An intelligent fault diagnosis method and system for electromechanical equipment

Abstract

The application provides a kind of intelligent fault diagnosis method and system for electromechanical equipment, and the method comprises: determining the equipment working flow of target electromechanical equipment, and calibrating important sub-process in equipment working flow;According to the preset component-sub-process mapping table, determine that each calibration result is based on the target component of target electromechanical equipment, and according to the contribution degree of corresponding calibrated sub-process and working attribute, the corresponding target component is detected layout;To the remaining components of target electromechanical equipment, routine layout is carried out;Based on detection layout, obtain first detection signal, at the same time, based on routine layout, obtain second detection signal, and based on first detection signal and second detection signal, carry out fault diagnosis to target electromechanical equipment.Guarantee the integrity detection of target component, improve diagnosis accuracy and diagnosis effectiveness.

Description

Technical Field

[0001] This invention relates to the field of equipment diagnostic technology, and in particular to an intelligent fault diagnosis method and system for electromechanical equipment. Background Technology

[0002] In the modern facilities of large and medium-sized enterprises such as steel, coal mining, and petrochemical industries, various large and complex mechanical equipment, such as sizing and reducing machines, motors, pumps, fans, air compressors, and gearboxes, play a crucial role. With the development of technology and the continuous modernization of equipment management, the analysis of machine failures has evolved from the initial simulation analysis to the current digital analysis technology, greatly improving the speed and performance of the analysis.

[0003] However, in the general diagnostic process, corresponding detection devices are set on different parts of electromechanical equipment according to human common sense, and fault diagnosis is performed based on the detection results. This result will have a certain degree of detection error set by humans, which greatly reduces the effectiveness of obtaining fault diagnosis results and thus cannot guarantee the rationality of fault diagnosis.

[0004] Therefore, this invention proposes an intelligent fault diagnosis method and system for electromechanical equipment. Summary of the Invention

[0005] This invention provides an intelligent fault diagnosis method and system for electromechanical equipment, which uses the equipment's workflow to perform relevant layout for different components of the electromechanical equipment, ensure the integrity detection of the target components, and then improve the accuracy and effectiveness of diagnosis by diagnosing faults through the detection signals.

[0006] This invention provides an intelligent fault diagnosis method for electromechanical equipment, comprising: Step 1: Determine the equipment workflow of the target electromechanical equipment and identify the important sub-processes in the equipment workflow; Step 2: According to the preset component-subprocess mapping table, determine the target component of each calibration result based on the target electromechanical equipment, and perform detection layout on the corresponding target component according to the contribution degree and working attributes of the corresponding calibrated subprocess; Step 3: Perform conventional layout for the remaining components of the target electromechanical equipment; Step 4: Based on the detection layout, acquire the first detection signal. At the same time, based on the conventional layout, acquire the second detection signal. Based on the first detection signal and the second detection signal, perform fault diagnosis on the target electromechanical equipment.

[0007] Preferably, the remaining components of the target electromechanical equipment are arranged in a conventional layout, including: Based on a standard fault list, high-frequency common faults of the remaining components of the target electromechanical equipment are extracted; Based on the fault attributes of each high-frequency common fault, the corresponding layout scheme is retrieved from the layout database, and the corresponding components are laid out in a conventional manner according to the layout scheme.

[0008] Preferably, the equipment workflow of the target electromechanical equipment is determined, including: Determine the current operating status of the target electromechanical equipment; The workflow that matches the current working state is retrieved and considered as the device workflow.

[0009] Preferably, the key sub-processes in the workflow of the device are identified, including: According to the standard equipment breakdown rules, the equipment workflow is broken down into several sub-processes; The importance of different sub-processes is determined based on the key standards for equipment leaving the factory, and important sub-processes are selected.

[0010] Preferably, after determining the target component of the target electromechanical equipment for each calibration result according to the preset component-subprocess mapping table, the method further includes: Based on the preset component-subprocess mapping table, obtain the parameter component set for each important subprocess, wherein the participating component set includes: the target component corresponding to each calibration result; Each set of participating components is determined based on the labeling layer of the target electromechanical equipment; Based on all annotation layers, a hierarchical annotation model is constructed. At the same time, according to all participating component sets, independent components are planned and the total number of the independent components is determined. Combined with the hierarchical annotation model, the overlap factor of each independent component is determined. Obtain the non-independent components and the first independent components present in each annotation layer, and determine the first number of the first independent components and the second number of the non-independent components in each annotation layer; Determine the first ratio of the first quantity to the total quantity and the second ratio of the second quantity to the total quantity in each annotation layer. At the same time, based on the overlap factor and the process attributes of the important sub-processes corresponding to each annotation layer, assign a weight factor to each annotation layer. The initial layer weights of the corresponding labeling layers are determined based on the first ratio, the second ratio, and the weighting factor for each labeling layer. Obtain the outline overlap ratio and area overlap ratio of each non-independent component and its corresponding independent component in each annotation layer; Based on the proportion of contour overlap and the proportion of area overlap, the proportion weight of the corresponding non-independent component is determined. The proportion weight is then adjusted according to the execution sub-work of the corresponding non-independent component and the execution total work of the corresponding independent component to obtain the first weight of the corresponding non-independent component. At the same time, a corresponding second weight is configured for the first independent component in the corresponding annotation layer according to the process attributes of the important sub-processes of the corresponding annotation layer. Based on all the first weights and all the second weights existing in the same annotation layer, the initial layer weights are adjusted to obtain the current layer weights; Establish a first mapping relationship between the current layer weight and all first weights in the corresponding annotation layer, and a second mapping relationship between the current layer weight and all second weights in the corresponding annotation layer; Based on the first mapping relationship and the second mapping relationship, the contribution level of the corresponding important sub-processes is determined.

[0011] Preferably, the corresponding target components are inspected and laid out according to their contribution level and working attributes as defined by the sub-processes, including: The contribution levels are sorted, and the corresponding first sub-processes are extracted from all the marked sub-processes in descending order. The first contribution factor is then assigned to the target components participating in the first sub-processes based on the working attributes of the first sub-processes. Extract all target components involved in the first sub-process, and number each target component according to the equipment structure of the target electromechanical equipment. Based on the numbering order, construct a blank row for each first sub-process, wherein there are n blank units in the blank row, and n is consistent with the total number of numbers. Based on the numbering results, the corresponding first contribution factors are sequentially filled into the blank cells of the corresponding blank rows, and the unfilled blank cells are zero-filled. Sort the filled blank rows by column according to their contribution level to create a row list; Based on the row list, obtain the maximum contribution value and total contribution value of each target component, and determine the detection level; According to the detection level and the component attributes of the target component, match the detection list of the target component from the level-attribute database; Based on the detection methods existing in the detection list, each detection method is arranged in a current layout to achieve the detection layout of the target component.

