Data processing method and apparatus for motor drive chain diagnosis

Abstract

The application discloses a data processing method and device for motor drive link diagnosis. The method comprises the following steps: acquiring to-be-processed data, including topology data and electrical parameter data of a to-be-diagnosed motor drive system; performing model construction processing on the topology data based on a motor drive link to obtain a drive link topology model, which is a model used for representing a link topology of the to-be-diagnosed motor system; performing link section-by-section reasoning processing on the electrical parameter data based on the drive link topology model to obtain first link diagnosis data; and performing diagnosis processing on the first link diagnosis data based on model self-calibration to obtain second link diagnosis data. By constructing a link topology model of the motor system, performing section-by-section reasoning on the link topology model according to electrical parameter data of the motor system, and performing diagnosis processing on diagnosis data obtained through the section-by-section reasoning based on model self-calibration, diagnosis results of the motor drive link are obtained, and the technical effect of improving the intelligent diagnosis accuracy of the motor drive system is achieved.

Description

Technical Field

[0001] This application relates to the field of intelligent diagnostics, and more specifically, to a data processing method and apparatus for diagnosing motor drive links. Background Technology

[0002] In industries such as steel and mining, typical equipment like belt conveyors, water pumps, and fans are driven by electric motors. These devices are numerous, widely distributed, and operate in complex environments. Analyzing the operating conditions and energy efficiency of motor-driven equipment, as well as diagnosing anomalies, are crucial factors in evaluating the performance of motor drive systems. Existing digital diagnostic methods suffer from the following shortcomings: High cost and poor reusability of whole-machine modeling. Traditional methods often involve "customized whole-machine modeling" for individual devices, requiring repeated modeling and parameter tuning for different manufacturers and structures, resulting in long delivery cycles, heavy reliance on experts, and difficulty in large-scale deployment. On-site data often only includes electrical parameters, making it difficult to recreate most mechanical operating conditions. High-frequency data that can be reliably acquired includes electrical parameters such as current, power, voltage, and frequency; while mechanical parameters such as flow rate, torque, shaft power, and load are often missing or have high deployment costs, making it difficult for traditional methods to accurately assess load conditions, efficiency levels, and degradation trends. Strong link coupling leads to weak diagnostic interpretability and difficulty in locating the faulty components. Electrical energy is converted segment by segment in the "inverter-motor-transmission-load" link, and an anomaly in one link can affect the overall electrical parameter performance. Existing black-box identification or threshold methods often only provide "overall anomaly / efficiency decline," making it difficult to pinpoint whether the anomaly belongs to the inverter, motor, drive, or load side. Maintenance / technical upgrade suggestions lack specificity and interpretability.

[0003] This application addresses the problems existing in the intelligent diagnosis of motor drive systems in the prior art. Summary of the Invention

[0004] The main objective of this application is to provide a data processing method and apparatus for diagnosing motor drive links, so as to solve the above-mentioned problems and achieve the technical effect of improving the accuracy of intelligent diagnosis of motor drive systems.

[0005] To achieve the above objectives, the first aspect of this application proposes a data processing method for motor drive link diagnosis, comprising: Acquire the data to be processed, wherein the data to be processed includes the topology data and electrical parameter data of the motor drive system to be diagnosed; The topology data is processed by model building based on the motor drive link to obtain a drive link topology model, wherein the drive link topology model is a model used to represent the link topology of the motor system to be diagnosed. The electrical parameter data is subjected to link-by-segment reasoning based on the drive link topology model to obtain first link diagnostic data, wherein the first link diagnostic data includes data on the operating status of each segment of the motor drive link; The first link diagnostic data is subjected to model self-calibration-based diagnostic processing to obtain the second link diagnostic data.

[0006] Furthermore, the topology data is processed to construct a model based on the motor drive link, resulting in a drive link topology model including: The topology data is processed based on device identifiers to obtain device identifier data, wherein the device identifier data is identifier data used to represent the devices in the motor system to be diagnosed; Match the device component model corresponding to the device identifier data in the preset component model database to obtain the device component model to be constructed; The device component model to be constructed is subjected to link construction processing based on topology relationships to obtain the driving link topology model.

