A method and system for managing the operating state of industrial park equipment

Abstract

The application relates to the technical field of device management, and discloses a management method and system for the running state of devices in an industrial park. The method comprises the following steps: obtaining device basic information, historical fault data and real-time running data of the devices; performing category analysis on the devices according to the device basic information to obtain the device categories; performing fault cycle prediction on the devices based on a pre-trained support vector regression model according to the historical fault data and the device categories to obtain the device fault cycles; performing health index calculation on the devices according to the real-time running data and the device categories to obtain the device health indexes; performing fault prediction according to the device categories, the device fault cycles and the device health indexes to obtain a predicted fault probability, and sending the predicted fault probability to an operation and maintenance end so that the operation and maintenance end performs device maintenance according to the predicted fault probability. The method can improve the device management efficiency of the industrial park and realize effective analysis and management of the device states of the industrial park.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of device management, and in particular to a management method and system for the running state of devices in an industrial park. BACKGROUND

[0002] With the continuous promotion and implementation of the concept of intelligent manufacturing, the information acquisition, analysis and management of devices in an industrial park have attracted more and more attention.

[0003] The running state of a device not only relates to the working cost of an industrial park, but also affects the production progress of the entire industrial park. In terms of the analysis of the running state of devices in the current industrial park, the traditional management method of an industrial park only maintains according to the historical maintenance of a device, and a preventive maintenance method mainly based on a fault tree makes corresponding predictions through the maintenance records of a device. However, this method cannot consider real-time data in the running process of a device. This maintenance mode belongs to scheduled maintenance and response maintenance, and cannot predict whether a device failure will occur in advance.

[0004] How to accurately predict the running state of devices in an industrial park to perform intelligent and proactive maintenance and improve the management level of device assets is a major focus and difficulty in the management of the running state of devices in an industrial park. However, the existing technology still has some deficiencies in comprehensively considering real-time data and historical maintenance data in the running of a device. SUMMARY

[0005] The present application provides a management method and system for the running state of devices in an industrial park, which can comprehensively consider historical failure data and real-time running data of a specific device category, and then obtain a predicted failure probability, so that the operation and maintenance end can perform device maintenance according to the predicted failure probability, and effectively analyze and manage the state of devices in an industrial park.

[0006] In a first aspect, to solve the above technical problems, the present application provides a management method for the running state of devices in an industrial park, comprising:

[0007] obtaining device basic information, historical failure data and real-time running data of a device;

[0008] performing category analysis on the device according to the device basic information to obtain a device category; wherein the device category includes a single-machine device, an intermittent process device and a continuous process device;

[0009] performing failure cycle prediction on the device based on a pre-trained support vector regression model according to the historical failure data and the device category to obtain a device failure cycle;

[0010] According to the real-time operation data and the equipment category, a health index of the equipment is calculated to obtain an equipment health index; wherein the equipment health index comprises a single-machine equipment health index, an intermittent process equipment health index and a continuous process equipment health index;

[0011] According to the equipment category, the equipment failure cycle and the equipment health index, a failure prediction is performed to obtain a predicted failure probability, and the predicted failure probability is sent to an operation and maintenance end, so that the operation and maintenance end performs equipment maintenance according to the predicted failure probability.

[0012] Preferably, the category analysis of the equipment according to the equipment basic information comprises:

[0013] According to the equipment basic information, a PCA dimension reduction operation is performed to obtain reduced dimension data;

[0014] According to the reduced dimension data, the equipment is classified based on a clustering algorithm to obtain the equipment category.

[0015] Preferably, the PCA dimension reduction operation according to the equipment basic information comprises:

[0016] According to the equipment basic information, a matrix construction is performed to obtain a feature matrix; wherein the feature matrix is composed of the equipment basic information, a row of the feature matrix represents each piece of equipment, and a column of the feature matrix represents a feature of the equipment;

[0017] The feature matrix is standardized to obtain a standardized matrix;

[0018] The covariance of the standardized matrix is calculated to obtain a covariance matrix;

[0019] The covariance matrix is decomposed in terms of eigenvalues and eigenvectors to obtain principal components;

[0020] According to the principal components, data screening is performed to obtain the reduced dimension data.

[0021] Preferably, the training process of the support vector regression model comprises:

[0022] The first failure cycle and the second failure cycle are obtained;

[0023] The first failure cycle and the second failure cycle are standardized to obtain first pre-processing data and second pre-processing data;

[0024] Based on the first pre-processing data and the second pre-processing data as input data, a support vector regression model is constructed based on a kernel function, the support vector regression model is trained to obtain a trained support vector regression model.

[0025] Preferably, the health index of the equipment is calculated according to the real-time operation data and the equipment category, including:

[0026] The health index of the single-machine equipment is calculated by the following formula:

[0027] ;

[0028] The health index of the intermittent-flow equipment is calculated by the following formula:

[0029] ;

[0030] The health index of the continuous-flow equipment is calculated by the following formula:

[0031] ;

[0032] In the formula, is the health index of the single-machine equipment; is the health index of the intermittent-flow equipment; is the health index of the continuous-flow equipment; is the running time of the i-th equipment; is the average running time of the equipment; is the failure state of the i-th running of the equipment, taking the value of 1 if a failure occurs, and taking the value of 0 if no failure occurs; is the total running number of the equipment.