[0012] Preferably, the process of acquiring the first detection signal based on the detection layout further includes: According to the detection layout, obtain the current layout of each detection method; Based on the current layout, establish the initial positional relationship with the corresponding target parts; Based on the initial positional relationship, the initial detection surfaces of the corresponding target parts based on different detection methods are obtained, and it is determined whether the corresponding initial detection surfaces meet the corresponding standard detection conditions. If satisfied, maintain the current layout of the corresponding initial detection surface. Otherwise, determine the positional differences between the corresponding initial detection surface and the preset standard detection points under the corresponding standard detection conditions, and obtain the outermost boundary line formed by all the corresponding difference points; Determine whether the outermost boundary line and the initial detection surface form an intersection region; If it exists, the initial detection surface is first expanded according to the outermost boundary line, and the lines in the intersection area are deleted to obtain the adjusted detection surface; If it does not exist, determine the minimum line connecting the outermost boundary line and the initial detection surface, and based on the minimum line, obtain the first intersection point with the outermost boundary line and the second intersection point with the initial detection surface; Obtain the first line connecting the second intersection point and the point of maximum difference, and obtain the third intersection point of the first line connecting the outermost boundary line; Determine whether there are other difference points in the boundary line corresponding to the third intersection point and the first intersection point. If so, obtain the specified difference point closest to the first intersection point as the first extension point. If it does not exist, the third intersection point will be used as the first extension point; Based on the line segment distance between the first extension point and the first intersection point, extend to the other side of the boundary line of the first intersection point by the same distance, and obtain the corresponding second extension point; Simultaneously, based on the equal distance length, the initial detection surface is extended from the first intersection point to the left and right boundary lines to obtain the third extension point and the fourth extension point; Connect the points on the same side of the initial detection surface and the outermost boundary line to obtain the connecting channel, and then obtain the adjusted detection surface.

[0013] Preferably, based on a conventional layout, the acquisition of the second detection signal includes: Based on the conventional layout, a conventional position is determined, and a second detection signal is obtained at the corresponding position by using a detection sensor arranged at the conventional position.

[0014] Preferably, fault diagnosis of the target electromechanical equipment based on the first detection signal and the second detection signal includes: Based on the first detection signal, a first detection matrix for the same first target component at the same timestamp is constructed; The working cycle of the target electromechanical equipment is determined, and the first detection matrix is ​​periodically split according to the working cycle. The first column vector in each split cycle is obtained, and the cumulative value of each first column vector is calculated according to the detection weight corresponding to each detection method in the first detection matrix. Where L represents the accumulated value of the corresponding first column vector; m represents the total number of elements in the corresponding first column vector; This represents the detection weight corresponding to the detection method at the row position of the i-th element in the first column vector; This represents the detection value of the i-th element in the first column vector; This represents the conversion coefficient used to convert the detection value of the i-th element to a standard format. This represents the adjustment factor of the detection method for the row position of the i-th element on the detection result; Compare the accumulated values ​​corresponding to all splitting periods contained in the first detection matrix, and mark the abnormal columns as the first label; Obtain the first period to which the labeled abnormal column belongs, and count the number of abnormal elements in each row vector of the period matrix corresponding to the first period. When the count value is greater than a preset value, perform a second labeling on the abnormal row of the first period. Based on the first annotation result and the second annotation result, the overlapping annotation elements in each row of the first detection matrix are obtained; According to the detection method of the row with overlapping labeled elements, the corresponding fault analysis model is retrieved, and the position distribution and element size of the overlapping labeled elements in the corresponding row are analyzed to obtain the first fault information of the corresponding first target part. Based on the second detection signal, a second detection matrix for the same second target component at the same time stamp is constructed, and based on a conventional analysis model, the detection elements in the second detection matrix are analyzed to obtain the second fault information corresponding to the second target part. Based on all the first fault information and the second fault information, the fault result of the target electromechanical equipment is obtained, and the fault level is determined. Where G represents the fault level; This indicates the effective weight of the information related to the first fault. This indicates the effective weight of the information regarding the second fault information, and m1 represents the total number of first target locations; m2 represents the total number of second target locations; m3 represents the total number of first fault information entries corresponding to the first target locations, and m3 is a variable, with different total number of first fault information entries corresponding to different first target locations; m4 represents the total number of second fault information entries corresponding to the second target locations, and m4 is a variable, with different total number of second fault information entries corresponding to different second target locations. This represents the weight of the j1th first target location; The fault factor represents the j3rd first fault information present in the j1st first target location, and its value range is [0, 1]. The fault factor represents the j4th second fault information existing in the j2nd second target part, and its value range is [0, 1]. This represents the maximum fault factor among all first fault information corresponding to all first target locations; This represents the maximum fault factor among all second fault information corresponding to all second target locations; Represents the logarithmic function with base e; When the fault level is greater than the preset level, the fault result is output in conjunction with the fault result, and the character representation of the fault level is enlarged and displayed.

[0015] This invention provides an intelligent fault diagnosis system for electromechanical equipment, comprising: The determination module is used to determine the equipment workflow of the target electromechanical equipment and identify the important sub-processes in the equipment workflow; The detection layout module is used to determine the target component of the target electromechanical equipment for each calibration result according to the preset component-subprocess mapping table, and to perform detection layout on the corresponding target component according to the contribution degree and working attributes of the corresponding calibrated subprocess. A conventional layout module is used to perform conventional layout of the remaining components of the target electromechanical equipment; The fault diagnosis module is used to acquire a first detection signal based on the detection layout, and simultaneously acquire a second detection signal based on the conventional layout, and perform fault diagnosis on the target electromechanical equipment based on the first detection signal and the second detection signal.

[0016] Compared with the prior art, the beneficial effects of this application are as follows: By analyzing the equipment's workflow, we can strategically position different components of the electromechanical equipment to ensure the integrity of the target components. Furthermore, by analyzing the detected signals, we can improve the accuracy and effectiveness of the diagnostics.