[0007] Further, the electrical parameter data undergoes link-by-segment reasoning processing based on the drive link topology model to obtain the first link diagnostic data, including: The driving link topology model is subjected to first-level component identification processing to obtain first-level component data, wherein the first-level component data is data used to represent the driving link entry component; The electrical parameter data is subjected to first-level inference processing based on the first-level component data to obtain the first-level component output data; The output data of the first-level component is subjected to hierarchical component inference processing based on the driving link topology model to obtain the output data of the load component, and the first link diagnostic data is obtained based on the output data of the load component.

[0008] Furthermore, the first link diagnostic data undergoes model self-calibration-based diagnostic processing to obtain the second link diagnostic data, which includes: The data to be processed is identified to obtain auxiliary diagnostic data, which includes historical operating data and auxiliary on-site data. The component model of the link inference is corrected by performing link inference on the historical operation data to obtain the first update-driven link topology model; The auxiliary field data is subjected to component model correction processing based on working condition constraints to obtain a second updated drive link topology model; The electrical parameter data is processed by link inference based on the update-driven link topology model to obtain the second link diagnostic data, wherein the update-driven link topology model is obtained according to the first update-driven link topology model and the second update-driven link topology model.

[0009] Furthermore, the historical operational data undergoes component model correction processing for link inference to obtain the first update-driven link topology model, which includes: The historical operating data is identified and processed to obtain historical sampled electrical parameter data and historical link load data; The historical sampled electrical parameter data is subjected to link-by-segment inference processing of the driving link topology model to obtain historical link prediction data; The historical link prediction data and the historical link load data are processed by component model correction to obtain the first update-driven link topology model.

[0010] Furthermore, the auxiliary field data undergoes component model correction processing based on operating condition constraints to obtain the second updated drive link topology model, which includes: The auxiliary field data is subjected to working condition clustering processing to obtain normal working condition data, wherein the normal working condition data is data used to represent the normal working condition domain obtained by clustering; Error construction processing is performed on the first link diagnostic data and the normal operating condition data to obtain operating condition error data; The component model of the drive link topology model is corrected based on the operating condition error data to obtain the second updated drive link topology model.

[0011] Furthermore, the drive link topology model includes multiple device component models, including power conversion and control components, motor components, transmission components, and load components.

[0012] According to a second aspect of this application, a data processing apparatus for diagnosing motor drive links is provided, comprising: The data acquisition module is used to acquire data to be processed, wherein the data to be processed includes topology data and electrical parameter data of the motor drive system to be diagnosed; The link construction module is used to perform model construction processing on the topology data based on the motor drive link to obtain a drive link topology model, wherein the drive link topology model is a model used to represent the link topology of the motor system to be diagnosed. The first diagnostic module is used to perform link-by-segment reasoning processing on the electrical parameter data based on the drive link topology model to obtain the first link diagnostic data, wherein the first link diagnostic data includes the operating status data of each segment of the motor drive link. The second diagnostic module is used to perform model self-calibration-based diagnostic processing on the first link diagnostic data to obtain the second link diagnostic data.

[0013] According to a third aspect of this application, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing the computer to perform the above-described data processing method for motor drive link diagnosis.

[0014] According to a fourth aspect of this application, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to cause the at least one processor to perform the above-described data processing method for motor drive link diagnosis.

