Data acquisition and monitoring control method based on intelligent industrial park

Abstract

The invention relates to the technical field of industrial park intelligent monitoring, and particularly discloses a data acquisition and monitoring control method based on an intelligent industrial park, which comprises the following steps: acquiring key state data of equipment in real time, including operation time, operation temperature and operation power; an advanced signal processing algorithm is used for calculating an operation temperature abnormal index and an operation power abnormal index, the system can comprehensively evaluate the state of equipment and accurately judge whether slight or serious abnormal conditions exist or not based on detailed data analysis results, and the system can not only give out instant early warning notifications but also give out alarm information according to the serious abnormal conditions. According to the method, a machine learning model is adopted to predict the future operation state of the equipment, the equipment state abnormal coefficient in each future monitoring period is predicted based on the equipment state data in the current period, and classification and dynamic regulation and control of the equipment state are carried out according to a preset threshold value.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent monitoring in industrial parks, and particularly to a data acquisition and supervisory control method based on a smart industrial park. Background Art

[0002] With the rapid development of industrial automation and intelligent manufacturing, the condition monitoring and health management of equipment have become increasingly important. Modern industrial parks are usually equipped with a large number of complex mechanical devices, which may experience various failures or performance degradation during long-term operation, thus affecting production efficiency and safety. To ensure the continuous and stable operation of the equipment, it is particularly crucial to monitor the key state parameters of the equipment in real time (such as running time, running temperature, and running power). Traditional monitoring methods mainly rely on manual inspections and regular maintenance, which are not only time-consuming and laborious but also difficult to detect potential problems in a timely manner, resulting in a high equipment failure rate and frequent unplanned outages. Therefore, there is an urgent need for a more intelligent and automated solution that can collect and analyze equipment status data in real time, predict potential failures, and take corresponding preventive measures.

[0003] The existing technologies have the following deficiencies:

[0004] The deficiencies of existing equipment monitoring technologies in dynamic early warning and control specifically include problems such as the lack of comprehensive analysis of multi-dimensional data, the prone occurrence of false alarms and missed alarms in simple threshold alarm mechanisms, and the lack of automated dynamic regulation measures. By collecting and analyzing data such as equipment running time, temperature, and power in real time, using advanced algorithms to evaluate the health status of the equipment, and automatically triggering regulation measures to improve the safety and reliability of equipment operation and reduce unplanned downtime. Summary of the Invention

[0005] The purpose of the present invention is to provide a data acquisition and supervisory control method based on a smart industrial park to solve the problems in the above background.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A data acquisition and supervisory control method based on a smart industrial park includes the following steps:

[0008] S1: During the monitoring period, collect the equipment status data in the park in real time, including: running time, running temperature, and running power;

[0009] S2: Analyze the update frequency of the equipment status data, and judge whether the quality of the equipment status data is qualified according to the consistency ratio and missing value ratio of the equipment status data;

[0010] S3: According to the judgment result, extract the qualified device status data, and evaluate the stability of the device status based on the running time, the degree of change in running temperature, and the stability of running power of the device, and judge whether the device status is abnormal;

[0011] S4: According to the judgment result, divide the device running status into normal, slightly abnormal, and severely abnormal;

[0012] S5: Conduct early warning processing on the devices in severely abnormal status, further monitor the devices in slightly abnormal status, predict whether the future running status of the devices is abnormal, and perform dynamic regulation of the devices based on the prediction results.

[0013] As a further solution of the present invention: The specific method for judging whether the quality of the device status data is qualified includes:

[0014] Obtain the update frequency of the device status data, calculate the quality coefficient of the device status data according to the consistency ratio and the missing value ratio of the device status data, judge whether the quality coefficient of the device status data is greater than or equal to a preset threshold. If so, it is recorded as qualified; if not, it is recorded as unqualified.

[0015] As a further solution of the present invention: The process of obtaining the quality coefficient is as follows:

[0016] Obtain the time series of the device status data, and use dynamic time warping to calculate the distance between the actual device status data and the ideal data;

[0017] The actual device status data is the device status data collected in real time, and the ideal data is the device status data of the preset normal standard;

[0018] Calculate the update frequency normalization function, the consistency ratio function, and the penalty function, comprehensively process the update frequency normalization function, the consistency ratio function, and the penalty function, and calculate the quality coefficient through the comprehensive calculation expression.

