of the same parameter type as the diagnosis parameters, and the number of parameters is also the same. The sample database is a collection of a certain number of sample machines. The sample database stores the sample parameters and fault information of all sample machines, and the sample parameters of the sample machines are associated with the fault information thereof.

[0081] In the embodiment, the sample air conditioner is also taken as an example that the energy efficiency test value is not within the qualified standard and the components of the sample air conditioner fail. The sample parameters of the sample air conditioner whose energy efficiency test value is not within the qualified standard are the working condition data of the sample air conditioner under the related working conditions, that is, the running data of the sample air conditioner under the related working conditions, specifically including the input parameters of the sample air conditioner and the output parameters of the sample air conditioner, and the specific parameters are completely the same as the types and quantities of the diagnostic parameters of the air conditioner to be diagnosed as the machine to be diagnosed, and also include the working condition capacity, power, indoor fan volume, indoor outlet dry bulb temperature, indoor outlet wet bulb temperature, return air temperature, exhaust air temperature, indoor heat exchanger inlet temperature, indoor heat exchanger middle temperature, indoor heat exchanger outlet temperature, outdoor heat exchanger middle temperature, and outdoor heat exchanger outlet temperature of the sample air conditioner. The sample database is a collection of a certain number of sample air conditioners, and the sample database stores the sample parameters of all sample air conditioners and the associated fault information, wherein the fault information includes the fault information of the indoor unit test and the outdoor unit test of the sample air conditioner, and specifically can be the problem components of the sample air conditioner that fail, including the heat exchanger of the indoor unit test, the air duct of the indoor unit test, the heat exchanger of the outdoor unit test, the compressor of the outdoor unit test, and the heat exchanger.

[0082] In the embodiment, the sample database is established in the fault diagnosis device, and after the diagnostic parameters of the machine to be diagnosed are obtained, the diagnostic parameters are compared with the sample parameters of each sample machine stored in the sample database one by one to obtain the matching result of the machine to be diagnosed and each sample machine.

[0083] Optionally, each diagnostic parameter of the machine to be diagnosed is compared with the same parameter in the sample parameters of a certain sample machine to obtain the matching result of each diagnostic parameter and the same parameter in the sample parameters of the sample machine, and the matching results of each diagnostic parameter and the same parameter in the sample parameters of the sample machine are integrated to obtain the matching result of the machine to be diagnosed and the sample machine. The above process is repeated to obtain the matching result of the machine to be diagnosed and all other sample machines in the sample database.

[0084] In the embodiment, the diagnostic parameters of the machine to be diagnosed can be divided into three parameter levels, specifically including level 1 / level 2 / level 3, wherein the parameter level refers to the relative level of the actual value of the diagnostic parameter relative to the reference value thereof, which is usually obtained by ratio and expressed in percentage, and the reference value refers to the normal value of the diagnostic parameter under the relevant working condition when the energy efficiency test value of the machine to be diagnosed is within the qualified standard and the components do not fail. The diagnostic parameters of the machine to be diagnosed are compared with the corresponding reference values, and the parameter levels corresponding to the diagnostic parameters are obtained by calculating the ratio of the actual value of the diagnostic parameter to the reference value thereof. The parameter level of the diagnostic parameter is specifically determined according to the following calculation method: level 1>101% reference value; 101% reference value≥level 2≥99% reference value; 99% reference value>level 3.

[0085] Similarly, each sample parameter of the sample machine in the sample database also has a sample parameter level, and the determination method of the sample parameter level of the sample parameter is exactly the same as that of the parameter level of the diagnostic parameter, which can be referred to the above content, and the embodiment will not be described here.

[0086] Further, the parameter level of each diagnostic parameter of the machine to be diagnosed is compared and matched with the parameter level of the same parameter of a sample machine, to obtain the comparison result of the parameter level of each diagnostic parameter and the parameter level of the same parameter of the sample machine, and the comparison results of the parameter levels of the diagnostic parameters and the parameter levels of the same parameters of the sample machine are integrated to obtain the matching result of the machine to be diagnosed and the sample machine.

[0087] Optionally, the matching result of the machine to be diagnosed and the sample machine is represented by red and blue ball data, wherein the matching result of the parameter level of the output parameter of the machine to be diagnosed and the parameter level of the same output parameter in the sample data of a sample machine is represented by the number of matching red balls, and the matching result of the parameter level of the input parameter in the diagnostic parameter and the parameter level of the same input parameter in the sample data of a sample machine is represented by the number of matching blue balls.