[0015] The technical solutions provided by the embodiments of this application may include the following beneficial effects: In this application, data to be processed is obtained, including topology data and electrical parameter data representing the motor drive system to be diagnosed. The topology data is then processed using a model construction based on the motor drive link to obtain a drive link topology model, which represents the link topology of the motor system to be diagnosed. The electrical parameter data is then processed using a link-by-segment inference process based on the drive link topology model to obtain first link diagnostic data. Finally, the first link diagnostic data is processed using a model self-calibration process to obtain second link diagnostic data. By constructing a link topology model of the motor system, performing segment-by-segment inference on the link topology model based on the electrical parameter data of the motor system, and performing model self-calibration diagnostic processing on the diagnostic data obtained from the segment-by-segment inference, diagnostic results for the motor drive link are obtained, thus achieving intelligent inference processing of the motor drive system. Attached Figure Description

[0016] The accompanying drawings, which form part of this application, are used to provide a further understanding of the application and to make other features, objects, and advantages of the application more apparent. The illustrative embodiments and descriptions of this application are used to explain the application and do not constitute an undue limitation of the application. In the drawings: Figure 1 A flowchart of a data processing method for diagnosing motor drive links provided in this application; Figure 2 A flowchart of a data processing method for diagnosing motor drive links provided in this application; Figure 3 A flowchart of a data processing method for diagnosing motor drive links provided in this application; Figure 4 A flowchart of a data processing method for diagnosing motor drive links provided in this application; Figure 5 This is a schematic diagram of a data processing device for diagnosing motor drive links, provided in this application. Detailed Implementation

[0017] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0018] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0019] In this application, the terms "upper," "lower," "left," "right," "front," "rear," "top," "bottom," "inner," "outer," "middle," "vertical," "horizontal," "lateral," and "longitudinal" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are primarily for the purpose of better describing this application and its embodiments, and are not intended to limit the indicated device, element, or component to having a specific orientation, or to be constructed and operated in a specific orientation.

[0020] Furthermore, in addition to indicating location or positional relationship, some of the aforementioned terms may also have other meanings. For example, the term "above" may also be used in some cases to indicate a certain dependency or connection relationship. Those skilled in the art can understand the specific meaning of these terms in this application based on the specific circumstances.

[0021] Furthermore, the terms "installation," "setup," "equipped with," "connection," "linked," and "socketing" should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral structure; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or an internal connection between two devices, components, or parts. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0022] In some optional embodiments of this application, a data processing method for diagnosing motor drive links is provided. Figure 1 A flowchart of a data processing method for motor drive link diagnosis provided in this application is shown below. Figure 1 As shown, the method includes the following steps: S101: Obtain the data to be processed; The data to be processed includes topology data and electrical parameter data of the motor drive system to be diagnosed. The data to be processed includes time series data of electrical parameters received from the grid side or inverter side from the field acquisition device or edge gateway, as well as the actual topology of the field devices in the field motor drive system, which is used to represent the connection relationship of each device in the motor drive system.

[0023] S102: Perform model building processing on the topology data based on the motor drive link to obtain the drive link topology model; The drive link topology model is a model used to represent the link topology of the motor system to be diagnosed; in some optional embodiments of this application, the drive link topology model includes multiple device component models, and the device components include power conversion and control components, motor components, transmission components, and load components.

[0024] In some optional embodiments of this application, a data processing method for motor drive link diagnosis is provided to construct a motor drive link topology model. Figure 2 A flowchart of a data processing method for motor drive link diagnosis provided in this application is shown below. Figure 2 As shown, the method includes the following steps: S201: Perform device identification processing on the topology data to obtain device identification data; Device identification data is used to represent the identification data of devices in the motor system to be diagnosed. The topology data includes the devices in the motor drive system and the connection relationships between the devices. The device identification of the motor devices in the topology data is identified to identify all the devices in the motor drive system so as to build the component link model corresponding to the motor drive system based on the identified devices.