[0019] As a further solution of the present invention: The specific method for evaluating the stability of the device status includes:

[0020] Collect the running time, running temperature, and running power of the device in real time, analyze the running temperature of the device, calculate the running temperature anomaly index according to the degree of change in the running temperature of the device, analyze the running power of the device, calculate the running power anomaly index according to the fluctuation range of the running power, comprehensively calculate and process the running time, the running temperature anomaly index, and the running power anomaly index of the device to obtain the device status anomaly coefficient, and judge the stability of the device status according to the device status anomaly coefficient.

[0021] As a further solution of the present invention: according to the evaluation results, the operating state of the device is divided into normal, slightly abnormal and severely abnormal, specifically including:

[0022] Compare the device state anomaly coefficient of the device with a preset first threshold. If the device state anomaly coefficient is greater than or equal to the preset first threshold, it is recorded as severely abnormal. If the device state anomaly coefficient is less than the preset first threshold and the device state anomaly coefficient is greater than the preset second threshold, it is recorded as slightly abnormal. If the device state anomaly coefficient is less than or equal to the preset second threshold, it is recorded as normal.

[0023] As a further solution of the present invention: the process of obtaining the operating temperature anomaly index is as follows:

[0024] Collect the temperature data of the device in real time according to the time series, perform Haar wavelet transform on the temperature data to obtain a series of approximation coefficients and detail coefficients;

[0025] Calculate the ratio of the energy of each detail coefficient to the total energy to obtain the energy ratio of each detail coefficient;

[0026] Obtain the mean value of the energy ratios of all decomposition layers of the Haar wavelet transform to obtain the operating temperature anomaly index.

[0027] As a further solution of the present invention: the process of obtaining the operating power anomaly index is as follows:

[0028] During the monitoring period, collect the operating power data of the device in real time according to the time series, perform fast Fourier transform on the operating power data to obtain a set of spectral coefficients in complex form;

[0029] Based on the transformed operating power data of the device, calculate the energy density of each frequency component. The calculation expression is:

[0030] E a =|F a | 2 ;

[0031] Wherein, E a is the energy density of the a-th frequency component, which represents a measure of the signal fluctuation intensity at the corresponding frequency, and F a represents the complex spectral coefficient of the a-th frequency component, and a represents the number of frequency components after the fast Fourier transform;

[0032] Calculate the ratio of each frequency component to the total energy to obtain the spectral energy ratio;

[0033] Calculate the average value of the energy ratios of all frequency components. Calculate the difference between each spectrum energy ratio and the average value of the energy ratios of all frequency components, and take the absolute value to obtain the difference of the spectrum energy ratios. Sum up all the differences to obtain the operating power anomaly index.

[0034] As a further solution of the present invention: further monitor the slightly abnormal device, which specifically includes:

[0035] Obtain the slightly abnormal device, extract the running time, running temperature anomaly index, and running power anomaly index of the device. Construct a comprehensive feature vector from the running time, running temperature anomaly index, and running power anomaly index as the input of the machine learning model. The output is the device status anomaly coefficient. According to the comparison between the device status anomaly coefficient and the preset threshold, predict whether the future running status of the device is abnormal.

[0036] As a further solution of the present invention: the construction process of the machine learning model is as follows:

[0037] Obtain the running time, running temperature anomaly index, and running power anomaly index. Construct a comprehensive feature vector from the running time, running temperature anomaly index, and running power anomaly index as the input of the machine learning model. Train the model with the goal of minimizing the error between the predicted device status anomaly coefficient and the actual device status anomaly coefficient. According to the training results, the model will output the device status anomaly coefficient in each future monitoring period. According to the comparison between the device status anomaly coefficient and the preset first threshold and the preset second threshold respectively, identify the device status in each monitoring period, including: normal, slightly abnormal, and severely abnormal.

[0038] As a further solution of the present invention: the device dynamic regulation based on the prediction result specifically includes:

[0039] For the device whose future status is detected to be severely abnormal, adjust the working load of the device in real time, optimize the operating parameters, start the standby system, and perform preventive maintenance tasks.