[0088] In the embodiment, the parameter level of each input parameter of the machine to be diagnosed is compared with the parameter level of the same input parameter in the sample parameter of a sample machine, and it is determined whether they are the same. If they are the same, the matching blue ball number of the machine to be diagnosed and the sample machine is increased by 1. The parameter level of each output parameter of the machine to be diagnosed is compared with the parameter level of the same output parameter in the sample parameter of a sample machine, and it is determined whether they are the same. If they are the same, the matching red ball number of the machine to be diagnosed and the sample machine is increased by 1. After the diagnosis parameters of the machine to be diagnosed and the sample parameters of a sample machine are completely matched, the final red ball data and blue ball data are taken as the matching result of the machine to be diagnosed and the sample machine. For example, the final comparison matching result of the machine to be diagnosed and a sample machine is 4 blue balls and 5 red balls, and the matching result is 4 blue balls and 5 red balls. The above process is repeated, and the matching results of the machine to be diagnosed and other sample machines can be obtained.

[0089] In the embodiment, taking an air conditioner as an example, the input parameters of the air conditioner to be diagnosed include 5 parameters of the working condition capacity, power, indoor unit air volume, indoor outlet air dry-bulb temperature and indoor outlet air wet-bulb temperature of the air conditioner to be diagnosed; and the output parameters include 7 parameters of the return air temperature, discharge air temperature, indoor heat exchanger inlet temperature, indoor heat exchanger middle temperature, indoor heat exchanger outlet temperature, outdoor heat exchanger middle temperature and outdoor heat exchanger outlet temperature of the air conditioner to be diagnosed.

[0090] In the embodiment, in the sample database, the sample parameters of each sample air conditioner also include 5 parameters of the working condition capacity, power, indoor unit air volume, indoor outlet air dry-bulb temperature and indoor outlet air wet-bulb temperature of the sample air conditioner; and the output parameters include 7 parameters of the return air temperature, discharge air temperature, indoor heat exchanger inlet temperature, indoor heat exchanger middle temperature, indoor heat exchanger outlet temperature, outdoor heat exchanger middle temperature and outdoor heat exchanger outlet temperature of the air conditioner.

[0091] In the embodiment, firstly, the parameter levels corresponding to the 12 diagnostic parameters of the air conditioner to be diagnosed and the parameter levels corresponding to the 12 parameters of the sample air conditioner are calculated according to the above parameter level determination manner, the parameter level of each input parameter of the air conditioner to be diagnosed is compared with the parameter level of the same parameter in the sample parameters of a sample air conditioner, whether the parameter levels are the same is determined, if the parameter levels are the same, the matching blue ball number of the air conditioner to be diagnosed and the sample air conditioner is added by 1, after the comparison of all the input parameters of the air conditioner to be diagnosed is completed, the total matching blue ball number of the air conditioner to be diagnosed and the sample air conditioner is obtained; the parameter level of each output parameter of the air conditioner to be diagnosed is compared with the parameter level of the same parameter in the sample parameters of a sample air conditioner, whether the parameter levels are the same is determined, if the parameter levels are the same, the matching red ball number of the air conditioner to be diagnosed and the sample air conditioner is added by 1, after the comparison of all the output parameters of the air conditioner to be diagnosed is completed, the total matching red ball number of the air conditioner to be diagnosed and the sample air conditioner is obtained, the total matching blue ball number and the total matching red ball number are determined as the matching result of the air conditioner to be diagnosed and the sample air conditioner, for example, the total matching blue ball number is 4 and the total matching red ball number is 5, so the matching result of the air conditioner to be diagnosed and the sample air conditioner is 4 blue balls and 5 red balls. The above process is repeated, and the matching results of the air conditioner to be diagnosed and other sample air conditioners can be obtained.

[0092] Alternatively, when the matching result of the machine to be diagnosed and the sample machine is obtained, all the diagnostic parameters of the machine to be diagnosed can be taken as a whole to determine the matching index of the machine to be diagnosed, and all the sample parameters of each sample machine can be taken as a whole to determine the matching index of the sample machine, the matching index of the machine to be diagnosed is compared with the matching index of each sample machine, and the matching result of the machine to be diagnosed and each sample machine is obtained according to the comparison result.

[0093] In the embodiment, taking the air conditioner as an example, a matching index A is obtained according to the working condition capacity, power, indoor unit air volume, indoor outlet air dry-bulb temperature, indoor outlet air wet-bulb temperature, return air temperature, exhaust air temperature, indoor heat exchanger inlet temperature, indoor heat exchanger middle temperature, indoor heat exchanger outlet temperature, outdoor heat exchanger middle temperature and outdoor heat exchanger outlet temperature of the air conditioner to be diagnosed; and a matching index B is obtained according to the working condition capacity, power, indoor unit air volume, indoor outlet air dry-bulb temperature, indoor outlet air wet-bulb temperature, return air temperature, exhaust air temperature, indoor heat exchanger inlet temperature, indoor heat exchanger middle temperature, indoor heat exchanger outlet temperature, outdoor heat exchanger middle temperature and outdoor heat exchanger outlet temperature of the sample air conditioner. The two matching indexes A and B are compared, and the matching result of the air conditioner to be diagnosed and the sample air conditioner is obtained.