[0033] Preferably, the failure probability is predicted according to the equipment category, the equipment failure cycle and the equipment health index, including:

[0034] The failure probability of the single-machine equipment is calculated by the following formula:

[0035] ;

[0036] The failure probability of the intermittent-flow equipment is calculated by the following formula:

[0037] ;

[0038] The failure probability of the continuous-flow equipment is calculated by the following formula:

[0039] ;

[0040] In the formula, is the failure probability of the single-machine equipment; is the failure probability of the intermittent-flow equipment; is the failure probability of the continuous-flow equipment; is the latest equipment failure cycle; is an average value of equipment historical failure cycles; is a single machine equipment health index; is an intermittent process equipment health index; is a continuous process equipment health index; is the i-th equipment failure cycle; is a total number of historical failure cycles; is a single machine equipment weight; is an intermittent process equipment weight; is a continuous process equipment weight; is a cumulative failure impact coefficient.

[0041] Preferably, the predicted failure probability is sent to an operation and maintenance end, so that the operation and maintenance end performs equipment maintenance according to the predicted failure probability, including:

[0042] When the predicted failure probability is greater than and equal to a first preset threshold, it is determined that the failure risk level is high risk, and an emergency maintenance operation is performed;

[0043] When the predicted failure probability is less than the first preset threshold and greater than a second preset threshold, it is determined that the failure risk level is medium risk, and a preventive maintenance operation is performed;

[0044] When the predicted failure probability is less than and equal to the second preset threshold, it is determined that the failure risk level is low risk, and a regular maintenance operation is performed.

[0045] In a second aspect, the present application provides an industrial park equipment operation state management system, comprising:

[0046] A data acquisition module is configured to acquire equipment basic information, historical failure data and real-time operation data of the equipment;

[0047] A category analysis module is configured to perform category analysis on the equipment according to the equipment basic information, and obtain an equipment category; wherein the equipment category includes single machine equipment, intermittent process equipment and continuous process equipment;

[0048] A cycle prediction module is configured to perform failure cycle prediction on the equipment based on a pre-trained support vector regression model according to the historical failure data and the equipment category, and obtain an equipment failure cycle;

[0049] A health calculation module is configured to perform health index calculation on the equipment according to the real-time operation data and the equipment category, and obtain an equipment health index; wherein the equipment health index includes a single machine equipment health index, an intermittent process equipment health index and a continuous process equipment health index;

[0050] A fault prediction module is configured to perform fault prediction according to the equipment category, the equipment failure cycle and the equipment health index, obtain a predicted failure probability, and send the predicted failure probability to the operation and maintenance end, so that the operation and maintenance end performs equipment maintenance according to the predicted failure probability.

[0051] In a third aspect, the present application further provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the management method of the running state of the equipment in the industrial park when executing the computer program.

[0052] In a fourth aspect, the present application further provides a computer readable storage medium, comprising a stored computer program, wherein the computer readable storage medium performs the management method of the running state of the equipment in the industrial park when the computer program runs.

[0053] Compared with the prior art, the present application has the following beneficial effects: the embodiments of the present application provide a management method and system for the running state of equipment in an industrial park. The method comprises: obtaining equipment basic information, historical failure data and real-time running data of the equipment; performing category analysis on the equipment according to the equipment basic information to obtain an equipment category; wherein the equipment category comprises single-machine equipment, intermittent process equipment and continuous process equipment; performing failure cycle prediction on the equipment based on a pre-trained support vector regression model according to the historical failure data and the equipment category to obtain an equipment failure cycle; performing health index calculation on the equipment according to the real-time running data and the equipment category to obtain an equipment health index; wherein the equipment health index comprises single-machine equipment health index, intermittent process equipment health index and continuous process equipment health index; performing fault prediction according to the equipment category, the equipment failure cycle and the equipment health index to obtain a predicted failure probability, and sending the predicted failure probability to an operation and maintenance end to enable the operation and maintenance end to perform equipment maintenance according to the predicted failure probability.

[0054] In the present application, the method can analyze the basic information, historical failure data and real-time running data of the equipment, and then infer the health state and failure cycle of the equipment. Specifically, the method comprises performing category analysis on the equipment, performing failure cycle prediction using a support vector regression model, and calculating the health index of the equipment according to real-time running data. In addition, the health index and failure cycle of the equipment are combined to perform fault prediction, thereby obtaining a predicted failure probability, and finally providing the prediction result to the operation and maintenance end for corresponding equipment maintenance operation. The method can improve the equipment management efficiency of the industrial park and realize effective analysis and management of the equipment state in the industrial park. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 is a flowchart of the management method of the running state of the equipment in the industrial park provided by the first embodiment of the present application;

[0056] Figure 2 is a structural diagram of the management system of the running state of the equipment in the industrial park provided by the second embodiment of the present application. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0058] With reference to Figure 1 , the first embodiment of the present application provides a management method of the running state of the equipment in the industrial park, comprising the following steps:

[0059] S11, obtaining equipment basic information, historical failure data and real-time running data of the equipment.