[0040] The beneficial effects of the present invention:

[0041] (1) The present invention collects key status data such as the running time, running temperature, and running power of devices in the park in real time, and uses advanced algorithms (such as Haar wavelet transform and fast Fourier transform) to deeply analyze these data, calculate the running temperature anomaly index and the running power anomaly index, so as to comprehensively evaluate the health status of the devices. Specifically, the Haar wavelet transform can accurately capture the change trend and fluctuation intensity of the running temperature, while the fast Fourier transform can convert the time series of the running power into a frequency-domain representation, reveal the energy distribution characteristics of the signal, and then identify potential abnormal fluctuations. Based on these detailed analysis results, the system can not only accurately judge whether there are minor or serious abnormal conditions in the devices, but also classify the device status according to the preset first threshold and second threshold. Especially for serious abnormal conditions, the system will not only issue an immediate warning notice, but also automatically trigger a series of dynamic regulation measures, including real-time adjustment of the device's workload, optimization of key operating parameters, and activation of standby systems. For example, in the power supply system, if a transformer malfunctions, the system will automatically switch to a standby transformer to ensure uninterrupted power supply. This forward-looking monitoring and regulation mechanism significantly improves the safety and reliability of device operation, reduces the unplanned downtime and economic losses caused by device failures. At the same time, the detailed anomaly reports generated by the system provide the specific time of the anomaly occurrence, the degree of parameter deviation from the normal range, and recommended corrective measures, helping technicians quickly locate problems and take effective maintenance measures, further enhancing the intelligent level and production efficiency of device management. This method is particularly applicable to the fields of industrial automation and intelligent manufacturing, effectively ensuring the continuity and stability of production and minimizing operation risks.

[0042] (2) The invention uses a machine learning model to predict the future operating status of devices. By learning historical data, the model can predict the device status anomaly coefficient in each future monitoring period based on the device status data in the current period (including running time, running temperature anomaly index, and running power anomaly index), and compare it with the preset threshold to identify possible abnormal conditions in advance. This method not only improves the intelligent level of device management, but also significantly enhances the fault prediction and prevention ability. By taking preventive maintenance measures in a timely manner, the occurrence of potential faults is avoided, thus greatly reducing the device downtime and improving the overall production efficiency. In addition, the detailed reports generated by the system provide the specific type, cause of the anomaly, and recommended maintenance measures, further guiding technicians to perform precise repair and maintenance to ensure the long-term stable operation of the devices. This data-driven decision support system provides an efficient and reliable device health management solution for the fields of industrial automation and intelligent manufacturing. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The present invention will be further described below with reference to the accompanying drawings.

[0044] Figure 1 It is a flowchart of the specific steps of a data collection and monitoring control method based on a smart industrial park of the present invention. Specific embodiments

[0045] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0046] Please refer to Figure 1 As shown, the present invention is a data collection and monitoring control method based on a smart industrial park, including the following steps:

[0047] S1: During the monitoring period, the device status data in the park is collected in real time, including: running time, running temperature, and running power;

[0048] S2: Analyze the update frequency of the device status data, and judge whether the quality of the device status data is qualified according to the consistency ratio and missing value ratio of the device status data;

[0049] S3: According to the judgment result, extract the qualified device status data, and evaluate the stability of the device status according to the running time, the degree of change in running temperature, and the running power stability of the device, and judge whether the device status is abnormal;

[0050] S4: According to the judgment result, divide the device operation status into normal, slightly abnormal, and severely abnormal;

[0051] S5: Perform early warning processing on the devices in the severely abnormal state, further monitor the slightly abnormal devices, predict whether the future operation status of the devices is abnormal, and perform dynamic regulation of the devices based on the prediction results.

[0052] In S1, during the monitoring period, the device status data in the park is collected in real time, including: running time, running temperature, and running power, specifically including:

[0053] The running time is recorded by the built-in timing module; the running temperature is collected by the thermocouple sensor, and these devices can accurately measure the temperature change on the surface of the device; the running power is monitored in real time by the power analyzer installed on the device power loop to ensure the acquisition of multi-dimensional data such as instantaneous power and cumulative energy consumption. These devices are all connected to the Internet of Things (IoT) network in the park through wired or wireless methods to form a comprehensively covered perception system.