[0094] Step S30, determining a sample machine matched with the machine to be diagnosed according to the matching result;

[0095] The fault diagnosis device selects a sample machine with better matching degree with the machine to be diagnosed according to each matching result after obtaining the matching result of the machine to be diagnosed with each sample machine. The sample machine with better matching degree with the machine to be diagnosed can be one or two. Of course, in other embodiments, the sample machine with better matching degree with the machine to be diagnosed can also be more than two, which can be determined according to actual needs. The embodiment does not limit this. It should be noted that generally, the more the number of sample machines with better matching degree with the machine to be diagnosed, the more accurate the diagnosis result after subsequent weighted calculation.

[0096] Specifically, the matching degrees of each matching result are sorted, the sample machine with the highest matching degree is determined as the sample machine with better matching degree with the machine to be diagnosed, or the sample machines with the highest and the second highest matching degrees are determined as the sample machines with better matching degree with the machine to be diagnosed. The rule of the sorting of the matching degrees is that the matching blue ball numbers in each matching result are sorted from high to low in priority, wherein the higher the matching blue ball number, the higher the matching degree of the machine to be diagnosed with the sample machine. When the matching blue ball numbers are the same, the matching red ball numbers are sorted from high to low, and the more the matching red ball numbers, the higher the matching degree. Referring to Figure 9 , Figure 9 The figure is a rule of the sorting of the matching degrees involved in the embodiment of the application. The solid ball represents the red ball and the hollow ball represents the blue ball.

[0097] In this embodiment, taking the machine to be diagnosed as an air conditioner as an example, the matching blue ball numbers in the matching results of the air conditioner to be diagnosed and each sample air conditioner are sorted from high to low in priority, wherein the higher the matching blue ball number, the higher the matching degree of the air conditioner to be diagnosed with the sample air conditioner. When the matching blue ball numbers are the same, the matching red ball numbers in the matching results of the air conditioner to be diagnosed and each sample air conditioner are sorted from high to low, and the more the matching red ball numbers, the higher the matching degree of the air conditioner to be diagnosed with the sample air conditioner. According to this sorting rule of the matching degrees, the sample air conditioner with better matching degree with the air conditioner to be diagnosed is obtained and determined as the sample air conditioner matched with the air conditioner to be diagnosed, or the sample air conditioners with the highest and the second highest matching degrees with the air conditioner to be diagnosed are obtained and determined as the sample air conditioners with better matching degree with the air conditioner to be diagnosed.

[0098] Step S40, obtaining the fault information of the sample machine, and determining the diagnosis result of the machine to be diagnosed according to the fault information of the sample machine, wherein the diagnosis result includes the problem component of the machine to be diagnosed, the problem influencing factor and / or the influence weight of the problem influencing factor.

[0099] The fault information is the fault condition information of the sample machine, and the fault information includes the problem component of the sample machine, i.e., the component information of the fault, and the diagnosis result is the diagnosis condition of the machine to be diagnosed, including the component information of the fault of the machine to be diagnosed, the influence factor of the problem and / or the influence weight of the problem influence factor.

[0100] After the fault diagnosis device obtains the sample machine with better matching degree, the problem component of the sample machine with better matching degree is sorted and output, and the diagnosis result of the machine to be diagnosed is determined according to the sorted and output problem component.

[0101] Optionally, if the sample machine with better matching degree obtained by matching is only one, the problem component of the sample machine is sorted and output, and the problem component of the machine to be diagnosed is the problem component of the sample machine, and the diagnosis result is obtained by weighted calculation of each working condition.

[0102] Optionally, if the sample machine with better matching degree obtained by matching is two (i.e., the sample machine with the highest matching degree and the sample machine with the second highest matching degree), the problem components of the two sample machines are sorted and output, and the diagnosis result is obtained by weighted calculation of each working condition. It should be noted that the sample machine with better matching degree obtained by matching can also be more than two, which can improve the accuracy of the diagnosis result.

[0103] Further, after obtaining the diagnosis result, the fault diagnosis device outputs the diagnosis result, and the research and development designers can optimize and adjust the machine to be diagnosed according to the diagnosis result. The research and development designers provide theoretical guidance and optimization direction for improving the energy efficiency stability of the machine to be diagnosed, and the optimization and adjustment mode can be to replace the problem component so that the energy efficiency test value of the machine to be diagnosed is within the qualified standard.

[0104] Further, after obtaining the diagnosis result, the fault diagnosis device associates the diagnosis parameter with the diagnosis result, and stores the associated diagnosis parameter and diagnosis result to the sample database, so that the sample database has the functions of self-updating and self-expanding.