[0060] S12, performing category analysis on the equipment according to the equipment basic information to obtain the equipment category.

[0061] S13, performing failure cycle prediction on the equipment based on a pre-trained support vector regression model according to the historical failure data and the equipment category to obtain the equipment failure cycle.

[0062] S14, performing health index calculation on the equipment according to the real-time running data and the equipment category to obtain the equipment health index.

[0063] S15, performing failure prediction according to the equipment category, the equipment failure cycle and the equipment health index to obtain a predicted failure probability, and sending the predicted failure probability to an operation and maintenance end to enable the operation and maintenance end to perform equipment maintenance according to the predicted failure probability.

[0064] It is worth noting that the failure prediction of the equipment in the industrial park is to predict the probability of possible failure of the equipment by using data analysis technology and model, so as to realize early warning and take corresponding maintenance measures. The core purpose is to identify potential failure risks in advance through continuous monitoring and analysis of equipment running data, to avoid production stagnation or equipment damage caused by sudden failure.

[0065] In order to facilitate the understanding of the present application, some preferred embodiments of the present application will be described further below.

[0066] In step S11, the device basic information, historical failure data and real-time operation data of the device are acquired.

[0067] Specifically, the device basic information includes characteristic data of the device, such as model, specification, operation parameter, working environment. Optionally, these information can be derived from the technical manual of the device, the document provided by the manufacturer, or collected through the device management system. It is worth noting that in the subsequent steps, in order to more efficiently process these information and classify, PCA (Principal Component Analysis) dimensionality reduction operation is applied to these device basic information. The purpose of dimensionality reduction is to reduce the dimension of data while retaining as much original information as possible.

[0068] Specifically, the historical failure data includes the failure records of the device in the past, including the time, type, severity, etc. of the failure. In the method of the present application, the historical failure data includes the first failure cycle, which is the time cycle of the first failure of the device, used to calibrate the initial failure cycle of the device; the second failure cycle, which refers to the time cycle of the second failure of the device. By analyzing the difference between the first and second failure cycles, the regularity of device failure can be better understood; the failure cycle refers to the cycle data of each failure, indicating the time interval from the last failure to the next failure. These data help to establish a prediction model to predict the future failure cycle. The historical failure data is used for failure cycle prediction. Specifically, by analyzing the historical failure data of the device, combined with the device category (such as single machine device, intermittent device, continuous device), the support vector regression model (SVR) is used to predict the possible future failure cycle of the device.

[0069] It is worth noting that the main application of these historical failure data in the present application is: for failure cycle prediction, by analyzing the regularity of historical failure cycle, the future failure cycle of the device is predicted; for calculating the failure probability of the device, especially in the failure probability calculation part of the formula, the historical failure data and the health index work together to evaluate the failure risk of the device.

[0070] In an implementable manner, the historical failure data can come from the repair records, failure logs and device management system of the device. These data are automatically generated by manual recording of the operation and maintenance personnel or the fault detection system of the device. For each failure, the specific time of occurrence, failure type and repair duration, etc. are recorded. These information can be obtained through the monitoring platform of the device or the maintenance management system, and form a long-term historical data archive.

[0071] Specifically, real-time operation data is dynamic working state data generated by the equipment during the current or recent operation process. These data include the running time of the equipment, the average running time, and whether the equipment has failed during operation. Specifically, it includes: the running time of the equipment, which refers to the running time of the equipment in a certain specific operation cycle. For example, the specific running time of the equipment in one work cycle; the average running time of the equipment, which refers to the average running time of the equipment within a certain time range. It is used to measure the overall working state and regular operation of the equipment; the fault state (fault state value) of the i-th running of the equipment. Each time the equipment runs, it will check whether the equipment has failed. If it fails, the fault state value is 1; if it does not fail, the fault state value is 0. This data helps to assess the health status of the equipment and identify failure modes.

[0072] It is worth noting that in the present application, the main use of real-time operation data is: for calculating the health index, combining running time and fault state to reflect the running condition of the equipment; for failure probability prediction, combining historical failure data to assess the risk of equipment failure in the next time period.

[0073] In an implementable manner, real-time operation data can be collected through various sensors on the equipment, such as temperature sensors, vibration sensors, running time meters, pressure sensors, etc. These sensors can monitor the working state of the equipment in real time and upload data to the equipment management system or cloud platform. Maintenance personnel can view the real-time data of the equipment through these systems, conduct state assessment, and take maintenance measures according to the actual situation.

[0074] In step S12, according to the equipment basic information, the equipment is analyzed by category to obtain the equipment category.

[0075] Preferably, the equipment category includes single-machine equipment, intermittent process equipment, and continuous process equipment.