[0054] To achieve efficient data collection, all sensors and monitoring devices are equipped with communication modules (such as 5G) to transmit the collected data to the central data processing platform in real time.

[0055] To ensure the accuracy and reliability of the collected data, all sensors and monitoring devices are strictly calibrated and tested before deployment to eliminate possible systematic errors.

[0056] In S2, analyze the update frequency of the device status data. According to the consistency ratio and missing value ratio of the device status data, judge whether the quality of the device status data is qualified, specifically including:

[0057] Obtain the update frequency of the device status data. The update frequency represents the frequency of collecting the device status data. Calculate the consistency ratio of the device status data, which represents the degree of consistency of the device status data at different time points and is a continuous variable between 0 and 1. A high consistency ratio indicates that the data remains stable throughout the time period. Calculate the missing value ratio of the device status data, which represents the proportion of missing values in the total device status data and is also a continuous variable between 0 and 1. A lower missing value ratio means the data is more complete.

[0058] Obtain the time series of the device status data and use dynamic time warping to calculate the distance between the actual device status data and the ideal data;

[0059] The actual device status data is the device status data collected in real time, and the ideal data is the device status data of the preset normal standard;

[0060] For the update frequency F, calculate the update frequency normalization function, and the calculation expression is: where f(F) represents the update frequency normalization function and F represents the update frequency;

[0061] For the consistency ratio C, calculate the consistency ratio function, and the calculation expression is: g(C)=C; where g(C) represents the consistency ratio function and C represents the consistency ratio;

[0062] For the missing value ratio M, calculate the penalty function, and the calculation expression is: h(M)=1 - M; where h(M) represents the penalty function and M represents the missing value ratio;

[0063] Comprehensively process the update frequency normalization function, the consistency ratio function and the penalty function, and calculate the quality coefficient through the comprehensive calculation expression. The comprehensive calculation expression is:

[0064]

[0065] Wherein, D(S, I) represents the dynamic time warping distance between the actual data sequence and the ideal sequence, and g(C), Q represent the quality coefficients.

[0066] It should be noted that: Using the dynamic time warping technology in combination with other key indicators to evaluate the quality of equipment status data can more comprehensively consider the time series characteristics and integrity of the data, so as to obtain a more accurate quality coefficient. This method is particularly suitable for application scenarios that require precise monitoring and evaluation of equipment status data.

[0067] In S3, according to the judgment result, extract the qualified equipment status data, and evaluate the stability of the equipment status according to the running time, the degree of change in running temperature, and the stability of running power of the equipment, and judge whether the equipment status is abnormal. Specifically, it includes:

[0068] Real-time collect the running time, running temperature, and running power of the equipment, analyze the running temperature of the equipment, calculate the running temperature anomaly index according to the degree of change in the running temperature of the equipment, analyze the running power of the equipment, calculate the running power anomaly index according to the fluctuation range of the running power, and comprehensively calculate and process the running time, running temperature anomaly index, and running power anomaly index of the equipment to obtain the equipment status anomaly coefficient;

[0069] Compare the equipment status anomaly coefficient of the equipment with a preset first threshold. If the equipment status anomaly coefficient is greater than or equal to the preset first threshold, it is recorded as a serious anomaly. If the equipment status anomaly coefficient is less than the preset first threshold and the equipment status anomaly coefficient is greater than the preset second threshold, it is recorded as a minor anomaly. If the equipment status anomaly coefficient is less than or equal to the preset second threshold, it is recorded as normal.

[0070] The process of obtaining the running temperature anomaly index is as follows:

[0071] Collect the temperature data of the equipment in real time according to the time series, perform Haar wavelet transform on the temperature data to obtain a series of approximation coefficients and detail coefficients;

[0072] Calculate the energy ratio of each detail coefficient to describe the degree of temperature change. The energy ratio is the ratio of the fluctuation intensity of the temperature data at the current scale to the total energy. The calculation expression of the energy ratio is:

[0073]

[0074] In the formula, E ratio,i represents the energy ratio of the detail coefficient of the i-th layer, n i represents the number of detail coefficients of the i-th layer, D i[j] is the value of the j-th detail coefficient in the i-th layer, where i represents the number of layers of the Haar wavelet transform, j represents the detail coefficient, L represents the total number of decomposition layers of the Haar wavelet transform, and k represents the current decomposition layer of the Haar wavelet transform being processed;

[0075] Obtain the mean value of the energy ratios of all decomposition layers of the Haar wavelet transform to get the operating temperature anomaly index.