[0017] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0018] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0019] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of an intelligent fault diagnosis method for electromechanical equipment according to an embodiment of the present invention; Figure 2 This is a structural diagram of an intelligent fault diagnosis system for electromechanical equipment according to an embodiment of the present invention; Figure 3 This is a structural diagram of the row list in an embodiment of the present invention; Figure 4 A structural diagram showing overlapping regions; Figure 5 This is a structure diagram where there are no overlapping regions. Detailed Implementation

[0020] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0021] Example 1: This invention provides an intelligent fault diagnosis method for electromechanical equipment, such as... Figure 1 As shown, it includes: Step 1: Determine the equipment workflow of the target electromechanical equipment and identify the important sub-processes in the equipment workflow; Step 2: According to the preset component-subprocess mapping table, determine the target component of each calibration result based on the target electromechanical equipment, and perform detection layout on the corresponding target component according to the contribution degree and working attributes of the corresponding calibrated subprocess; Step 3: Perform conventional layout for the remaining components of the target electromechanical equipment; Step 4: Based on the detection layout, acquire the first detection signal. At the same time, based on the conventional layout, acquire the second detection signal. Based on the first detection signal and the second detection signal, perform fault diagnosis on the target electromechanical equipment.

[0022] In this embodiment, the target electromechanical equipment can be a motor, pump, fan, air compressor, automation equipment, etc. For example, when acquiring the working process of the fan, such as the working process in different gear states, the working process in the low gear state includes: the initial rotation stage after turning on - the continuous rotation stage - the rotation stage after turning off, and each stage can be regarded as a sub-process.

[0023] For example, the continuous rotation phase identified at this point is the key monitoring phase, and this phase can be calibrated and regarded as an important sub-process.

[0024] In this embodiment, the preset component-subprocess mapping table includes: a mapping relationship between a component and the subprocesses involved in that component. This mapping relationship can be one-to-many or one-to-one. This is mainly to facilitate the determination of the corresponding target component according to the marked subprocesses. For example, the workflow of the target electromechanical equipment 1 includes sub-processes 1, 2, and 3. In this case, according to the preset component-sub-process mapping table, sub-process 1 is mapped to components 1 and 2, sub-process 2 is mapped to components 2 and 3, and sub-process 3 is mapped to component 3. Among them, sub-process 3 is an important sub-process. In this case, component 3 is the target component. The contribution level of process 3 is determined (that is, the contribution of component 3 based on the entire equipment workflow). The work attribute refers to the work content of the sub-process, such as maintaining the fan to maintain high-intensity heat dissipation.

[0025] In this embodiment, the detection layout refers to the arrangement of different types of detection sensors on the target component. Since different key sub-processes may involve the same target part, the detection method and detection deployment of the target part can be determined according to the contribution level and working attributes of different key sub-processes, thereby realizing the detection layout.

[0026] For example, detection method 1 detects the temperature at different locations on the target part, detection method 2 detects the wear condition at different detection points on the target part, and detection method 3 detects the vibration information at different detection points on the target part, etc., in order to ensure the reliability of detection on the same target part.

[0027] In this embodiment, conventional layout refers to detecting the remaining components according to conventional detection methods, such as detecting a single location point, like detecting vibration.

[0028] In this embodiment, both the first detection signal and the second detection signal are obtained based on sensors deployed at the target location. By analyzing the signals, such as performing Fourier analysis on the vibration signal, it is determined whether there is abnormal vibration, and thus the vibration fault in that location is identified. Based on the detection of all locations, the electromechanical equipment can be diagnosed, for example, by detecting problems such as imbalance, misalignment, and gear meshing in the electromechanical equipment.

[0029] In this embodiment, the detection layout and conventional layout refer to the use of corresponding sensor devices to set up at the corresponding detection points after determining the detection points of different detection methods for the target part, so as to obtain different detection signals and perform fault diagnosis.

[0030] The beneficial effects of the above technical solution are: by using the equipment workflow to arrange the relevant components of the electromechanical equipment, the integrity of the target components can be guaranteed, and then the accuracy and effectiveness of diagnosis can be improved by diagnosing faults through the detection signals.

[0031] Example 2: Based on Embodiment 1, the remaining components of the target electromechanical equipment are arranged in a conventional manner, including: Based on a standard fault list, high-frequency common faults of the remaining components of the target electromechanical equipment are extracted; Based on the fault attributes of each high-frequency common fault, the corresponding layout scheme is retrieved from the layout database, and the corresponding components are laid out in a conventional manner according to the layout scheme.

[0032] In this embodiment, the standard fault list is determined at the factory of the target electromechanical equipment. For example, component 1 in the motor of the target electromechanical equipment is prone to failure and is a high-frequency common fault. Then, according to the fault attribute of component 1, that is, what type of fault it is, the layout scheme, that is, the scheme that can detect the type of fault, is obtained from the layout database to arrange component 1 and obtain the detection information of component 1 when it is working.

[0033] The beneficial effects of the above technical solution are: by identifying high-frequency common faults and making conventional arrangements for the corresponding components, not only can the components be reasonably tested, but the testing efficiency can also be improved, thereby enhancing the effectiveness of subsequent fault diagnosis.

[0034] Example 3: Based on Example 1, the equipment workflow of the target electromechanical equipment is determined, including: Determine the current operating status of the target electromechanical equipment; The workflow that matches the current working state is retrieved and considered as the device workflow.

[0035] In this embodiment, the electromechanical equipment is in different working states and has different workflows.

[0036] The beneficial effects of the above technical solution are: by matching the workflow with the working status, it facilitates subsequent detection and analysis of the target parts.

[0037] Example 4: Based on Example 1, the important sub-processes in the workflow of the equipment are identified, including: According to the standard equipment breakdown rules, the equipment workflow is broken down into several sub-processes; The importance of different sub-processes is determined based on the key standards for equipment leaving the factory, and important sub-processes are selected.

[0038] In this embodiment, the standard equipment splitting rules are pre-set and are related to the working and execution functions of the target electromechanical equipment, such as the equipment workflow executed in a working state. After splitting the workflow, processes 1, 2, and 3 are obtained. The factory importance standard is set to determine the importance of different processes 1, 2, and 3. For example, the importance of sub-processes 1, 2, and 3 is 0.2, 0.3, and 0.6, respectively. At this time, the importance of 0.6 is greater than the preset importance of 0.5, so sub-process 3 is regarded as an important sub-process.

[0039] The beneficial effects of the above technical solution are: by breaking down and filtering the process, it is easier to obtain important sub-processes, which facilitates the subsequent layout of different parts, improves the effectiveness of detection, and further improves the reliability of subsequent fault diagnosis.