[0105] In the embodiment, the machine to be diagnosed is an air conditioner, and after the fault diagnosis device obtains the sample air conditioner matched with the air conditioner to be diagnosed, the fault diagnosis device sorts and outputs the problem component of the sample air conditioner, i.e., the problem component, and obtains the problem component of the air conditioner to be diagnosed, the influence factor of the problem and / or the influence weight of the problem influence factor by weighted calculation of each working condition. Figure 10 , Figure 10 The diagnosis result table of the air conditioner X / Y to be diagnosed is involved in the embodiment of the present application.

[0106] Optionally, if there is only one sample air conditioner that matches the air conditioner to be diagnosed well, then the problematic component of that sample air conditioner is sorted out and output. The problematic component of the air conditioner to be diagnosed is the problematic component of that sample air conditioner. The problematic component of the air conditioner to be diagnosed, the influencing factors of the problem, and / or the influence weight of the influencing factors are obtained by weighted calculation under various operating conditions.

[0107] Optionally, if there are two sample air conditioners that match the air conditioner to be diagnosed well (i.e., the highest and second highest matching degree mentioned above), then the problematic components of these two sample air conditioners are identified, and the problematic components of the air conditioner to be diagnosed, the influencing factors of the problem, and / or the influence weight of the influencing factors of the problem are obtained through weighted calculation under each operating condition.

[0108] Furthermore, after obtaining the diagnostic results from the air conditioner under test, the fault diagnosis device outputs these results. Based on these results, R&D personnel can optimize and adjust the air conditioner. Optimization and adjustment can include replacing faulty components to ensure the energy efficiency test value of the machine under test is within the acceptable standard. (Refer to...) Figure 11 , Figure 11 This is a table showing the energy efficiency data of the air conditioner before and after X / Y diagnosis, as per an embodiment of the present invention. Figure 11 As can be seen, after optimization based on the diagnostic results, the energy efficiency test value of the air conditioner under diagnosis is within the qualified standard, which effectively improves the energy efficiency stability of the air conditioner under diagnosis.

[0109] In the technical solution provided in this embodiment, the fault diagnosis device obtains the diagnostic parameters of the machine to be diagnosed, matches the diagnostic parameters with the sample parameters of each sample machine in the sample database, obtains the matching results between the machine to be diagnosed and each sample machine, then determines the sample machine that matches the machine to be diagnosed based on the matching results, finally obtains the fault information of the sample machine, and determines the diagnosis result of the machine to be diagnosed based on the fault information of the sample machine. In this way, this solution uses the sample database to trace the source of the problem of the problematic component of the machine to be diagnosed with an unknown problem, providing theoretical guidance and optimization direction for improving the energy efficiency stability of the machine to be diagnosed, and effectively solving the problem of low energy efficiency stability of machine equipment.

[0110] Reference Figure 3 , Figure 3 This is a flowchart illustrating the second embodiment of the fault diagnosis method of the present invention. Based on embodiment one, the steps of S20 above include:

[0111] Step S21: Obtain the parameter level corresponding to the diagnostic parameter, wherein the parameter level is the relative level of the actual value of the diagnostic parameter relative to the reference value of the diagnostic parameter;

[0112] Step S22, judging whether the parameter level of the diagnosis parameter is same as the parameter level of the sample parameter of each sample machine, to determine the matching result of the machine to be diagnosed and each sample machine.

[0113] In the embodiment, the diagnosis parameters of the machine to be diagnosed can be divided into three parameter levels, including level 1 / level 2 / level 3, wherein the parameter level refers to the relative level of the actual value of the diagnosis parameter relative to the reference value, which is usually obtained by ratio and expressed in percentage, and the reference value refers to the normal value of the diagnosis parameter under the related working condition when the energy efficiency test value of the machine to be diagnosed is within the qualified standard and the components do not fail. The diagnosis parameters of the machine to be diagnosed are compared with the corresponding reference values, and the parameter level of each diagnosis parameter is obtained by calculating the ratio of the actual value of the diagnosis parameter to the reference value. The parameter level of the diagnosis parameter is determined according to the following calculation method: level 1>101% reference value; 101% reference value≥level 2≥99% reference value; 99% reference value>level 3.

[0114] Similarly, each sample parameter of the sample machine in the sample database also has a sample parameter level, and the determination method of the sample parameter level of the sample parameter is exactly the same as that of the parameter level of the diagnosis parameter, which can be referred to the above content, and will not be repeated here.

[0115] Further, the parameter level of each diagnosis parameter of the machine to be diagnosed is compared with the parameter level of the same parameter of a sample machine, to obtain the comparison result of the parameter level of each diagnosis parameter and the parameter level of the same parameter of the sample machine, and the matching result of the machine to be diagnosed and the sample machine is obtained by comprehensively considering the comparison results of the parameter levels of the diagnosis parameters and the parameter levels of the same parameters of the sample machine.