[0076] It should be noted that: The anomaly index of the equipment operating temperature is calculated by using the Haar wavelet transform in combination with the energy ratio of the detail coefficients. By calculating the mean value of the overall energy ratio, the anomaly situation during the entire monitoring period of the equipment operation is shown. It is applicable to application scenarios that require precise monitoring of the equipment health status, such as the fields of industrial automation and intelligent manufacturing. It can more accurately identify potential equipment failures or operational anomalies, thereby improving the reliability and security of the system.

[0077] The process of obtaining the operating power anomaly index is as follows:

[0078] During the monitoring period, in accordance with the time series, the operating power data of the equipment are collected in real time. The real-time operating power data of the equipment are subjected to a fast Fourier transform to obtain a frequency-domain representation, including: A set of complex spectral coefficients is obtained by performing a fast Fourier transform on the time-series operating power data. The fast Fourier transform can convert a time-domain signal into a frequency-domain signal, facilitating the analysis of the energy distribution of the signal.

[0079] Based on the transformed operating power data of the equipment, calculate the energy density of each frequency component. The calculation expression is:

[0080] E a =|F a | 2 ;

[0081] where, E a is the energy density of the a-th frequency component, representing a measure of the signal fluctuation intensity at the corresponding frequency, F a represents the complex spectral coefficient of the a-th frequency component, and a represents the number of frequency components after the fast Fourier transform;

[0082] Calculate the ratio of each frequency component to the total energy to obtain the spectral energy ratio;

[0083] Calculate the average value of the energy ratios of all frequency components, calculate the difference between each spectral energy ratio and the average value of the energy ratios of all frequency components, and take the absolute value to obtain the difference of the spectral energy ratio. Sum up all the differences to obtain the operating power anomaly index.

[0084] It should be noted that by converting the power time series to the frequency domain through fast Fourier transform and calculating the energy density and energy ratio of each frequency component, abnormal conditions in power fluctuations can be effectively identified.

[0085] The process of obtaining the abnormal coefficient of the device state is as follows:

[0086] Calculate the ratio of the device's running duration to the maximum continuous usage duration allowed for the device to obtain the running duration ratio. Then, calculate the ratios of the running duration ratio to the running power abnormal index and the running temperature abnormal index respectively and sum them up to obtain the abnormal coefficient of the device state.

[0087] In S5, warning processing is performed on devices in a severe abnormal state, and devices with minor abnormalities are further monitored to predict whether the future operating state of the device is abnormal. Based on the prediction results, dynamic regulation of the device is carried out, specifically including:

[0088] The warning processing for devices in a severe abnormal state specifically includes:

[0089] Once it is determined that the device is in a severe abnormal state, the system will automatically trigger a series of warning notification measures. First, the system will send instant alerts to relevant maintenance personnel, communicating information through multiple methods such as text messages, emails, or push notifications via mobile applications. At the same time, the system will also display a detailed abnormal report on the user interface, including information such as the time when the abnormality occurred, specific parameter values, and the degree of deviation from the normal range, to help technicians quickly locate the problem. In addition, according to the pre-established emergency plan, the system can recommend or directly execute some preliminary corrective measures, including: adjusting the device's workload or pausing operation to prevent the failure from deteriorating further. Finally, the maintenance team needs to conduct a comprehensive inspection based on the provided data and take repair measures to ensure that the device resumes normal operation. The entire process emphasizes the principle of combining automation and manual intervention, aiming to minimize downtime and economic losses. This systematic warning processing not only improves the efficiency and reliability of device management but also effectively reduces the risks brought by device failures, providing a solid guarantee for industrial production and intelligent manufacturing.