[0040] Example 5: Based on Example 1, after determining the target component of the target electromechanical equipment for each calibration result according to the preset component-subprocess mapping table, the method further includes: Based on the preset component-subprocess mapping table, obtain the parameter component set for each important subprocess, wherein the participating component set includes: the target component corresponding to each calibration result; Each set of participating components is determined based on the labeling layer of the target electromechanical equipment; Based on all annotation layers, a hierarchical annotation model is constructed. At the same time, according to all participating component sets, independent components are planned and the total number of the independent components is determined. Combined with the hierarchical annotation model, the overlap factor of each independent component is determined. Obtain the non-independent components and the first independent components present in each annotation layer, and determine the first number of the first independent components and the second number of the non-independent components in each annotation layer; Determine the first ratio of the first quantity to the total quantity and the second ratio of the second quantity to the total quantity in each annotation layer. At the same time, based on the overlap factor and the process attributes of the important sub-processes corresponding to each annotation layer, assign a weight factor to each annotation layer. The initial layer weights of the corresponding labeling layers are determined based on the first ratio, the second ratio, and the weighting factor for each labeling layer. Obtain the outline overlap ratio and area overlap ratio of each non-independent component and its corresponding independent component in each annotation layer; Based on the proportion of contour overlap and the proportion of area overlap, the proportion weight of the corresponding non-independent component is determined. The proportion weight is then adjusted according to the execution sub-work of the corresponding non-independent component and the execution total work of the corresponding independent component to obtain the first weight of the corresponding non-independent component. At the same time, a corresponding second weight is configured for the first independent component in the corresponding annotation layer according to the process attributes of the important sub-processes of the corresponding annotation layer. Based on all the first weights and all the second weights existing in the same annotation layer, the initial layer weights are adjusted to obtain the current layer weights; Establish a first mapping relationship between the current layer weight and all first weights in the corresponding annotation layer, and a second mapping relationship between the current layer weight and all second weights in the corresponding annotation layer; Based on the first mapping relationship and the second mapping relationship, the contribution level of the corresponding important sub-processes is determined.

[0041] In this embodiment, for example, there are important sub-processes 1 and 2, and important sub-process 1 includes participating component 1, participating component 2, and participating component 3, and important sub-process 2 includes participating component 1, participating component 4, and participating component 5, and the corresponding target components are: target 1, 2, 3, 4, and 5. Based on the important sub-process 1, the annotation layer composed of participating components 1, 2, and 3 is obtained. By placing all the annotation layers in sequence (one layer at a time), a hierarchical annotation model is constructed.

[0042] In this embodiment, a participating component in the participating component set can be a small component within a larger component. For example, device 1 includes motor 1 and motor 2, and the motor includes a stator, rotor, and accessories. For instance, suppose device 1 includes component one and component two, where component one and component two are independent components. Component one includes component 11, component 12, and component 13, and component two includes component 21, component 22, and component 23. If participating component set 1 includes components 11 and 12, and participating component set 2 includes components 13 and 21, then the planned independent component includes components one and two, with a total quantity of 2. That is, as long as a participating component corresponds to a new independent component, planning can be performed based on that participating component to determine the corresponding independent component. For example, if participating component 12 belongs to component one, then component one can be considered an independent component.

[0043] In this embodiment, the overlap factor is for the labeling of the same independent component at different levels. If the labeling of the same independent component is complete at each level, the corresponding overlap factor is 1. That is, the labeling of the same independent component is related to the labeling position, labeling area, and labeling level of each level, and the value range is 0 to 1.

[0044] In this embodiment, for example, the first number of independent components in the annotation layer 1 is 1, and the total number of independent components determined based on the set of participating components is 2. At this time, the corresponding first ratio is 1 / 2. At this time, the second number of incomplete independent components in the corresponding annotation layer is 1, and the corresponding second ratio is 1 / 2. In addition, the incomplete independent component refers to the independent component that is not fully annotated in the annotation layer, but only a part of the independent component where the incomplete independent component is located is annotated.

[0045] The process attributes are related to the importance of participation in the corresponding important sub-processes. The greater the importance of participation, the greater the overlap factor, and the greater the weight factor of the corresponding annotation layer. The initial layer weight is determined based on the first ratio, the second ratio, and the weight factor.

[0046] for example: Where b1 represents the first ratio and b2 represents the second ratio. denoted as weight factor, and c represents the initial layer weight.

[0047] In this embodiment, each non-independent component has its corresponding independent component and is a part of the independent component. The outline overlap ratio and area overlap ratio are mainly used to determine the proportional relationship between the corresponding non-independent component and the corresponding independent component.

[0048] In this embodiment, the greater the proportion of contour overlap and the greater the proportion of area overlap, the greater the corresponding proportion weight. The proportion weight is adjusted by determining the execution of sub-work and the execution of total work to obtain the first weight. Then, the second weight of the first independent component of the same annotation layer is obtained.

[0049] In this embodiment, by obtaining the weights of non-independent components and the weights of the first independent component, the main purpose is to determine the contribution of the sub-process to the failure. Then, by setting the weights, the effective detection settings for the components involved in different sub-processes can be effectively obtained.

[0050] In this embodiment, for example, if there are weights b1 of non-independent component 1, weight b2 of non-independent component 1, weight b3 of first independent component 3, and weight b4 of second independent component 4 in annotation layer 1, the initial layer weights can be adjusted. For example, a c1 can be calculated based on the weights b1 of non-independent component 1, weight b2 of non-independent component 1, weight b3 of first independent component 3, and weight b4 of second independent component 4, and c can be adjusted based on c1 as the current layer weight.

[0051] In this embodiment, when the lower layer weight c2 is connected with the first weights b1 and b2, a first mapping relationship is established; when the lower layer weight c2 is connected with the second weights b3 and b4, a second mapping relationship is established.

[0052] Based on the first and second mapping relationships, the contribution level is obtained from the preset contribution list. The preset contribution list is related to the target electromechanical equipment and includes different mapping ranges and their corresponding contribution levels. Therefore, the contribution level can be obtained.

[0053] The beneficial effects of the above technical solution are as follows: by determining the annotation layer, it is convenient to build the model; by determining the number of independent parts and obtaining parameters such as the ratio of the number of different annotation layers, the initial layer weights can be obtained; finally, by determining the weights of non-independent parts and independent parts in the annotation layer, adjustments can be made to facilitate the subsequent construction of mapping relationships, thereby obtaining the degree of contribution. This effectively determines the contribution of different sub-processes to the fault detection of the target electromechanical equipment, thereby improving the rationality of the detection layout and the effectiveness of subsequent fault diagnosis.