[0076] It is worth noting that single-machine equipment refers to equipment that runs independently and does not depend on other equipment. This type of equipment usually performs a single work task and does not directly affect the operation of other equipment when a failure occurs. Their distinctive feature is that they can independently complete a specific work task and usually have no coordination requirements with other equipment. Since they work independently, the failure of single-machine equipment will not immediately affect the operation of the entire production line, and the failure will only affect the normal operation of a single device. The maintenance of single-machine equipment is usually based on the running state and failure record of the equipment itself, and is independent of the running state of other equipment.

[0077] It is worth mentioning that the batch process equipment refers to those equipment that work and stop in a certain period. Such equipment is usually not continuously running in the production process, but is started and stopped periodically according to the production needs. For example, in the chemical industry, food processing and other industries, some equipment may only run in a certain production stage, and stop running in other stages. Their remarkable feature is to run in cycles in the production process, after each run, there is a downtime, and then start again. Due to its non-continuous operation, the failure of batch process equipment may not be easily discovered during non-operation period.

[0078] It is worth mentioning that the continuous process equipment refers to those equipment that continuously and uninterruptedly run, usually work continuously for 24 hours, and will not stop due to periodic factors. Such equipment is particularly common in chemical, petroleum, steel and other industries, and its running mode is continuous, usually uninterrupted production tasks. The feature of continuous process equipment is that such equipment almost never stops in the production process, and may run for weeks, months or even longer, until maintenance or repair. The maintenance of continuous process equipment is usually based on real-time data, and requires continuous state monitoring and prediction to discover potential failures as early as possible and take appropriate preventive measures.

[0079] Preferably, the category analysis of the equipment according to the equipment basic information obtains the equipment category, which comprises:

[0080] According to the equipment basic information, the PCA dimension reduction operation is performed to obtain the dimension reduction data;

[0081] According to the dimension reduction data, the equipment classification is performed based on a clustering algorithm to obtain the equipment category.

[0082] Preferably, the PCA dimension reduction operation according to the equipment basic information obtains the dimension reduction data, which comprises:

[0083] According to the equipment basic information, a matrix is constructed to obtain a feature matrix; wherein the feature matrix is composed of the equipment basic information, a row of the feature matrix represents each equipment, and a column of the feature matrix represents the characteristics of the equipment;

[0084] The feature matrix is standardized to obtain a standardized matrix;

[0085] The covariance of the standardized matrix is calculated to obtain a covariance matrix;

[0086] The covariance matrix is decomposed into eigenvalues and eigenvectors to obtain principal components;

[0087] According to the principal components, data filtering is performed to obtain the dimension reduction data.

[0088] Specifically, according to the device basic information, first of all, the device needs to be classified to determine whether it belongs to a single device, an intermittent process device or a continuous process device. This process is to ensure that different management strategies can be taken according to different device types in subsequent fault prediction and health management.

[0089] Specifically, the classification of the device is mainly carried out through PCA dimension reduction operation. It is worth noting that PCA (Principal Component Analysis) dimension reduction is a statistical method for data dimension reduction, which extracts the main features of the device by reducing the data dimension. The goal of this process is to extract the most important features from a large amount of device information, so as to help the classification algorithm better identify and distinguish the device categories.

[0090] First of all, when performing PCA dimension reduction operation, the detailed features of each device need to be extracted from the device basic information. These features constitute a feature matrix, each row of which represents a device and each column represents a feature of the device. The feature matrix contains various data generated by the device during operation, such as device model, running time, load, temperature. These data help the algorithm understand the running status and working characteristics of the device. Next, the feature matrix is standardized. Standardization is to eliminate the scale difference between different features to ensure that each feature is compared in the same range. The standardized matrix is called standardized matrix, and each column of it represents a standardized device feature. The standardization process can effectively avoid some features from dominating the dimension reduction process due to their large or small scales. After obtaining the standardized matrix, the covariance matrix is calculated. The covariance matrix reflects the correlation between features, and the larger the covariance value, the stronger the relationship between features. By calculating the covariance matrix, the algorithm can identify which features have strong linear relationships, and thus help us identify the features that need to be focused on in dimension reduction.

[0091] Specifically, after calculating the covariance matrix, eigenvalue and vector decomposition is performed, which is the core step of PCA dimensionality reduction. By performing eigenvalue decomposition on the covariance matrix, we can obtain a set of principal components. These principal components are the "new coordinate axes" of the data, which represent the most important directions in the original feature space. In the dimensionality reduction process, the principal components corresponding to the preset number of largest eigenvalues are selected, which contain most of the data information, while ignoring the secondary features that contribute less to device classification. It is worth noting that selecting the principal components corresponding to the preset number of largest eigenvalues is the data screening process. The final dimensionality reduction data is obtained, which contains the most important feature information of the device and can help the subsequent clustering algorithm to classify the device. The clustering algorithm will use these dimensionality reduced data to divide the device into different categories (such as single machine device, intermittent process device and continuous process device), and provide a basis for subsequent fault prediction, health index calculation and other tasks.