[0090] The further monitoring of devices with minor abnormalities specifically includes:

[0091] Obtain devices with minor abnormalities, extract the running time, running temperature abnormal index, and running power abnormal index of the devices, construct a comprehensive feature vector from the running time, running temperature abnormal index, and running power abnormal index, use it as the input of the machine learning model, and the output is the abnormal coefficient of the device state. According to the comparison between the abnormal coefficient of the device state and the preset threshold, predict whether the future operating state of the device is abnormal;

[0092] The construction process of the machine learning model is as follows:

[0093] Obtain the running time, running temperature anomaly index, and running power anomaly index. Construct a comprehensive feature vector from the running time, running temperature anomaly index, and running power anomaly index as the input of the machine learning model. Take minimizing the error between the predicted device status anomaly coefficient and the actual device status anomaly coefficient as the goal for model training. According to the training results, the model will output the device status anomaly coefficient for each future monitoring period. Based on the comparison of the device status anomaly coefficient with the preset first threshold and the preset second threshold respectively, identify the device status for each monitoring period, including: normal, slightly abnormal, and severely abnormal. Specifically:

[0094] By obtaining the running time, running temperature anomaly index, and running power anomaly index in real time from the device monitoring system, these key indicators reflect the health status of the device in different dimensions. To effectively integrate this information for subsequent analysis, we construct them into a comprehensive feature vector. Using a historical dataset that contains the labeled device status anomaly coefficients (i.e., actual values), we train a machine learning model with the goal of minimizing the error between the predicted device status anomaly coefficient and the actual device status anomaly coefficient. This step is achieved by selecting an appropriate loss function (such as mean squared error MSE) and using an optimization algorithm (such as gradient descent) for parameter adjustment. After sufficient training, the model can accurately predict the device status anomaly coefficient for each future monitoring period based on the input feature vector. After completing the model training, in actual application, whenever a new monitoring period arrives, the system automatically collects the running time, running temperature anomaly index, and running power anomaly index within the current period and constructs them into a feature vector to input into the trained model. The model outputs the predicted device status anomaly coefficient for this period. According to the preset first threshold and second threshold, we can classify the status of the device. The system not only provides simple status labels but also generates detailed reports, including the specific type of anomaly, possible reasons, and recommended maintenance measures.

[0095] It should be noted that: This advanced early warning mechanism not only improves the intelligent level of device management but also significantly enhances the fault prevention ability, helps reduce unplanned downtime, and improves production efficiency and safety. The entire process emphasizes the importance of data-driven decision-making and demonstrates how to achieve precise device health management through advanced machine learning techniques.

[0096] The device dynamic regulation based on the prediction results specifically includes:

[0097] When the machine learning model predicts that the equipment status anomaly coefficient in a future monitoring period is greater than or equal to the preset first threshold, indicating that the equipment is about to enter a severe abnormal state, the system automatically triggers a series of predefined dynamic regulation measures. These measures include real-time adjustment of the equipment's workload, including: the system will identify bottleneck links in the current workflow through sensor data, such as high-load motors or overheated components; the system will automatically adjust the task allocation on the production line to reduce the pressure on key components. If the spindle temperature of a machine tool rises abnormally, the system can temporarily reduce its processing tasks or transfer some workpieces to other available machine tools for processing.

[0098] Optimizing operating parameters Based on historical data and the current state, the system will re-evaluate and optimize the key operating parameters of the equipment, such as speed, pressure, temperature, etc. For example, for a pump system, if it is predicted that the motor may be damaged due to overload, the system will appropriately reduce the speed of the pump to reduce current and heat generation. The system will continuously monitor the effect of the adjusted parameters and further fine-tune according to the actual feedback to ensure that production efficiency is maintained while reducing risks.

[0099] Starting the standby system and performing preventive maintenance tasks, aiming to quickly reduce the pressure on the equipment and eliminate potential risks, thus avoiding serious failures and unplanned downtimes, including: once a potential failure risk of the equipment is detected, the system will immediately activate the pre-configured standby unit;

[0100] Exemplary: In a power supply system, if a transformer malfunctions, the system will automatically switch to the standby transformer to ensure uninterrupted power supply;

[0101] This process not only relies on accurate state prediction but also integrates automation control technology and an intelligent decision support system to ensure that the optimal action plan can be taken at the first time, maximizing the continuous and stable operation of the equipment, and significantly improving production efficiency and safety. Through this forward-looking dynamic regulation strategy, the present invention provides an efficient and reliable equipment health management solution for the fields of industrial automation and intelligent manufacturing.