[0054] Example 6: Based on Example 1, the corresponding target components are inspected and laid out according to the contribution level and working attributes of the corresponding calibrated sub-processes, including: The contribution levels are sorted, and the corresponding first sub-processes are extracted from all the marked sub-processes in descending order. The first contribution factor is then assigned to the target components participating in the first sub-processes based on the working attributes of the first sub-processes. Extract all target components involved in the first sub-process, and number each target component according to the equipment structure of the target electromechanical equipment. Based on the numbering order, construct a blank row for each first sub-process, wherein there are n blank units in the blank row, and n is consistent with the total number of numbers. Based on the numbering results, the corresponding first contribution factors are sequentially filled into the blank cells of the corresponding blank rows, and the unfilled blank cells are zero-filled. Sort the filled blank rows by column according to their contribution level to create a row list; Based on the row list, obtain the maximum contribution value and total contribution value of each target component, and determine the detection level; According to the detection level and the component attributes of the target component, match the detection list of the target component from the level-attribute database; Based on the detection methods existing in the detection list, each detection method is arranged in a current layout to achieve the detection layout of the target component.

[0055] In this embodiment, the degree of contribution refers to the degree of contribution of each important sub-process component. This facilitates the setting of a reasonable detection layout for the components. By sorting the components by their degree of contribution, the first sub-process can be extracted in sequence. Combined with the work attributes, i.e., the work execution content, the contribution factor of the target component involved in the sub-process can be determined. At this time, the sum of the contribution factors of the target component involved in the sub-process is equal to the degree of contribution. Moreover, the more critical the content executed by the target component, the larger the first contribution factor is allocated.

[0056] In this embodiment, numbering all target components facilitates the construction of blank rows and the filling of each blank unit within the blank rows.

[0057] In this embodiment, n is 5, corresponding to 5 blank cells in blank row 1. The blank cells are numbered 01, 02, 03, 04, and 05 respectively. The contribution factor for cell 01 is 0.1, for cell 02 it is 0.01, for cell 03 it has no contribution factor, for cell 04 it is 0.02, and for cell 05 it has no contribution factor. Cells with no contribution factor are zero-filled. Blank rows 2 and 3 also exist. See details. Figure 3 It can be used as a row list.

[0058] In this embodiment, the maximum contribution value and total contribution value of the corresponding target component can be determined according to the column elements in the row and column.

[0059] In this embodiment, the detection level is determined based on two indicators: the maximum contribution value and the total contribution value.

[0060] In this embodiment, the level-attribute database includes: detection level, component attributes (related to the function to be executed), and various corresponding detection methods. This allows the acquisition of a detection list for the corresponding target component, and the layout of the target component can be achieved based on the existing detection methods.

[0061] In this embodiment, for example, the vibration and pressure of the detection component are monitored, and relevant devices are arranged in relevant locations to achieve the detection.

[0062] The beneficial effects of the above technical solution are as follows: by assigning contribution factors to the components involved in the sub-process and combining them with the component numbers, a process list can be constructed. By determining the detection level and attributes, a detection list can be obtained, and then the detection layout can be carried out. This can effectively perform multiple types of detection on a component, improve detection reliability, and ensure the accuracy of subsequent fault diagnosis.

[0063] Example 7: Based on Embodiment 1, and based on the detection layout, the process of acquiring the first detection signal further includes: According to the detection layout, obtain the current layout of each detection method; Based on the current layout, establish the initial positional relationship with the corresponding target parts; Based on the initial positional relationship, the initial detection surfaces of the corresponding target parts based on different detection methods are obtained, and it is determined whether the corresponding initial detection surfaces meet the corresponding standard detection conditions. If satisfied, maintain the current layout of the corresponding initial detection surface. Otherwise, determine the positional differences between the corresponding initial detection surface and the preset standard detection points under the corresponding standard detection conditions, and obtain the outermost boundary line formed by all the corresponding difference points; Determine whether the outermost boundary line and the initial detection surface form an intersection region; If it exists, the initial detection surface is first expanded according to the outermost boundary line, and the lines in the intersection area are deleted to obtain the adjusted detection surface; If it does not exist, determine the minimum line connecting the outermost boundary line and the initial detection surface, and based on the minimum line, obtain the first intersection point with the outermost boundary line and the second intersection point with the initial detection surface; Obtain the first line connecting the second intersection point and the point of maximum difference, and obtain the third intersection point of the first line connecting the outermost boundary line; Determine whether there are other difference points in the boundary line corresponding to the third intersection point and the first intersection point. If so, obtain the specified difference point closest to the first intersection point as the first extension point. If it does not exist, the third intersection point will be used as the first extension point; Based on the line segment distance between the first extension point and the first intersection point, extend to the other side of the boundary line of the first intersection point by the same distance, and obtain the corresponding second extension point; Simultaneously, based on the equal distance length, the initial detection surface is extended from the first intersection point to the left and right boundary lines to obtain the third extension point and the fourth extension point; Connect the points on the same side of the initial detection surface and the outermost boundary line to obtain the connecting channel, and then obtain the adjusted detection surface.

[0064] In this embodiment, the detection layout includes detection methods for different target parts, and each detection method has a corresponding current layout to facilitate placement at the target part, and the layout is realized based on the initial position relationship.

[0065] In this embodiment, the initial detection surface is determined by the corresponding initial positional relationship. For example, if the target part needs to be calibrated by detection points 1, 2, 3, 4, and 5, then if the initial detection surface contains detection points 1, 2, 3, 4, and 5, it is considered to meet the corresponding standard detection conditions, and the current layout remains unchanged. Otherwise, the positional difference is determined, that is, the detection points 1, 2, and 4 that are not in the initial detection surface are determined to construct the outermost boundary line.

[0066] In this embodiment, such as Figure 4 As shown, the initial detection surface 1 and the outermost boundary line 01 have an intersection region 001. At this time, the line 002 in the intersection region is deleted to obtain the adjusted detection surface.

[0067] like Figure 5 As shown, the initial detection surface 1 and the outermost boundary line 01 have no intersection area. At this time, d1 represents the minimum connection, d2 represents the first intersection point, d3 represents the second intersection point, and the point of maximum difference is detection point 2. At this time, the corresponding first connection is d4, and the corresponding third intersection point is d5. At this time, there are no other difference points between the first intersection point and the third intersection point. Therefore, the third intersection point is also the first extension point. The corresponding second extension point is d6, the third extension point is d7, and the fourth extension point is d8. The corresponding extension points on the same side are the first extension point and the third extension point, the second extension point and the fourth extension point. This adjustment yields the detection surface.

[0068] The beneficial effects of the above technical solution are as follows: By adjusting the detection surface, the integrity of the final acquired detection information is ensured, avoiding the reduction in the accuracy of fault analysis due to the lack of detection signals at a certain location. By expanding the boundary line and the detection surface, the consistency between the area affecting the detection results and the corresponding detection surface can be effectively established, ensuring that data omissions are avoided in subsequent analysis because the two areas are not together. In the process of determining the expansion point, on the one hand, it is to connect the area, and on the other hand, it is to minimize the occupation of the target component area under the corresponding layout of different detection methods. It can effectively curb the multiple designated placement of different detection devices at the same location point, which can cause confusion in the detection of that location point. Therefore, by using the expansion point, the integrity of data detection can be guaranteed on the basis of the integrity of the area formed by the detection surface, thus ensuring the accuracy of subsequent fault diagnosis.