[0116] Optionally, the matching result of the machine to be diagnosed and the sample machine is represented by red and blue ball data, wherein the matching result of the parameter level of the output parameter of the machine to be diagnosed and the parameter level of the same output parameter in the sample data of a sample machine is represented by the number of matching red balls, and the matching result of the parameter level of the input parameter in the diagnosis parameter and the parameter level of the same input parameter in the sample data of a sample machine is represented by the number of matching blue balls.

[0117] In the embodiment, if the parameter level of an input parameter in the diagnosis parameter is the same as the parameter level of the same input parameter in the sample parameter of a sample machine, it is determined that the matching blue ball number of the machine to be diagnosed and the sample machine is increased by 1; if the parameter level of an output parameter in the diagnosis parameter is the same as the parameter level of the same output parameter in the sample parameter of a sample machine, it is determined that the matching red ball number of the machine to be diagnosed and the sample machine is increased by 1. After the diagnosis parameter of the machine to be diagnosed and the sample parameter of a sample machine are completely matched, the final red and blue ball data are taken as the matching result of the machine to be diagnosed and the sample machine. For example, the final matching result of the machine to be diagnosed and a sample machine is 4 blue balls and 5 red balls, and the matching result is 4 blue balls and 5 red balls. The above process is repeated to obtain the matching result of the machine to be diagnosed and other sample machines.

[0118] In the technical scheme provided in the embodiment, the parameter level of the diagnosis parameter of the machine to be diagnosed is obtained, and whether the parameter level of the diagnosis parameter is the same as the parameter level of the sample parameter of each sample machine is determined to determine the matching result of the machine to be diagnosed and each sample machine. In this way, the matching accuracy can be improved by setting the parameter level and comparing the parameter levels, and the fault diagnosis accuracy is improved.

[0119] Reference Figure 4 , Figure 4 FIG. 3 is a flowchart of a third embodiment of the fault diagnosis method of the present application. Based on the second embodiment, the step S21 includes the following steps.

[0120] In step S23, the reference value of the diagnosis parameter is obtained.

[0121] In step S24, the ratio of the actual value of the diagnosis parameter to the reference value is obtained.

[0122] In step S25, the parameter level of the diagnosis parameter is determined according to the ratio.

[0123] In the embodiment, the reference value refers to the energy efficiency test value of the machine to be diagnosed within the qualified standard, and the normal value of the diagnosis parameter under the related working condition when the parts do not fail.

[0124] Each diagnosis parameter of the machine to be diagnosed is compared with the corresponding reference value, and the parameter level corresponding to each diagnosis parameter is obtained by calculating the ratio of the actual value of the diagnosis parameter to the reference value. The parameter level of the diagnosis parameter is determined according to the following calculation method: level 1>101% reference value; 101% reference value≥level 2≥99% reference value; 99% reference value>level 3.

[0125] The technical scheme provided in the embodiment can accurately determine the relative level of the diagnostic parameter relative to the reference value by determining the parameter level based on the reference value, thereby improving the accuracy of fault diagnosis.

[0126] With reference to Figure 5 , Figure 5 FIG. 4 is a flowchart of a fourth embodiment of the fault diagnosis method of the present application, which is based on the third embodiment. The step S22 in the above embodiment includes the following steps.

[0127] In step S26, it is determined whether the parameter level of the input parameter of the machine to be diagnosed is the same as the parameter level of the same input parameter of each of the sample machines, so as to determine the first matching result of the machine to be diagnosed and each of the sample machines.

[0128] In the present embodiment, the first matching result is the matching result of the parameter level of the input parameter of the machine to be diagnosed and the parameter level of the input parameter of a sample machine, and the first matching result is represented by the number of matching blue balls.

[0129] Specifically, the parameter level of each input parameter of the machine to be diagnosed is compared with the parameter level of the same input parameter in the sample parameters of a sample machine, and it is determined whether the two are the same. If they are the same, the number of matching blue balls of the machine to be diagnosed and the sample machine is increased by 1. After the comparison of all the input parameters of the machine to be diagnosed is completed, the total number of matching blue balls obtained finally is the first matching result of the machine to be diagnosed and the sample machine. If the number of matching blue balls obtained finally is 4, the first matching result is 4 blue balls. The above process is repeated, and the first matching result of the machine to be diagnosed and other sample machines can be obtained.

[0130] In step S27, it is determined whether the parameter level of the output parameter of the machine to be diagnosed is the same as the parameter level of the same output parameter of each of the sample machines, so as to determine the second matching result of the machine to be diagnosed and each of the sample machines.

[0131] In the present embodiment, the second matching result is the matching result of the parameter level of the output parameter of the machine to be diagnosed and the parameter level of the output parameter of a sample machine, and the second matching result is represented by the number of matching red balls.