[0102] Working principle of the present invention: By real-time monitoring the running time, running temperature and running power of the device, the device status is evaluated and potential abnormalities are predicted. During the monitoring period, the running time is recorded by the built-in timing module, the running temperature is collected by the thermocouple sensor, the running power is monitored by the power analyzer, and these data are transmitted to the central processing platform through the 5G communication module. The quality of the collected data is evaluated, and whether the data quality is qualified is judged by calculating the consistency ratio and the missing value ratio. For qualified data, the degree of change in the running temperature and the amplitude of fluctuation in the running power of the device are further analyzed, and the running temperature anomaly index and the running power anomaly index are calculated respectively by using Haar wavelet transform and fast Fourier transform. The device status anomaly coefficient is comprehensively calculated in combination with the running time. According to the preset first threshold and second threshold, the device status is classified as normal, slightly abnormal or severely abnormal. For severely abnormal devices, the system automatically triggers warning notification measures, including sending an immediate alarm to the maintenance personnel and suggesting preliminary corrective measures to prevent the deterioration of the failure; for slightly abnormal devices, the future running status is predicted through a machine learning model, and the working load of the device is dynamically adjusted, the running parameters are optimized or the standby system is started according to the prediction results to ensure the continuous and stable operation of the device. The whole process integrates automation control technology and intelligent decision support system, which not only improves the intelligent level of device management, but also significantly enhances the fault prevention ability, reduces the unplanned downtime, and improves the production efficiency and safety. This method is particularly applicable to the fields of industrial automation and intelligent manufacturing, and can effectively monitor the health status of the device, identify potential faults in advance, and ensure the continuity and stability of production.

[0103] The above formulas are all dimensionless and take their numerical values for calculation. The formula is obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0104] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that contains one or more collections of available media. The available media can be magnetic media (such as floppy disks, hard disks, magnetic tapes), optical media (such as DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0105] It should be understood that the term "and / or" in this document is merely a description of the association relationship between associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. In addition, the character " / " in this document generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, which can be specifically understood by referring to the context.

[0106] It should be understood that in various embodiments of the present application, the order numbers of the above processes do not indicate the sequence of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0107] The above has described in detail an embodiment of the present invention, but the content described is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. Any equivalent changes and improvements made within the scope of the application of the present invention should still fall within the scope covered by the patent of the present invention.

Claims

1. A data collection and supervisory control method based on an intelligent industrial park, characterized in that, It includes the following steps: S1: During the monitoring period, the device status data in the park is collected in real time, including: running time, running temperature, and running power; S2: Analyze the update frequency of the device status data, and judge whether the quality of the device status data is qualified according to the consistency ratio and missing value ratio of the device status data; S3: According to the judgment result, extract the qualified device status data, and evaluate the stability of the device status according to the running time, the degree of change in running temperature, and the stability of running power of the device, and judge whether the device status is abnormal; S4: According to the judgment result, divide the device running status into normal, slightly abnormal, and severely abnormal; S5: Carry out early warning processing on the devices in the severely abnormal state, further monitor the slightly abnormal devices, predict whether the future running status of the devices is abnormal, and perform dynamic regulation of the devices based on the prediction result.

2. The data acquisition and supervisory control method based on a smart industrial park according to claim 1, characterized in that, Judging whether the quality of the device status data is qualified specifically includes: Obtain the update frequency of the device status data, calculate the quality coefficient of the device status data according to the consistency ratio and missing value ratio of the device status data, and judge whether the quality coefficient of the device status data is greater than or equal to a preset threshold. If so, it is recorded as qualified; if not, it is recorded as unqualified.

3. A data acquisition and supervisory control method based on an intelligent industrial park according to claim 1, characterized in that, The process of obtaining the quality coefficient is: Obtain the time series of the device status data, and use dynamic time warping to calculate the distance between the actual device status data and the ideal data; The actual device status data is the device status data collected in real time, and the ideal data is the device status data of the preset normal standard; Calculate the update frequency normalization function, the consistency ratio function, and the penalty function, comprehensively process the update frequency normalization function, the consistency ratio function, and the penalty function, and calculate the quality coefficient through the comprehensive calculation expression.