[0025] S202: Match the device component model corresponding to the device identification data in the preset component model database to obtain the device component model to be constructed; The pre-set component model database stores multiple component models. Each component model is a functional unit in the motor drive chain that can independently describe the "energy / motion / state transfer relationship." These component models include: power conversion and control components (such as frequency converters, soft starters, and other power regulation / control devices); motor components (such as asynchronous / synchronous / permanent magnet motors and other electromechanical conversion components); transmission components (mechanical transmission components such as couplings, pulleys, reducers, gearboxes, hydraulic couplings, brakes / clutches, etc.); and load components (driven load components such as pumps, fans, belt conveyors, and other motor-driven equipment). Furthermore, each component model is encapsulated as a unified input / output interface. The input interface receives the output of the previous level component and the necessary state characteristics of the current level component; the output interface receives the output of the current level component. Component models can be mechanistic / empirical models, data fitting models, and / or hybrid models. They can be obtained from equipment manufacturer manuals, manufacturer performance curves, short-term field calibration, and long-term operational data statistical analysis results, and are stored in the pre-set component model database. If the component models stored in the preset component model database are incomplete or missing due to manufacturer curves / parameters, they can be self-calibrated based on the accessed operating data. This process corrects the missing parameters in the component models, continuously improving the similarity between the component models and the actual equipment's operating characteristics during the diagnosis of the motor drive system, thereby enhancing the accuracy of the diagnosis.

[0026] In some optional embodiments of this application, the following device component models are provided, including: The model construction of power conversion components, such as frequency converter components, involves obtaining the input power and equivalent losses on the motor side based on the available electrical parameter data from the grid side / frequency converter side. The input interface data includes: active power P_in, current, voltage, power factor, etc., from the grid side / frequency converter input side; output frequency f_out (or obtained from frequency / speed signals); frequency converter rated power, control method, manufacturer's efficiency curve, etc. The output interface data includes: the equivalent input power P_vfd_out from the frequency converter to the motor; the equivalent losses L_vfd and efficiency η_vfd of the frequency converter. The calculation relationship of the component model is P_vfd_out = P_in. η_vfd(f_out, load).

[0027] The electro-mechanical conversion component model is constructed, such as the motor component, which converts the input power on the motor side into mechanical characteristics such as shaft power / speed, and estimates the loss composition and health-related features. The input interface data includes: motor input power P_vfd_out (or motor-side active power), frequency f_out, motor nameplate parameters: rated power, rated efficiency level, rated voltage / current, number of pole pairs / synchronous speed, power factor, etc., as well as winding temperature / temperature rise, vibration, etc. The output interface output data includes: motor shaft power P_shaft, motor efficiency η_m(f, load) and equivalent loss L_m, speed n. The calculation relationship of the component model is: P_shaft = P_vfd_out η_m(f_out, load).

[0028] Mechanical transmission component models, such as couplings and reducers, are constructed to reflect the transmission of mechanical energy in the transmission chain, the changes in transmission ratio, and the effects of efficiency / deterioration. The input interface includes: power / speed of the previous stage shaft (P_in, n_in), and transmission structure parameters: transmission ratio i, rated efficiency, and type (gear / belt / hydraulic, etc.). The output interface includes: power / speed of the next stage shaft (P_out, n_out); transmission efficiency η_tr; and equivalent loss / friction characteristic L_tr. The calculation relationship of the component model is n_out = n_in / i, P_out = P_in. η_tr(load, wear).

[0029] The driven load equipment model is constructed, such as water pumps, fans, and belt conveyors. The power / speed of the end shaft is mapped to the load operating range, efficiency level, and health / abnormal characteristics. The input interface includes: load-side shaft power P_load, speed n_load (or frequency), and equipment static parameters. The output interface outputs: equipment operating range (e.g., the flow-head / air volume-pressure range for pumps / fans, the load / resistance range for belt conveyors), efficiency / energy consumption level, and load health characteristics. The calculation relationships of the component models are as follows: For the water pump model, the operating point range and efficiency level are estimated based on the manufacturer's performance curves, combined with shaft power and speed, and health indicators related to scaling / clogging / cavitation / internal leakage are output; the fan model is based on the fan QP-η... Curves or historical operating condition domains, combined with end shaft power and speed, yield airflow / pressure ranges and characterizations such as dust accumulation and surge risk; Belt conveyor model: Based on the relationship between belt conveyor resistance, load, speed and shaft power, load / resistance ranges are obtained, and combined with historical no-load / heavy-load characteristic domains, outputting no-load, abnormal feeding and other state characterizations.

[0030] S203: Perform link construction processing based on topology relationships on the device component model to be constructed to obtain the driving link topology model.