[0069] Example 8: Based on Example 1, and using a conventional layout, the second detection signal is acquired, including: Based on the conventional layout, a conventional position is determined, and a second detection signal is obtained at the corresponding position by using a detection sensor arranged at the conventional position.

[0070] The beneficial effects of the above technical solution are: through conventional layout, it is easy to obtain detection signals and improve the accuracy of fault diagnosis.

[0071] Example 9: Based on Embodiment 1, fault diagnosis of the target electromechanical equipment is performed based on the first detection signal and the second detection signal, including: Based on the first detection signal, a first detection matrix for the same first target component at the same timestamp is constructed; The working cycle of the target electromechanical equipment is determined, and the first detection matrix is ​​periodically split according to the working cycle. The first column vector in each split cycle is obtained, and the cumulative value of each first column vector is calculated according to the detection weight corresponding to each detection method in the first detection matrix. Where L represents the accumulated value of the corresponding first column vector; m represents the total number of elements in the corresponding first column vector; This represents the detection weight corresponding to the detection method at the row position of the i-th element in the first column vector; This represents the detection value of the i-th element in the first column vector; This represents the conversion coefficient used to convert the detection value of the i-th element to a standard format. This represents the adjustment factor of the detection method for the row position of the i-th element on the detection result; Compare the accumulated values ​​corresponding to all splitting periods contained in the first detection matrix, and mark the abnormal columns as the first label; Obtain the first period to which the labeled abnormal column belongs, and count the number of abnormal elements in each row vector of the period matrix corresponding to the first period. When the count value is greater than a preset value, perform a second labeling on the abnormal row of the first period. Based on the first annotation result and the second annotation result, the overlapping annotation elements in each row of the first detection matrix are obtained; According to the detection method of the row with overlapping labeled elements, the corresponding fault analysis model is retrieved, and the position distribution and element size of the overlapping labeled elements in the corresponding row are analyzed to obtain the first fault information of the corresponding first target part. Based on the second detection signal, a second detection matrix for the same second target component at the same time stamp is constructed, and based on a conventional analysis model, the detection elements in the second detection matrix are analyzed to obtain the second fault information corresponding to the second target part. Based on all the first fault information and the second fault information, the fault result of the target electromechanical equipment is obtained, and the fault level is determined. Where G represents the fault level; This indicates the effective weight of the information related to the first fault. This indicates the effective weight of the information regarding the second fault information, and m1 represents the total number of first target locations; m2 represents the total number of second target locations; m3 represents the total number of first fault information entries corresponding to the first target locations, and m3 is a variable, with different total number of first fault information entries corresponding to different first target locations; m4 represents the total number of second fault information entries corresponding to the second target locations, and m4 is a variable, with different total number of second fault information entries corresponding to different second target locations. This represents the weight of the j1th first target location; The fault factor represents the j3rd first fault information present in the j1st first target location, and its value range is [0, 1]. The fault factor represents the j4th second fault information existing in the j2nd second target part, and its value range is [0, 1]. This represents the maximum fault factor among all first fault information corresponding to all first target locations; This represents the maximum fault factor among all second fault information corresponding to all second target locations; Represents the logarithmic function with base e; When the fault level is greater than the preset level, the fault result is output in conjunction with the fault result, and the character representation of the fault level is enlarged and displayed.

[0072] In this embodiment, the preset level is pre-set, and its value is generally 0.3.

[0073] In this embodiment, for example, the same timestamp includes three splitting cycles. The first target component is a component determined by an important sub-process. Based on the detection signal, the detection matrix of the first component can be obtained. By splitting the matrix according to the cycle, the first column vector is determined to determine the cumulative value. The first detection matrix is ​​composed of the detection signals of the same target part under different detection methods. The rows of the detection matrix represent the detection results of the same detection method under different cycles, and the columns of the detection matrix represent the detection results of multiple detection methods at the same time point.

[0074] In this embodiment, the detection weight is related to the importance of the corresponding detection method in detecting the target component; the more important it is, the greater the corresponding detection weight.

[0075] In this embodiment, the conversion coefficient is calculated by converting the detected value according to the numerical conversion standard, and the adjustment factor ranges from 0.01 to 0.03.

[0076] In this embodiment, for example, there are three periods. The first column vector in the first period is the abnormal column, and the period matrix refers to the matrix contained in the first period. The second annotation is to annotate a certain row in the first period. For example, there are 5 rows in the first period, and each row has 10 elements. Among them, 3 abnormal elements are counted in the first row. At this time, the second annotation is performed on the first row. At this time, the first row and first column element in the first period is the corresponding annotation overlapping element.

[0077] In this embodiment, each row is related to the detection method. Therefore, the corresponding fault analysis model is retrieved, and the position distribution and element size of the overlapping labeled elements are analyzed to determine the fault in the first target part. The fault information is based on the fault of the target part determined under the corresponding detection method.

[0078] In this embodiment, both the fault analysis model and the conventional analysis model are pre-trained according to the corresponding detection signals, thereby effectively obtaining fault information under the relevant signals.

[0079] In this embodiment, the second target part refers to the part corresponding to the non-important sub-process, that is, the remaining part excluding the first target part.

[0080] The beneficial effects of the above technical solution are as follows: by determining the detection matrix of the first target component and dividing it into cycles according to the working cycle, and calculating the cumulative value of the first column vector respectively, abnormal columns can be effectively filtered out. Furthermore, by statistically analyzing abnormal elements, the position distribution and element size of overlapping elements in each row of the matrix can be effectively obtained. Based on the model, fault information can be obtained. Through the calculation of fault level, it is convenient to enlarge the fault level representation characters, improve the effective reminder of fault information and intuitive understanding of fault severity, and improve the accuracy of fault diagnosis.

[0081] Example 10: This invention provides an intelligent fault diagnosis system for electromechanical equipment, such as... Figure 2 As shown, it includes: The determination module is used to determine the equipment workflow of the target electromechanical equipment and identify the important sub-processes in the equipment workflow; The detection layout module is used to determine the target component of the target electromechanical equipment for each calibration result according to the preset component-subprocess mapping table, and to perform detection layout on the corresponding target component according to the contribution degree and working attributes of the corresponding calibrated subprocess. A conventional layout module is used to perform conventional layout of the remaining components of the target electromechanical equipment; The fault diagnosis module is used to acquire a first detection signal based on the detection layout, and simultaneously acquire a second detection signal based on the conventional layout, and perform fault diagnosis on the target electromechanical equipment based on the first detection signal and the second detection signal.