4. A data acquisition and supervisory control method based on an intelligent industrial park according to claim 1, characterized in that Evaluating the stability of the device status specifically includes: Collect the running time, running temperature, and running power of the device in real time, analyze the running temperature of the device, calculate the running temperature anomaly index according to the degree of change in the running temperature of the device, analyze the running power of the device, calculate the running power anomaly index according to the fluctuation range of the running power, and comprehensively calculate and process the running time, the running temperature anomaly index, and the running power anomaly index of the device to obtain the device status anomaly coefficient, and judge the stability of the device status according to the device status anomaly coefficient.

5. A data acquisition and supervisory control method based on an intelligent industrial park according to claim 4, characterized in that Dividing the device running status into normal, slightly abnormal, and severely abnormal according to the evaluation result specifically includes: Compare the device status anomaly coefficient of the device with a preset first threshold. If the device status anomaly coefficient is greater than or equal to the preset first threshold, it is recorded as severely abnormal. If the device status anomaly coefficient is less than the preset first threshold and the device status anomaly coefficient is greater than the preset second threshold, it is recorded as slightly abnormal. If the device status anomaly coefficient is less than or equal to the preset second threshold, it is recorded as normal.

6. A data acquisition and supervisory control method based on an intelligent industrial park according to claim 4, characterized in that The process of obtaining the running temperature anomaly index is: Collect the temperature data of the device in real time according to the time series, perform Haar wavelet transform on the temperature data to obtain a series of approximation coefficients and detail coefficients; Calculate the ratio of the energy of each detail coefficient to the total energy to obtain the energy ratio of each detail coefficient. Obtain the mean value of the energy ratios of all decomposition levels of the Haar wavelet transform to obtain the operating temperature anomaly index.

7. A data acquisition and supervisory control method based on an intelligent industrial park according to claim 4, characterized in that The process of obtaining the operating power anomaly index is as follows: During the monitoring period, collect the operating power data of the device in real time according to the time series, perform a fast Fourier transform on the operating power data to obtain a set of spectral coefficients in complex form. Based on the transformed operating power data of the device, calculate the energy density of each frequency component. The calculation expression is: E a = |F a | 2 ; Among them, E a is the energy density of the a-th frequency component, representing a measure of the signal fluctuation intensity at the corresponding frequency, and F a represents the complex spectral coefficient of the a-th frequency component, and a represents the number of frequency components after the fast Fourier transform; Calculate the ratio of each frequency component to the total energy to obtain the spectral energy ratio. Calculate the average value of the energy ratios of all frequency components, calculate the difference between each spectral energy ratio and the average value of the energy ratios of all frequency components, take the absolute value to obtain the difference of the spectral energy ratio, and sum up all the differences to obtain the operating power anomaly index.

8. A data acquisition and supervisory control method based on an intelligent industrial park according to claim 1, characterized in that The further monitoring of the slightly abnormal device specifically includes: Obtain the slightly abnormal device, extract the operating time, operating temperature anomaly index, and operating power anomaly index of the device, construct the operating time, operating temperature anomaly index, and operating power anomaly index into a comprehensive feature vector as the input of the machine learning model, and the output is the device status anomaly coefficient. According to the comparison between the device status anomaly coefficient and the preset threshold, predict whether the future operating status of the device is abnormal.

9. A data acquisition and supervisory control method based on an intelligent industrial park according to claim 8, characterized in that, The construction process of the machine learning model is as follows: Obtain the operating time, operating temperature anomaly index, and operating power anomaly index, construct the operating time, operating temperature anomaly index, and operating power anomaly index into a comprehensive feature vector as the input of the machine learning model, and perform model training with the goal of minimizing the error between the predicted device status anomaly coefficient and the actual device status anomaly coefficient. According to the training results, the model will output the device status anomaly coefficient in each future monitoring period. According to the comparison between the device status anomaly coefficient and the preset first threshold and the preset second threshold respectively, identify the device status in each monitoring period, including: normal, slightly abnormal, and severely abnormal.

10. A data acquisition and supervisory control method based on an intelligent industrial park according to claim 1, characterized in that, The device dynamic regulation based on the prediction results specifically includes: For the device whose future status is predicted to be severely abnormal, adjust the working load of the device in real time, optimize the operating parameters, start the standby system, and perform preventive maintenance tasks.

Images