[0031] The multiple equipment component models in the above-mentioned equipment component model to be constructed are assembled according to the actual link topology on site, such as: frequency converter-motor-coupling-reducer-fan. Based on the above topology link, the above equipment component models are assembled in series, and the output of the previous level component is received through the above-mentioned input interface, and the output of the current level component is output to the next level component through the above-mentioned output interface.

[0032] S103: Perform link-by-link reasoning processing on the electrical parameter data based on the drive link topology model to obtain the first link diagnostic data; The first link diagnostic data includes data on the operating status of each segment of the motor drive link, including: input / output power, efficiency / loss characterization of each segment of the link; load-side operating condition range and energy efficiency level; healthy / abnormal locations (link link level) and trends.

[0033] The electrical parameter data are time series data of electrical parameters received from the grid side or frequency converter side by the field acquisition device or edge gateway, including: active power P_in(k), voltage U(k), current I(k), power factor cosφ(k), output frequency f_out(k) (or equivalent speed signal), where k is the sampling number.

[0034] The above time series is subjected to denoising and outlier removal processing, such as using moving median filtering, moving mean filtering or three sigma filtering to remove isolated spikes, to obtain the smoothed electrical parameter sequence: P_in_tilde(k), U_tilde(k), I_tilde(k), f_out_tilde(k).

[0035] In some optional embodiments of this application, a data processing method for motor drive link diagnosis is provided to achieve link-by-link reasoning of the motor drive link topology model. Figure 3 A flowchart of a data processing method for motor drive link diagnosis provided in this application is shown below. Figure 3 As shown, the method includes the following steps: S301: Perform first-level component identification processing on the drive link topology model to obtain first-level component data; The first-level component data is used to represent the data of the drive link entry component. The first-level component is the link entry component of the drive link. For example, if the frequency converter is the first-level component, the first-level component data corresponding to the frequency converter is obtained. The input state vector of the first-level component is defined as: x1(k) = [P_in_tilde(k),U_tilde(k),I_tilde(k),f_out_tilde(k),cosφ(k),s1(k)], where s1(k) is the state feature (such as operating mode, start-stop state, etc.) related to the first-level component. S302: Perform first-level inference processing on the electrical parameter data based on the first-level component data to obtain the first-level component output data; Based on the aforementioned electrical parameter data and the first-level component model, the link undergoes component-level forward inference processing. For the i-th component in the link (e.g., inverter, motor, reducer, or load device), the formula is: y_i(k) = f_i(x_i(k), theta_i), where x_i(k): the input state vector of the i-th component at sampling time k (containing at least input power, input speed, or frequency); y_i(k): the output state vector of the i-th component at sampling time k (containing at least output power, output speed, and efficiency / loss); and theta_i: the parameter vector related to the i-th component, such as rated efficiency, empirical efficiency curve parameters, wear offset factor, etc. Adjacent components are connected by the following recursive relationship: x_(i+1)(k) = g_i( y_i(k) ), and the output power, speed and other states of the i-th stage component are directly used as the input boundary conditions of the i+1-th stage component.

[0036] S303: Perform hierarchical component inference processing based on the driving link topology model on the output data of the first-level component to obtain the output data of the load component, and obtain the first link diagnostic data based on the output data of the load component.

[0037] In an optional embodiment of this application, the component-level segmented reasoning of the drive link topology model includes: inputting electrical parameters / active power from the grid side or the inverter side; inferring the input power from the motor side using the inverter model; inferring the motor shaft power and speed characteristics using the motor model; transmitting the shaft power / speed segment by segment to the load side using the mechanical transmission model; and outputting the load operating range, efficiency level, and health characteristics from the load model to obtain the aforementioned first link diagnostic data.

[0038] S104: Perform model self-calibration-based diagnostic processing on the first link diagnostic data to obtain the second link diagnostic data.

[0039] The second diagnostic data includes energy efficiency assessment of the motor drive system and anomaly location data, etc. The location processing of abnormal components is realized based on the input and output of different component models in the second link.