[0132] Specifically, the parameter level of each output parameter of the machine to be diagnosed is compared with the parameter level of the same output parameter in the sample parameters of a sample machine, and it is determined whether the two are the same. If they are the same, the number of matching red balls of the machine to be diagnosed and the sample machine is increased by 1. After the comparison of all the output parameters of the machine to be diagnosed is completed, the total number of matching red balls obtained finally is the second matching result of the machine to be diagnosed and the sample machine. If the number of matching red balls obtained finally is 5, the second matching result is 5 red balls. The above process is repeated, and the second matching result of the machine to be diagnosed and other sample machines can be obtained.

[0133] Step S28, determining the matching result of the machine to be diagnosed and each sample machine according to the first matching result and the second matching result.

[0134] In the embodiment, the matching result is the final matching result after the comparison of all the diagnostic parameters of the machine to be diagnosed and all the sample parameters of the sample machine, and the matching result is represented by the red ball and blue ball data of the machine to be diagnosed and a sample machine. The machine to be diagnosed and each sample machine in the sample database will generate a matching result. If the first matching result is 4 blue balls and the second matching result is 5 red balls, the matching result is 4 blue balls 5 red balls.

[0135] In the technical scheme provided in the embodiment, the first matching result and the second matching result of the machine to be diagnosed and the sample machine are calculated respectively, and the first matching result and the second matching result are combined to form the matching result of the machine to be diagnosed and the sample machine, so that the matching method is more reasonable, the accuracy of the matching result is improved, and the accuracy of the fault diagnosis is improved.

[0136] Reference Figure 6 , Figure 6 FIG. 5 is a flowchart of a fifth embodiment of the fault diagnosis method of the present application. Based on the first embodiment, the step S10 includes:

[0137] Step S11, obtaining the input parameter of the machine to be diagnosed;

[0138] In the embodiment, the input parameter is the parameter value of the input part of the machine to be diagnosed. The input parameter is part of the diagnostic parameter of the machine to be diagnosed. The fault diagnosis device obtains the input parameter of the machine to be diagnosed through a sensor. The machine to be diagnosed is provided with a sensor corresponding to the input parameter. The sensor is connected with the fault diagnosis device. The sensor collects the input parameter of the machine to be diagnosed, and sends the collected input parameter to the fault diagnosis device. The fault diagnosis device receives the input parameter collected by the sensor. It can be understood that the input parameter that cannot be collected by the sensor can also be calculated by manual experiment.

[0139] Step S12, inputting the input parameter of the machine to be diagnosed into a target neural network model for prediction to obtain the output parameter of the machine to be diagnosed;

[0140] In the embodiment, the output parameter is a parameter value of an output part of the machine to be diagnosed, and the output parameter is also part of the diagnosis parameter of the machine to be diagnosed; the target neural network model is a neural network model that is pre-established and trained, and the target neural network model can be a BP (back propagation) neural network model, of course, in other embodiments, the target neural network model can be determined according to actual conditions, and the embodiment does not limit this. After the fault diagnosis device obtains the input parameter of the machine to be diagnosed, the input parameter is input into the pre-established and trained target neural network model to obtain the output parameter of the machine to be diagnosed. It can be understood that the output parameter can also be obtained through a sensor and manual experiment calculation.

[0141] Step S13, determining the input parameter and the output parameter as the diagnosis parameter of the machine to be diagnosed.

[0142] In the embodiment, after the fault diagnosis device obtains the input parameter and the output parameter of the machine to be diagnosed, the input parameter of the machine to be diagnosed and the output parameter obtained according to the input parameter and the BP (back propagation) neural network model are determined as the diagnosis parameter of the machine to be diagnosed.

[0143] In the technical scheme provided by the embodiment, the input parameter of the machine to be diagnosed is obtained, the input parameter is input into the pre-established and trained target neural network model to obtain the output parameter of the machine to be diagnosed, and the input parameter of the machine to be diagnosed and the output parameter obtained based on the target neural network model are used as the diagnosis parameter of the machine to be diagnosed, which can reduce the number of sensors in the machine to be diagnosed, save hardware resources, and improve the efficiency and accuracy of obtaining the diagnosis parameter.

[0144] Reference Figure 7 , Figure 7 FIG. 6 is a flowchart of a sixth embodiment of the fault diagnosis method of the application, and based on the fifth embodiment, the step S11 includes the following steps before the step S11.

[0145] Step S14, obtaining a preset neural network model.

[0146] In the embodiment, the preset neural network model is a neural network model that is pre-established in the fault diagnosis device but not trained, and the preset neural network model can be a BP (back propagation) neural network model.

[0147] Specifically, taking the establishment of a BP (back propagation) neural network model as an example, the neural network correlation between the input parameter and the output parameter of the machine to be diagnosed is established in the following manner:

[0148] (1) setting an input layer: taking the input parameters of the machine to be diagnosed as the input layer of the preset neural network model, the number of input parameters being equal to the number of layers of the input layer.