[0092] In step S13, according to the historical fault data and the device category, the device is predicted for failure cycle based on a pre-trained support vector regression model, to obtain a device failure cycle

[0093] Preferably, the training process of the support vector regression model comprises:

[0094] Obtain the first failure cycle and the second failure cycle;

[0095] Standardize the first failure cycle and the second failure cycle to obtain first preprocessed data and second preprocessed data;

[0096] Based on the first preprocessed data and the second preprocessed data as input data, a support vector regression model is constructed based on a kernel function, the support vector regression model is trained to obtain a trained support vector regression model.

[0097] Specifically, the training process of the support vector regression (SVR) model aims to predict the future failure cycle of the device using historical fault data. Specifically, this process includes the following key steps.

[0098] First, the first failure cycle and the second failure cycle need to be obtained. The first failure cycle refers to the time interval when the device first fails, and the second failure cycle refers to the time interval when the device fails for the second time. Through these two time periods, we can understand the early patterns and trends of device failure. These failure cycle data are the basic input of the support vector regression model, helping the model to learn the rules of device failure.

[0099] Next, the obtained first failure cycle and second failure cycle need to be standardized. This is to eliminate the scale difference between different failure cycle data, so that the data can be compared in the same range. The standardization process will convert the original data into standard data with a mean of zero and a variance of one, which can ensure that each data feature contributes equally to the training of the model, and avoid the unnecessary influence of some large-scale data features on the model training. After standardization, the first preprocessed data and the second preprocessed data are obtained respectively.

[0100] Then, based on the two preprocessed data (i.e. the first and second failure cycle after standardization), as input data, a support vector regression model is constructed using a kernel function. The kernel function is the core part of the SVR model, which maps the original data to a high-dimensional space, so that an optimal hyperplane can be found in this space to make regression prediction. Selecting different kernel functions (such as linear kernel, radial basis kernel) will affect the prediction ability and adaptability of the model. Through the kernel function, SVR can better capture the complex patterns of equipment failure cycle in high-dimensional space.

[0101] Finally, the support vector regression model is trained to learn how to predict future equipment failure cycles based on input failure cycle data. The training process includes minimizing error and optimizing model parameters to ensure that the model can accurately fit the input data and has good generalization ability. After training, an optimized support vector regression model is obtained, which can predict the failure cycle based on the historical failure data of the equipment.

[0102] In step S14, the health index of the equipment is calculated according to the real-time running data and the equipment category, and the health index of the equipment is obtained.

[0103] Preferably, the health index of the equipment includes single machine health index, intermittent process equipment health index and continuous process equipment health index.

[0104] Preferably, the health index of the equipment is calculated according to the real-time running data and the equipment category, and the health index of the equipment is obtained.

[0105] The single machine health index is calculated by the following formula:

[0106] ;

[0107] The intermittent process equipment health index is calculated by the following formula:

[0108] ;

[0109] The continuous process equipment health index is calculated by the following formula:

[0110] ;

[0111] wherein, is the single machine equipment health index; is the intermittent process equipment health index; is the continuous process equipment health index; is the running time of the i-th equipment; is the average running time of the equipment; is the failure status of the i-th running of the equipment, taking the value of 1 if a failure occurs, and 0 if no failure occurs; is the total running number of the equipment.

[0112] Specifically, in the formula for calculating the single machine equipment health index, the running time represents the duration of each running of the equipment, and the average running time is the average running duration of the equipment within a certain period of time. The calculates the deviation of each equipment running time relative to the average time. If the running time of the equipment deviates greatly from the average running time, it indicates that the health condition of the equipment may be poor. By squaring these deviations, the influence of larger deviations on the health index can be emphasized. The failure status indicates whether a failure occurs during each running of the equipment. If a failure occurs, the failure status is 1; if there is no failure, the failure status is 0. The number of failures has a direct negative impact on the health index, and frequent failures will significantly reduce the health index of the equipment. The total running number is the number of equipment runs, i.e., the number of all running cycles of the equipment within a certain period of time. By accumulating the above and dividing by the total running number, the health index of the equipment is calculated. The closer the value of the health index is to 1, the better the health condition of the equipment; if the value is low, it indicates that the health condition of the equipment is poor.

[0113] Specifically, in the formula for calculating the intermittent process equipment health index, the meanings of the running time and the average running time are the same as those of the single machine equipment, but here the (i.e., the absolute difference of the running time) is used. Since the running of the intermittent process equipment is periodically changing, using the absolute difference can more accurately reflect the volatility of each running of the equipment. If the running time of the equipment fluctuates greatly each time, it will affect the calculation of the health index. In addition, the meanings of the failure status and the total running number are the same as those of the single machine equipment. By summing up the absolute differences of all equipment running times and the failure status, and dividing by the product of the total running number and the average running time, the health index of the equipment is finally obtained. The closer the value of the health index is to 1, the better the health condition of the equipment.