[0040] In some optional embodiments of this application, a data processing method for motor drive link diagnosis is provided to achieve link diagnosis processing based on model self-calibration. Figure 4 A flowchart of a data processing method for motor drive link diagnosis provided in this application is shown below. Figure 4 As shown, the method includes the following steps: S401: Identify the data to be processed to obtain auxiliary diagnostic data; Auxiliary diagnostic data includes historical operating data and auxiliary field data. Historical operating data is periodic historical operating data, which includes historical sampled electrical parameter data and at least one historical observation data that can characterize the link load status or component operating status. Auxiliary field data is multi-sensor observation data when multiple other sensors are configured on-site, such as pressure / differential pressure, flow rate, temperature, vibration, valves, start / stop signals, etc.

[0041] S402: Perform component model correction processing on historical operation data for link inference to obtain the first update-driven link topology model; Historical operating data is sampled using a time window. A time window W of length T is selected, and forward inference is performed on all samples within the window to obtain the corresponding predicted quantity z_j_pred(k). This predicted quantity is then compared with the actual observation z_j_meas(k) of the historical operating data to correct the component model.

[0042] In some optional embodiments of this application, a data processing method for motor drive link diagnosis is provided to achieve component model correction based on link inference. The method includes: Historical operating data is identified and processed to obtain historical sampled electrical parameter data and historical link load data; the historical sampled electrical parameter data is processed by link-by-link inference of the driving link topology model to obtain historical link prediction data; the historical link prediction data and historical link load data are processed by component model correction to obtain the first updated driving link topology model.

[0043] S403: Perform component model correction processing based on working condition constraints on the auxiliary field data to obtain the second updated drive link topology model; If the complete historical operating data of the motor drive system to be diagnosed is not available, multi-sensor data collected by multiple sensors, including historical multi-sensor data, is acquired, and component model correction processing based on operating condition constraints is performed based on this auxiliary field data.

[0044] In some optional embodiments of this application, a data processing method for motor drive link diagnosis is provided to achieve component model correction based on operating condition constraints. The method includes: The auxiliary field data is subjected to working condition clustering to obtain normal working condition data, which is used to represent the normal working condition domain obtained by clustering. The first link diagnostic data and normal working condition data are subjected to error construction processing to obtain working condition error data. The driving link topology model is modified according to the working condition error data to obtain the second updated driving link topology model.

[0045] In an optional embodiment of this application, the above-mentioned component correction process based on link inference includes: performing consistency error construction processing based on the above-mentioned predicted quantities and actual observations, where the errors of various physical quantities are e_j(k) = z_j_pred(k). z_j_meas(k) is the weighted mean square error within the time window: J(theta_c) = Σ_j w_j Σ_{k∈W} e_j(k)^2, where e_j(k): the consistency error of the j-th type of physical quantity at time k; w_j: the weight coefficient corresponding to the j-th type of physical quantity; theta_c: the subset of parameters to be corrected (including parameters related to the corresponding sensors in the inverter, motor, transmission and load models); J(theta_c): the total consistency error function within the current time window.

[0046] In an optional embodiment of this application, the above-mentioned component correction process based on operating condition constraints includes: performing consistency error construction processing according to the above-mentioned normal operating condition domain, e_domain(k) = d(w(k), Omega_normal), where w(k): operating condition feature points composed of multidimensional prediction quantities (e.g., [Q, H, eta]); Omega_normal: normal operating condition domain formed by historical normal data; e_domain(k) = d(w(k), Omega_normal) is the operating condition domain error between the prediction point and the normal operating condition domain.

[0047] Based on the consistency error obtained above, the parameters of each model component in the drive link topology model are updated. For example, the overall offset and shape parameters of the load performance curve are corrected using load-side pressure / flow; the copper loss / iron loss and friction loss coefficient in the motor loss model are corrected using motor temperature / vibration; and the wear offset in the transmission efficiency is corrected using gearbox temperature / vibration.