[0149] (2) setting a hidden layer: the node bias of the hidden layer being Qj=10, the number of hidden layers being 2, it being noted that the node bias of the hidden layer and the number of layers can be determined according to actual needs, which are not limited in the embodiment.

[0150] (3) setting an output layer: taking the output parameters of the machine to be diagnosed as the output layer of the preset neural network model, the number of output parameters being equal to the number of layers of the output layer.

[0151] After the input layer, the hidden layer and the output layer are set, the preset neural network model (BP neural network model) is obtained. Referring to Figure 12 , Figure 12 The BP neural network model involved in the embodiment of the application is shown in the schematic diagram.

[0152] According to the neural network model, after the input data is input, the input layer is transmitted to the front (hidden layer and output layer), and the calculation formula is:

[0153]

[0154] where I j is the value of the next transmission unit (waiting for nonlinear transformation), W ij is the weight between each unit and the next unit, and Q j is the bias. Through nonlinear transformation, the output value can be obtained.

[0155] Step S15: inputting the input parameters of the training machine in the training database into the preset neural network model to obtain the actual output value of the output parameters of the training machine.

[0156] In the embodiment, the training machine is a machine device of the same type as the machine device of the machine to be diagnosed, and the training database is a collection of a certain number of training machines, including the input parameters and the output parameters of the training machine, and the training database is used to obtain the target neural network model from the preset neural network model.

[0157] Specifically, the training database is pre-established in the fault diagnosis device, and after the fault diagnosis device obtains the preset neural network model, the input parameters in the training database are input into the preset neural network model to obtain the actual output value of the training machine.

[0158] Step S16: obtaining the expected output value of the output parameters of the training machine.

[0159] In the embodiment, the expected output value is an ideal output value expected to be obtained by inputting the input parameter of the training machine into the preset neural network model, and the expected output value of the output parameter of the training machine is stored in the fault diagnosis device in advance, and the fault diagnosis device automatically acquires the expected output value of the output parameter of the training machine.

[0160] In step S17, when the actual output value is equal to the expected output value, the preset neural network model is determined as the target neural network model.

[0161] In the embodiment, after the fault diagnosis device acquires the actual output value and the expected output value of the input parameter of the training machine, the actual output value and the expected output value are compared to determine whether the actual output value is equal to the expected output value, and when the actual output value is equal to the expected output value, the preset neural network model is determined as the target neural network model.

[0162] Further, when the actual output value is not equal to the expected output value, an error value between the actual output value and the expected output value is calculated, and then the weight of the network node bias in the preset neural network model is corrected according to the error value.

[0163] Further, the preset neural network model is updated as the corrected neural network model, the input parameter of the training machine in the training database is input into the updated preset neural network model to obtain the actual output value of the training machine, and the above process is repeated until the actual output value of the output parameter of the training machine is equal to the expected output value, and finally the preset neural network model after correction is determined as the target neural network model.

[0164] In the technical scheme provided in the embodiment, the preset neural network model is trained by the training database to obtain the target neural network model, which can improve the prediction accuracy of the target neural network model, and further improve the diagnosis efficiency and accuracy of fault diagnosis.

[0165] It should be noted that in this document, the terms "comprising", "including", or any other variant thereof are intended to cover non-exclusive inclusions, so that a process, method, article or system including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such a process, method, article or system. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of another identical element in the process, method, article or system including the element.

[0166] The above-mentioned serial numbers of the embodiments of the application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0167] Those skilled in the art can clearly understand the above-mentioned embodiment method can be realized by means of software and the necessary general hardware platform, of course, also can be through hardware, but in many cases the former is the better embodiment. Based on such understanding, the technical solutions of the present application essentially or say the part of the contribution to the prior art can be embodied in the form of software products, the computer software product is stored in a storage medium (such as ROM / RAM, magnetic disc, optical disc) as described above, including a number of instructions to make a terminal device (may be a mobile phone, computer, server, or network equipment, etc.) executes the method described in various embodiments of the present application.