[0082] The beneficial effects of the above technical solution are: by using the equipment workflow to arrange the relevant components of the electromechanical equipment, the integrity of the target components can be guaranteed, and then the accuracy and effectiveness of diagnosis can be improved by diagnosing faults through the detection signals.

[0083] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for intelligent fault diagnosis of electromechanical equipment, characterized in that, include: Step 1: Determine the equipment workflow of the target electromechanical equipment and identify the important sub-processes in the equipment workflow; Step 2: According to the preset component-subprocess mapping table, determine the target component of each calibration result based on the target electromechanical equipment, and perform detection layout on the corresponding target component according to the contribution degree and working attributes of the corresponding calibrated subprocess; Step 3: Perform conventional layout for the remaining components of the target electromechanical equipment; Step 4: Based on the detection layout, acquire the first detection signal; simultaneously, based on the conventional layout, acquire the second detection signal; and perform fault diagnosis on the target electromechanical equipment based on the first and second detection signals. After determining the target component of the target electromechanical equipment for each calibration result according to the preset component-subprocess mapping table, the process further includes: Based on the preset component-subprocess mapping table, obtain the set of participating components for each important subprocess, wherein the set of participating components includes: the target component corresponding to each calibration result; Each set of participating components is determined based on the labeling layer of the target electromechanical equipment; Based on all annotation layers, a hierarchical annotation model is constructed. At the same time, according to all participating component sets, independent components are planned and the total number of the independent components is determined. Combined with the hierarchical annotation model, the overlap factor of each independent component is determined. Obtain the non-independent components and the first independent components present in each annotation layer, and determine the first number of the first independent components and the second number of the non-independent components in each annotation layer; Determine the first ratio of the first quantity to the total quantity and the second ratio of the second quantity to the total quantity in each annotation layer. At the same time, based on the overlap factor and the process attributes of the important sub-processes corresponding to each annotation layer, assign a weight factor to each annotation layer. The initial layer weights of the corresponding labeling layers are determined based on the first ratio, the second ratio, and the weighting factor for each labeling layer. Obtain the outline overlap ratio and area overlap ratio of each non-independent component and its corresponding independent component in each annotation layer; Based on the proportion of contour overlap and the proportion of area overlap, the proportion weight of the corresponding non-independent component is determined. The proportion weight is then adjusted according to the execution sub-work of the corresponding non-independent component and the execution total work of the corresponding independent component to obtain the first weight of the corresponding non-independent component. At the same time, a corresponding second weight is configured for the first independent component in the corresponding annotation layer according to the process attributes of the important sub-processes of the corresponding annotation layer. Based on all the first weights and all the second weights existing in the same annotation layer, the initial layer weights are adjusted to obtain the current layer weights; Establish a first mapping relationship between the current layer weight and all first weights in the corresponding annotation layer, and a second mapping relationship between the current layer weight and all second weights in the corresponding annotation layer; Based on the first mapping relationship and the second mapping relationship, the contribution level of the corresponding important sub-processes is determined.

2. The intelligent fault diagnosis method for electromechanical equipment as described in claim 1, characterized in that, The remaining components of the target electromechanical equipment are to be laid out in a conventional manner, including: Based on a standard fault list, high-frequency common faults of the remaining components of the target electromechanical equipment are extracted; Based on the fault attributes of each high-frequency common fault, the corresponding layout scheme is retrieved from the layout database, and the corresponding components are laid out in a conventional manner according to the layout scheme.

3. The intelligent fault diagnosis method for electromechanical equipment as described in claim 1, characterized in that, Determine the equipment workflow of the target electromechanical equipment, including: Determine the current operating status of the target electromechanical equipment; The workflow that matches the current working state is retrieved and considered as the device workflow.

4. The intelligent fault diagnosis method for electromechanical equipment as described in claim 1, characterized in that, Identify the key sub-processes in the workflow of the aforementioned equipment, including: According to the standard equipment breakdown rules, the equipment workflow is broken down into several sub-processes; The importance of different sub-processes is determined based on the key standards for equipment leaving the factory, and important sub-processes are selected.

5. The intelligent fault diagnosis method for electromechanical equipment as described in claim 1, characterized in that, Based on the contribution level and work attributes of the corresponding calibrated sub-processes, the corresponding target components are inspected and laid out, including: The contribution levels are sorted, and the corresponding first sub-processes are extracted from all the marked sub-processes in descending order. The first contribution factor is then assigned to the target components participating in the first sub-processes based on the working attributes of the first sub-processes. Extract all target components involved in the first sub-process, and number each target component according to the equipment structure of the target electromechanical equipment. Based on the numbering order, construct a blank row for each first sub-process, wherein there are n blank units in the blank row, and n is consistent with the total number of numbers. Based on the numbering results, the corresponding first contribution factors are sequentially filled into the blank cells of the corresponding blank rows, and the unfilled blank cells are zero-filled. Sort the filled blank rows by column according to their contribution level to create a row list; Based on the row list, obtain the maximum contribution value and total contribution value of each target component, and determine the detection level; According to the detection level and the component attributes of the target component, match the detection list of the target component from the level-attribute database; Based on the detection methods existing in the detection list, each detection method is arranged in a current layout to achieve the detection layout of the target component.

6. The intelligent fault diagnosis method for electromechanical equipment as described in claim 1, characterized in that, Based on the detection layout, the process of acquiring the first detection signal also includes: According to the detection layout, obtain the current layout of each detection method; Based on the current layout, establish the initial positional relationship with the corresponding target parts; Based on the initial positional relationship, the initial detection surfaces of the corresponding target parts based on different detection methods are obtained, and it is determined whether the corresponding initial detection surfaces meet the corresponding standard detection conditions. If satisfied, maintain the current layout of the corresponding initial detection surface. Otherwise, determine the positional differences between the corresponding initial detection surface and the preset standard detection points under the corresponding standard detection conditions, and obtain the outermost boundary line formed by all the corresponding difference points; Determine whether the outermost boundary line and the initial detection surface form an intersection region; If it exists, the initial detection surface is first expanded according to the outermost boundary line, and the lines in the intersection area are deleted to obtain the adjusted detection surface; If it does not exist, determine the minimum line connecting the outermost boundary line and the initial detection surface, and based on the minimum line, obtain the first intersection point with the outermost boundary line and the second intersection point with the initial detection surface; Obtain the first line connecting the second intersection point and the point of maximum difference, and obtain the third intersection point of the first line connecting the outermost boundary line; Determine whether there are other difference points in the boundary line corresponding to the third intersection point and the first intersection point. If so, obtain the specified difference point closest to the first intersection point as the first extension point. If it does not exist, the third intersection point will be used as the first extension point; Based on the line segment distance between the first extension point and the first intersection point, extend to the other side of the boundary line of the first intersection point by the same distance, and obtain the corresponding second extension point; Simultaneously, based on the equal distance length, the initial detection surface is extended from the first intersection point to the left and right boundary lines to obtain the third extension point and the fourth extension point; Connect the points on the same side of the initial detection surface and the outermost boundary line to obtain the connecting channel, and then obtain the adjusted detection surface.