[0048] Among them, through the formula: theta_c = argmin_theta_c [ J(theta_c) + lambda ||theta_c Solve for theta_c0||^2, where theta_c To calibrate the corresponding component model parameters, batch calibration based on multi-sensor data is achieved. Furthermore, when continuous online self-calibration is required, recursive algorithms such as Recursive Least Squares (RLS) are employed, driven by the time window comprehensive error, to perform small-step incremental updates on theta_c, and the weight of the latest data is increased through a forgetting factor. Here, theta_c0 represents the current (before calibration) parameter vector; theta_c : The updated parameter vector; lambda: Regularization coefficient, used to limit the magnitude of parameter shift and avoid overfitting; ||·||: Norm of the parameter vector (e.g., L2 norm); RLS: Recursive Least Squares algorithm, used for online recursive parameter estimation.

[0049] S404: The electrical parameter data is processed by link inference based on the updated drive link topology model to obtain the second link diagnostic data.

[0050] The updated driving link topology model is obtained based on the first and second updated driving link topology models.

[0051] In an optional embodiment of this application, the confidence level of the first link diagnostic data is output using the updated drive link topology model described above. Based on the residual level, the number of samples involved in the correction, and the magnitude of parameter changes, the reliability of the inference result is classified into confidence levels A / B / C / D, and output along with the energy efficiency diagnosis, health diagnosis, and link location results. When there is limited field sensor data, the confidence level of the first link diagnostic data can be improved and output by using a model correction based on the step-by-step inference from historical operating data. When there is abundant field sensor data, the model correction can be based on the operating condition constraints of the field auxiliary data to achieve consistency correction processing based on multi-sensor data, thereby improving the accuracy of the link diagnostic data.

[0052] In some optional embodiments of this application, a data processing device for diagnosing motor drive links is provided. Figure 5 A schematic diagram of a data processing device for diagnosing motor drive links provided in this application is shown below. Figure 5 As shown, the device includes: The data acquisition module 51 is used to acquire data to be processed, wherein the data to be processed includes topology data and electrical parameter data representing the motor system to be diagnosed; Link construction module 52 is used to perform model construction processing on topology data based on motor drive link to obtain drive link topology model, wherein the drive link topology model is a model used to represent the link topology of the motor system to be diagnosed. The first diagnostic module 53 is used to perform link-by-link reasoning processing on the electrical parameter data based on the driving link topology model to obtain the first link diagnostic data. The second diagnostic module 54 is used to perform model self-calibration-based diagnostic processing on the first link diagnostic data to obtain the second link diagnostic data.

[0053] The specific methods of execution of each unit in the above embodiments have been described in detail in the embodiments of the method, and will not be elaborated here.

[0054] In summary, the process involves acquiring data to be processed, including topology data and electrical parameter data representing the motor drive system to be diagnosed; performing model construction processing based on the motor drive link on the topology data to obtain a drive link topology model, wherein the drive link topology model is a model representing the link topology of the motor system to be diagnosed; performing link-by-segment inference processing based on the drive link topology model on the electrical parameter data to obtain first link diagnostic data; and performing model self-calibration diagnostic processing on the first link diagnostic data to obtain second link diagnostic data. By constructing a link topology model of the motor system, performing segment-by-segment inference on the link topology model based on the electrical parameter data of the motor system, and performing model self-calibration diagnostic processing on the diagnostic data obtained from the above segment-by-segment inference, the diagnostic results of the motor drive link are obtained, realizing intelligent inference processing of the motor drive system.

[0055] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0056] Obviously, those skilled in the art should understand that the various units or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device, or fabricating them separately as individual integrated circuit modules, or fabricating multiple modules or steps into a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0057] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A data processing method for diagnosing motor drive links, characterized in that, include: Acquire the data to be processed, wherein the data to be processed includes the topology data and electrical parameter data of the motor drive system to be diagnosed; The topology data is processed by model building based on the motor drive link to obtain a drive link topology model, wherein the drive link topology model is a model used to represent the link topology of the motor system to be diagnosed. The electrical parameter data is subjected to link-by-segment reasoning based on the drive link topology model to obtain first link diagnostic data, wherein the first link diagnostic data includes data on the operating status of each segment of the motor drive link; The first link diagnostic data is subjected to model self-calibration-based diagnostic processing to obtain the second link diagnostic data.