[0168] The above is only an optional embodiment of the present application, and does not limit the patent scope of the present application, any equivalent structure or equivalent flow transformation made by using the content of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A failure diagnosis method characterized by comprising: The fault diagnosis method comprises: obtaining a diagnosis parameter of a machine to be diagnosed, wherein the diagnosis parameter comprises an operating parameter of the machine to be diagnosed under a related working condition, and the operating parameter comprises an input parameter and an output parameter of the machine to be diagnosed; matching the diagnosis parameter with a sample parameter of each sample machine in a sample database to obtain a matching result of the machine to be diagnosed and each sample machine, wherein the sample parameter comprises an input parameter and an output parameter of the sample machine, a matching red ball number is used to represent a matching result of a parameter level of an output parameter of the machine to be diagnosed and a parameter level of a same output parameter in sample data of a sample machine, and a matching blue ball number is used to represent a matching result of a parameter level of an input parameter in the diagnosis parameter and a parameter level of a same input parameter in sample data of a sample machine; performing a high-low sorting of matching degrees of each matching result, determining a sample machine with a highest matching degree as a sample machine matched with the machine to be diagnosed, or determining sample machines with a highest matching degree and a second highest matching degree as sample machines matched with the machine to be diagnosed, wherein a rule of the high-low sorting of the matching degrees is that: the matching blue ball numbers in each matching result are preferentially sorted from high to low, wherein the higher the matching blue ball number is, the higher the matching degree of the machine to be diagnosed and the sample machine is, when the matching blue ball numbers are the same, the matching red ball numbers are sorted from high to low, and the more the matching red ball numbers are, the higher the matching degree is; obtaining fault information of the sample machine, and determining a diagnosis result of the machine to be diagnosed according to the fault information, wherein the diagnosis result comprises a problem component of the machine to be diagnosed, a problem influencing factor and / or an influence weight of the problem influencing factor.

2. The failure diagnosis method according to claim 1, characterized by, The step of matching the diagnosis parameter with the sample parameter of each sample machine in the sample database to obtain the matching result of the machine to be diagnosed and each sample machine comprises: obtaining a parameter level of the diagnosis parameter, wherein the parameter level is a relative level of an actual value of the diagnosis parameter relative to a reference value of the diagnosis parameter; determining whether the parameter level of the diagnosis parameter and a parameter level of the sample parameter of each sample machine are the same to determine the matching result of the machine to be diagnosed and each sample machine.

3. The failure diagnosing method according to claim 2, characterized by, The step of obtaining the parameter level of the diagnosis parameter comprises: obtaining a reference value of the diagnosis parameter; obtaining a ratio of an actual value of the diagnosis parameter to the reference value; determining the parameter level of the diagnosis parameter according to the ratio.

4. The failure diagnosing method according to claim 3, characterized by, The step of determining whether the parameter level of the diagnosis parameter and the parameter level of the sample parameter of each sample machine are the same to determine the matching result of the machine to be diagnosed and each sample machine comprises: determining whether the parameter level of the input parameter of the machine to be diagnosed and a parameter level of a same input parameter of each sample machine are the same to determine a first matching result of the machine to be diagnosed and each sample machine; determining whether the parameter level of the output parameter of the machine to be diagnosed and a parameter level of a same output parameter of each sample machine are the same to determine a second matching result of the machine to be diagnosed and each sample machine. determining whether the parameter level of the output parameter of the machine to be diagnosed is same as the parameter level of the same output parameter of each of the sample machines, to determine a second matching result of the machine to be diagnosed and each of the sample machines; determining a matching result of the machine to be diagnosed and each of the sample machines according to the first matching result and the second matching result.

5. The failure diagnostic method according to Claim 1, characterized by, The step of obtaining the diagnosis parameters of the machine to be diagnosed comprises: obtaining input parameters of the machine to be diagnosed; inputting the input parameters of the machine to be diagnosed into a target neural network model for prediction to obtain output parameters of the machine to be diagnosed; determining the input parameters and the output parameters as the diagnosis parameters of the machine to be diagnosed.

6. The failure diagnostic method according to claim 5, characterized by, The step of obtaining the input parameters of the machine to be diagnosed further comprises: obtaining a preset neural network model; inputting input parameters of a training machine in a training database into the preset neural network model to obtain actual output values of output parameters of the training machine; obtaining expected output values of the output parameters of the training machine; determining the preset neural network model as the target neural network model when the actual output values are equal to the expected output values.

7. The failure diagnosing method according to claim 6, wherein The step of obtaining the expected output values of the output parameters of the training machine further comprises: when the actual output values are not equal to the expected output values, obtaining error values between the expected output values and the actual output values; correcting weights of network node biases in the preset neural network model according to the error values; updating the preset neural network model into a corrected neural network model, and returning to execute the step of obtaining the preset neural network model.

8. The failure diagnostic method according to Claim 1, wherein The step of obtaining the fault information of the sample machines, and determining the diagnosis result of the machine to be diagnosed according to the fault information of the sample machines, wherein the diagnosis result comprises a problem component of the machine to be diagnosed, a problem influencing factor and / or an influence weight of the problem influencing factor, further comprises: associating the diagnosis parameters and the diagnosis result of the machine to be diagnosed; storing the associated diagnosis parameters and the diagnosis result of the machine to be diagnosed into the sample database.

9. A failure diagnosing apparatus characterized by comprising: The fault diagnosis apparatus comprises a memory, a processor, and a fault diagnosis program stored on the memory and executable on the processor, and the fault diagnosis program, when executed by the processor, implements the steps of the fault diagnosis method according to any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a fault diagnosis program, and the fault diagnosis program, when executed by the processor, implements the steps of the fault diagnosis method according to any one of claims 1-8.

Images