7. The intelligent fault diagnosis method for electromechanical equipment as described in claim 1, characterized in that, Based on a conventional layout, a second detection signal is obtained, including: Based on the conventional layout, a conventional position is determined, and a second detection signal is obtained at the corresponding position by using a detection sensor arranged at the conventional position.

8. The intelligent fault diagnosis method for electromechanical equipment as described in claim 1, characterized in that, Fault diagnosis of the target electromechanical equipment based on the first detection signal and the second detection signal includes: Based on the first detection signal, a first detection matrix for the same first target component at the same timestamp is constructed; The working cycle of the target electromechanical equipment is determined, and the first detection matrix is ​​periodically split according to the working cycle. The first column vector in each split cycle is obtained, and the cumulative value of each first column vector is calculated according to the detection weight corresponding to each detection method in the first detection matrix. ; Where L represents the accumulated value of the corresponding first column vector; m represents the total number of elements in the corresponding first column vector; This represents the detection weight corresponding to the detection method at the row position of the i-th element in the first column vector; This represents the detection value of the i-th element in the first column vector; This represents the conversion coefficient used to convert the detection value of the i-th element to a standard format. This represents the adjustment factor of the detection method for the row position of the i-th element on the detection result; Compare the accumulated values ​​corresponding to all splitting periods contained in the first detection matrix, and mark the abnormal columns as the first label; Obtain the first period to which the labeled abnormal column belongs, and count the number of abnormal elements in each row vector of the period matrix corresponding to the first period. When the count value is greater than a preset value, perform a second labeling on the abnormal row of the first period. Based on the first annotation result and the second annotation result, the overlapping annotation elements in each row of the first detection matrix are obtained; According to the detection method of the row with overlapping labeled elements, the corresponding fault analysis model is retrieved, and the position distribution and element size of the overlapping labeled elements in the corresponding row are analyzed to obtain the first fault information of the corresponding first target part. Based on the second detection signal, a second detection matrix for the same second target component at the same time stamp is constructed, and based on a conventional analysis model, the detection elements in the second detection matrix are analyzed to obtain the second fault information corresponding to the second target part. Based on all the first fault information and the second fault information, the fault result of the target electromechanical equipment is obtained, and the fault level is determined. ； Where G represents the fault level; This indicates the effective weight of the information related to the first fault. This indicates the effective weight of the information regarding the second fault information, and m1 represents the total number of first target locations; m2 represents the total number of second target locations; m3 represents the total number of first fault information entries corresponding to the first target locations, and m3 is a variable, with different total number of first fault information entries corresponding to different first target locations; m4 represents the total number of second fault information entries corresponding to the second target locations, and m4 is a variable, with different total number of second fault information entries corresponding to different second target locations. This represents the weight of the j1th first target location; The fault factor represents the j3rd first fault information present in the j1st first target location, and its value range is [0, 1]. The fault factor represents the j4th second fault information existing in the j2nd second target part, and its value range is [0, 1]. This represents the maximum fault factor among all first fault information corresponding to all first target locations; This represents the maximum fault factor among all second fault information corresponding to all second target locations; ln() represents a logarithmic function with base e. When the fault level is greater than the preset level, the fault result is output in conjunction with the fault result, and the character representation of the fault level is enlarged and displayed.

9. An intelligent fault diagnosis system for electromechanical equipment, characterized in that, include: The determination module is used to determine the equipment workflow of the target electromechanical equipment and identify the important sub-processes in the equipment workflow; The detection layout module is used to determine the target component of the target electromechanical equipment for each calibration result according to the preset component-subprocess mapping table, and to perform detection layout on the corresponding target component according to the contribution degree and working attributes of the corresponding calibrated subprocess. A conventional layout module is used to perform conventional layout of the remaining components of the target electromechanical equipment; The fault diagnosis module is used to acquire a first detection signal based on the detection layout, and simultaneously acquire a second detection signal based on the conventional layout, and perform fault diagnosis on the target electromechanical equipment based on the first detection signal and the second detection signal. After determining the target component of the target electromechanical equipment for each calibration result according to the preset component-subprocess mapping table, the process further includes: Based on the preset component-subprocess mapping table, obtain the set of participating components for each important subprocess, wherein the set of participating components includes: the target component corresponding to each calibration result; Each set of participating components is determined based on the labeling layer of the target electromechanical equipment; Based on all annotation layers, a hierarchical annotation model is constructed. At the same time, according to all participating component sets, independent components are planned and the total number of the independent components is determined. Combined with the hierarchical annotation model, the overlap factor of each independent component is determined. Obtain the non-independent components and the first independent components present in each annotation layer, and determine the first number of the first independent components and the second number of the non-independent components in each annotation layer; Determine the first ratio of the first quantity to the total quantity and the second ratio of the second quantity to the total quantity in each annotation layer. At the same time, based on the overlap factor and the process attributes of the important sub-processes corresponding to each annotation layer, assign a weight factor to each annotation layer. The initial layer weights of the corresponding labeling layers are determined based on the first ratio, the second ratio, and the weighting factor for each labeling layer. Obtain the outline overlap ratio and area overlap ratio of each non-independent component and its corresponding independent component in each annotation layer; Based on the proportion of contour overlap and the proportion of area overlap, the proportion weight of the corresponding non-independent component is determined. The proportion weight is then adjusted according to the execution sub-work of the corresponding non-independent component and the execution total work of the corresponding independent component to obtain the first weight of the corresponding non-independent component. At the same time, a corresponding second weight is configured for the first independent component in the corresponding annotation layer according to the process attributes of the important sub-processes of the corresponding annotation layer. Based on all the first weights and all the second weights existing in the same annotation layer, the initial layer weights are adjusted to obtain the current layer weights; Establish a first mapping relationship between the current layer weight and all first weights in the corresponding annotation layer, and a second mapping relationship between the current layer weight and all second weights in the corresponding annotation layer; Based on the first mapping relationship and the second mapping relationship, the contribution level of the corresponding important sub-processes is determined.

Images