2. The data processing method according to claim 1, characterized in that, The topology data is processed to construct a model based on the motor drive link, resulting in a drive link topology model including: The topology data is processed based on device identifiers to obtain device identifier data, wherein the device identifier data is identifier data used to represent the devices in the motor system to be diagnosed; Match the device component model corresponding to the device identifier data in the preset component model database to obtain the device component model to be constructed; The device component model to be constructed is subjected to link construction processing based on topology relationships to obtain the driving link topology model.

3. The data processing method according to claim 1, characterized in that, The electrical parameter data is subjected to link-by-segment inference processing based on the drive link topology model to obtain the first link diagnostic data, which includes: The driving link topology model is subjected to first-level component identification processing to obtain first-level component data, wherein the first-level component data is data used to represent the driving link entry component; The electrical parameter data is subjected to first-level inference processing based on the first-level component data to obtain the first-level component output data; The output data of the first-level component is subjected to hierarchical component inference processing based on the driving link topology model to obtain the output data of the load component, and the first link diagnostic data is obtained based on the output data of the load component.

4. The data processing method according to claim 1, characterized in that, The first link diagnostic data is processed based on model self-calibration to obtain the second link diagnostic data, which includes: The data to be processed is identified to obtain auxiliary diagnostic data, which includes historical operating data and auxiliary on-site data. The component model of the link inference is corrected by performing link inference on the historical operation data to obtain the first update-driven link topology model; The auxiliary field data is subjected to component model correction processing based on working condition constraints to obtain a second updated drive link topology model; The electrical parameter data is processed by link inference based on the update-driven link topology model to obtain the second link diagnostic data, wherein the update-driven link topology model is obtained according to the first update-driven link topology model and the second update-driven link topology model.

5. The data processing method according to claim 4, characterized in that, The component model of the link inference is corrected by performing link inference on the historical running data to obtain the first update-driven link topology model, which includes: The historical operating data is identified and processed to obtain historical sampled electrical parameter data and historical link load data; The historical sampled electrical parameter data is subjected to link-by-segment inference processing of the driving link topology model to obtain historical link prediction data; The historical link prediction data and the historical link load data are processed by component model correction to obtain the first update-driven link topology model.

6. The data processing method according to claim 4, characterized in that, The auxiliary field data is subjected to component model correction processing based on operating condition constraints to obtain the second updated drive link topology model, which includes: The auxiliary field data is subjected to working condition clustering processing to obtain normal working condition data, wherein the normal working condition data is data used to represent the normal working condition domain obtained by clustering; Error construction processing is performed on the first link diagnostic data and the normal operating condition data to obtain operating condition error data; The component model of the drive link topology model is corrected based on the operating condition error data to obtain the second updated drive link topology model.

7. The data processing method according to claim 1, characterized in that, The drive link topology model includes multiple device component models, including power conversion and control components, motor components, transmission components, and load components.

8. A data processing device for diagnosing motor drive links, characterized in that, include: The data acquisition module is used to acquire data to be processed, wherein the data to be processed includes topology data and electrical parameter data of the motor drive system to be diagnosed; The link construction module is used to perform model construction processing on the topology data based on the motor drive link to obtain a drive link topology model, wherein the drive link topology model is a model used to represent the link topology of the motor system to be diagnosed. The first diagnostic module is used to perform link-by-segment reasoning processing on the electrical parameter data based on the drive link topology model to obtain the first link diagnostic data, wherein the first link diagnostic data includes the operating status data of each segment of the motor drive link. The second diagnostic module is used to perform model self-calibration-based diagnostic processing on the first link diagnostic data to obtain the second link diagnostic data.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the data processing method for motor drive link diagnosis as described in any one of claims 1-7.

10. An electronic device, characterized in that, include: At least one processor; And a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to cause the at least one processor to perform the data processing method for motor drive link diagnosis as described in any one of claims 1-7.

Images