[0114] Specifically, in the formula for calculating the continuous process equipment health index, the calculation of the health index introduces an exponential function, which makes the health index more sensitive to equipment failures and running time deviations. Since the continuous process equipment is always running, its failures and running deviations have a greater impact on health, so by exponential processing, the impact of failures and deviations can be more significant. In this formula, the running time represents the length of each run, and the average running time represents the average running length of the equipment. The deviation of the equipment running time from the average running time is calculated, which is similar to the calculation method of single-machine equipment, but under the action of the exponential function, the impact of the deviation on the health index is further amplified. The failure state represents whether the equipment has failed, and the total running time is the total running period of the equipment. All deviation and failure state values are subjected to exponential operation to obtain the final health index. The closer the value of the health index is to 1, the better the health of the equipment, and higher failure frequency or running time fluctuations will cause a significant drop in the health index.

[0115] In step S15, according to the equipment category, the equipment failure period and the equipment health index, a failure prediction is performed to obtain a predicted failure probability, and the predicted failure probability is sent to the operation and maintenance end to enable the operation and maintenance end to perform equipment maintenance according to the predicted failure probability.

[0116] Preferably, the failure prediction according to the equipment category, the equipment failure period and the equipment health index to obtain a predicted failure probability comprises:

[0117] The single-machine equipment failure probability is calculated by the following formula:

[0118] ;

[0119] The intermittent equipment failure probability is calculated by the following formula:

[0120] ;

[0121] The continuous equipment failure probability is calculated by the following formula:

[0122] ;

[0123] In the formula, is the single-machine equipment failure probability; is the intermittent equipment failure probability; is the continuous equipment failure probability; is the last equipment failure period; is the average value of the equipment historical failure period; is the single-machine equipment health index; is the intermittent process equipment health index; is a continuous process equipment health index; is an ith equipment failure cycle; is a total number of historical failure cycles; is a single machine equipment weight; is an intermittent process equipment weight; is a continuous process equipment weight; is a cumulative failure influence coefficient.

[0124] Specifically, in the failure prediction, the failure probabilities of single machine equipment, intermittent process equipment and continuous process equipment are calculated respectively according to the category, failure cycle and health index of the equipment. For single machine equipment, the calculation of the failure probability comprehensively considers the deviation of the last failure cycle from the average failure cycle, the health index of the equipment and the weight of the single machine equipment. When the difference between the last failure cycle and the average cycle is large and the health index of the equipment is low, the failure probability will significantly increase, reflecting that the equipment is in potential failure risk. For intermittent process equipment, the calculation of the failure probability mainly depends on the sum of the deviation of each failure cycle from the average failure cycle, as well as the health index of the equipment and the weight of the intermittent equipment. By accumulating the deviation values of the historical failure cycles and combining the adjustment of the health index on the result, the failure probability of the equipment can be accurately predicted. The failure probability of the intermittent equipment can reflect the influence of the periodic fluctuations in the operation process of the equipment on the failure risk. The failure probability of the continuous process equipment is calculated by accumulating the total sum of the failure cycles, and a cumulative failure influence coefficient is introduced to evaluate the influence of the historical failure cycles on the future failure probability. This calculation method particularly emphasizes the long-term operation characteristics of the equipment, and combined with the health index and the category weight of the equipment, it can more sensitively capture the influence of the decline of the health state of the equipment or the abnormality of the historical failure cycles on the failure risk.

[0125] Preferably, the predicted failure probability is sent to an operation and maintenance end, so that the operation and maintenance end performs equipment maintenance according to the predicted failure probability, including:

[0126] When the predicted failure probability is greater than and equal to a first preset threshold value, it is determined that the failure risk level is high risk, and an emergency maintenance operation is performed;

[0127] When the predicted failure probability is less than the first preset threshold value and greater than a second preset threshold value, it is determined that the failure risk level is medium risk, and a preventive maintenance operation is performed;

[0128] When the predicted failure probability is less than and equal to the second preset threshold value, it is determined that the failure risk level is low risk, and a regular maintenance operation is performed.

[0129] Specifically, in the equipment maintenance process, information is sent to the operation and maintenance end according to the predicted failure probability, helping the operation and maintenance team to take appropriate maintenance measures in a timely manner. The classification standard of failure probability is divided into three categories: high risk, medium risk and low risk, each category corresponds to different maintenance strategies to ensure the normal operation of the equipment and reduce the risk of sudden failure.

[0130] When the predicted failure probability is greater than or equal to the first preset threshold, the equipment is determined to be in a high-risk state. This means that the equipment has a very high possibility of failure, which may have a serious impact on the normal operation of the production line or system. In this case, immediate emergency maintenance operations, including shutdown inspection, replacement of parts or comprehensive overhaul, are needed to eliminate potential risks and restore the normal state of the equipment.

[0131] When the predicted failure probability is lower than the first preset threshold but higher than the second preset threshold, the equipment is determined to be in a medium-risk state. In this case, although the equipment has a certain risk of failure, it does not reach an emergency state. In order to prevent the risk from further expanding, preventive maintenance measures are recommended. This can include replacing wear-and-tear parts, adjusting operating parameters or conducting detailed inspections, thereby delaying the occurrence of equipment failure and improving its operating stability.

[0132] When the predicted failure probability is less than or equal to the second preset threshold, the equipment is determined to be in a low-risk state. This indicates that the equipment is running relatively stably and has a low possibility of failure. For this case, regular maintenance can be arranged to ensure the normal operation of the equipment while avoiding resource waste caused by excessive maintenance. Regular maintenance usually includes routine inspection, cleaning and basic maintenance.

[0133] In summary, the present application provides a management method for the running state of equipment in an industrial park, which includes: obtaining equipment basic information, historical failure data and real-time running data of the equipment; performing category analysis on the equipment according to the equipment basic information to obtain the equipment category; wherein the equipment category includes single-machine equipment, intermittent process equipment and continuous process equipment; performing failure cycle prediction on the equipment based on a pre-trained support vector regression model according to the historical failure data and the equipment category to obtain the equipment failure cycle; performing health index calculation on the equipment according to the real-time running data and the equipment category to obtain the equipment health index; wherein the equipment health index includes single-machine equipment health index, intermittent process equipment health index and continuous process equipment health index; performing failure prediction according to the equipment category, the equipment failure cycle and the equipment health index to obtain the predicted failure probability, and sending the predicted failure probability to the operation and maintenance end to enable the operation and maintenance end to perform equipment maintenance according to the predicted failure probability.

[0134] In the present application, the method can analyze the basic information, historical failure data and real-time operation data of the equipment, and further infer the health state and failure cycle of the equipment. Specifically, the method comprises classifying the equipment, predicting the failure cycle by using a support vector regression model, and calculating the health index of the equipment according to the real-time operation data. In addition, the failure prediction is combined with the health index and failure cycle of the equipment to obtain the predicted failure probability, and the prediction result is finally provided to the operation and maintenance end for corresponding equipment maintenance operation. The method can improve the equipment management efficiency of the industrial park and realize effective analysis and management of the equipment state of the industrial park.

[0135] Referring to Figure 2 The second embodiment of the present application provides a management system for the running state of equipment in an industrial park, comprising:

[0136] A data acquisition module is configured to acquire the basic information, historical failure data and real-time operation data of the equipment.

[0137] A classifying module is configured to classify the equipment according to the basic information of the equipment to obtain the equipment category, wherein the equipment category comprises single-machine equipment, intermittent process equipment and continuous process equipment.

[0138] A cycle prediction module is configured to predict the failure cycle of the equipment based on a pre-trained support vector regression model according to the historical failure data and the equipment category.

[0139] A health calculation module is configured to calculate the health index of the equipment according to the real-time operation data and the equipment category to obtain the equipment health index, wherein the equipment health index comprises the health index of single-machine equipment, the health index of intermittent process equipment and the health index of continuous process equipment.

[0140] A failure prediction module is configured to predict the failure according to the equipment category, the equipment failure cycle and the equipment health index to obtain the predicted failure probability, and send the predicted failure probability to the operation and maintenance end to enable the operation and maintenance end to perform equipment maintenance according to the predicted failure probability.

[0141] It should be noted that the management system for the running state of equipment in an industrial park provided by the embodiments of the present application is used to execute all process steps of the management method for the running state of equipment in an industrial park of the above-mentioned embodiments, and the working principles and beneficial effects of the two are one-to-one corresponding, thus not being repeated.

[0142] The embodiment of the present application further provides an electronic device. The electronic device comprises a processor, a memory, and a computer program stored in the memory and executable on the processor, for example, the management method of the industrial park equipment running state program. The processor implements the steps in each of the above-mentioned industrial park equipment running state management method embodiments when executing the computer program, for example Figure 1 The processor implements the functions of each module / unit in each of the above-mentioned apparatus embodiments when executing the computer program, for example, the cycle prediction module.

[0143] For example, the computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present application. The one or more modules / units can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program in the electronic device.

[0144] The electronic device can be a desktop computer, a notebook computer, a palm computer, a smart tablet and the like. The electronic device can include, but is not limited to, a processor, a memory. Those skilled in the art can understand that the above components are only examples of the electronic device and do not constitute a limitation on the electronic device, and can include more or fewer components than the above, or combine certain components, or different components, for example, the electronic device can also include an input / output device, a network access device, a bus and the like.

[0145] The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor and the like. The processor is the control center of the electronic device, and connects all parts of the electronic device through various interfaces and lines.

[0146] The memory can be used to store the computer program and / or modules, and the processor realizes various functions of the electronic device by running or executing the computer program and / or modules stored in the memory, and calling data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required by a function (such as a sound playing function, an image playing function, etc.), and the like; and the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), and the like. In addition, the memory can include a high-speed random access memory, and can also include a nonvolatile memory, for example, a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state memory devices.

[0147] The modules / units integrated in the electronic device, if realized in the form of software function units and sold or used as independent products, can be stored in a computer readable storage medium. Based on this understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. The computer program can realize the steps of the above-mentioned various method embodiments when executed by a processor. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or some intermediate forms, etc. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the contents included in the computer readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.

[0148] It should be noted that the apparatus embodiments described above are merely illustrative, and the units described as separate units can or can not be physically separate, and the units displayed as units can or can not be physical units, i.e. can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. In addition, the connection relationship between the modules in the apparatus embodiment provided by the present application indicates that there is a communication connection between them, which can be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement it without creative labor.

[0149] The above specific embodiments further illustrate the purpose, technical scheme and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the protection scope of the present application. It is particularly pointed out that any modification, equivalent replacement, improvement, etc. made by those skilled in the art within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method of managing the operating state of industrial park equipment, characterized by, The method comprises the following steps: obtaining device basic information, historical failure data and real-time running data of the device; performing category analysis on the device according to the device basic information to obtain a device category; wherein the device category comprises single-machine device, intermittent process device and continuous process device; performing failure cycle prediction on the device based on a pre-trained support vector regression model according to the historical failure data and the device category to obtain a device failure cycle; performing health index calculation on the device according to the real-time running data and the device category to obtain a device health index; wherein the device health index comprises single-machine device health index, intermittent process device health index and continuous process device health index; performing failure prediction according to the device category, the device failure cycle and the device health index to obtain a predicted failure probability, and sending the predicted failure probability to an operation and maintenance end to enable the operation and maintenance end to perform device maintenance according to the predicted failure probability; wherein the training process of the support vector regression model comprises: obtaining a first failure cycle and a second failure cycle; standardizing the first failure cycle and the second failure cycle to obtain first pre-processing data and second pre-processing data; training the support vector regression model based on the first pre-processing data and the second pre-processing data as input data and based on a kernel function to obtain a trained support vector regression model; wherein the health index calculation on the device according to the real-time running data and the device category to obtain a device health index comprises: calculating the single-machine device health index by the following formula: ； calculating the intermittent process device health index by the following formula: ; calculating the continuous process device health index by the following formula: ； wherein, is the single machine equipment health index; is the intermittent process equipment health index; is the continuous process equipment health index; is the running time of the ith equipment; is the average time of equipment running; is the failure state of the ith running equipment, taking the value of 1 if a failure occurs, and taking the value of 0 if no failure occurs; is the total number of equipment running; wherein the failure prediction according to the device category, the device failure cycle and the device health index to obtain a predicted failure probability comprises: calculating the single-machine device failure probability by the following formula: ; calculating the intermittent device failure probability by the following formula: ； calculating the continuous device failure probability by the following formula: ; wherein, is the single machine equipment failure probability; is the intermittent equipment failure probability; is the continuous equipment failure probability; is the last equipment failure cycle; is the average of the equipment historical failure cycles; is the single machine equipment health index; is the intermittent process equipment health index; is the continuous process equipment health index; is the ith equipment failure cycle; is the total number of historical failure cycles; is the single machine equipment weight; is the intermittent process equipment weight; is the continuous process equipment weight; is the cumulative failure impact factor.

2. The management method of the industrial park facility operation state according to claim 1, characterized by, the category analysis on the device according to the device basic information to obtain a device category comprises: performing PCA dimension reduction operation according to the device basic information to obtain dimension reduction data; performing device classification based on a clustering algorithm according to the dimension reduction data to obtain a device category.

3. The management method of the operational status of the industrial park facility according to claim 2, characterized by, the PCA dimension reduction operation according to the device basic information to obtain dimension reduction data comprises: performing matrix construction according to the device basic information to obtain a feature matrix; wherein the feature matrix is composed of the device basic information, a row of the feature matrix represents each device, and a column of the feature matrix represents a feature of the device; standardizing the feature matrix to obtain a standardized matrix; calculating the covariance of the standardized matrix to obtain a covariance matrix; performing eigenvalue and vector decomposition on the covariance matrix to obtain principal components; performing data screening according to the principal components to obtain dimension reduction data.

4. An industrial park facility operation state management system characterized by comprising: The method for managing the running state of the device in the industrial park according to any one of claims 1 to 3 comprises: The data acquisition module is configured to acquire device basic information, historical fault data, and real-time operation data of the device. The category analysis module is configured to perform category analysis on the device according to the device basic information, and obtain a device category, wherein the device category includes a single-machine device, an intermittent process device, and a continuous process device. The cycle prediction module is configured to perform fault cycle prediction on the device based on a pre-trained support vector regression model according to the historical fault data and the device category, and obtain a device fault cycle. The health calculation module is configured to perform health index calculation on the device according to the real-time operation data and the device category, and obtain a device health index, wherein the device health index includes a single-machine device health index, an intermittent process device health index, and a continuous process device health index. The fault prediction module is configured to perform fault prediction according to the device category, the device fault cycle, and the device health index, obtain a predicted fault probability, and send the predicted fault probability to an operation and maintenance terminal, so that the operation and maintenance terminal performs device maintenance according to the predicted fault probability.

5. An electronic device, comprising: The computer readable storage medium comprises a stored computer program, wherein the computer program controls a device in which the computer readable storage medium is located to perform the management method of the industrial park device operation state according to any one of claims 1 to 3 when the computer program is running.

6. A computer-readable storage medium, characterized in that, The computer readable storage medium comprises a stored computer program, wherein the computer program controls a device in which the computer readable storage medium is located to perform the management method of the industrial park device operation state according to any one of claims 1 to 3 when the computer program is running.

Images