Industrial Internet of Things platform based on equipment data real-time monitoring

Abstract

The invention relates to the technical field of industrial Internet of Things, is used for solving the problem that an existing industrial Internet of Things platform has defects in bidirectional intelligent management of production and equipment, and particularly discloses an industrial Internet of Things platform based on real-time monitoring of equipment data. Comprising a cloud database, a data acquisition module, a production analysis module, a production prediction module, an equipment analysis module, an equipment judgment module, a regulation and control processing module and a display terminal, according to the method, the production trend coefficient is extracted through STL time sequence decomposition, the model is built in combination with historical yield data to accurately predict the future yield, meanwhile, the equipment state is monitored in real time, the health value is generated, the abnormal signal is captured, and the equipment operation parameters are dynamically adjusted according to the prediction result and the abnormal signal, so that the dynamic linkage of production and equipment management is realized; and the production efficiency and the operation intelligence level are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial Internet of Things, and specifically to an industrial Internet of Things platform based on real-time monitoring of device data. Background Art

[0002] With the development of Industry 4.0, the intelligent transformation of industrial production has been accelerated. As an important pillar of industrial intelligence, the technology of Industrial Internet of Things (IIoT) is profoundly changing the production management mode of traditional industries. Through real-time data collection and analysis, IIoT technology can not only optimize the production process, but also realize intelligent monitoring of the operating status of devices, and at the same time improve the accuracy of production prediction, providing data-driven decision-making support for enterprises. However, the existing industrial Internet of Things platforms still have the following deficiencies in the two-way intelligent management of production and devices:

[0003] 1. Current production prediction methods usually rely on simple historical data trend analysis, lacking in-depth analysis of real-time production fluctuations, resulting in insufficient prediction accuracy. This limitation affects the accuracy of enterprise production plans and increases the complexity of inventory management and production scheduling;

[0004] 2. Traditional device monitoring means mainly focus on independent parameter evaluation, failing to make full use of multi-dimensional data for comprehensive health analysis, making it difficult to detect potential operating anomalies of devices in a timely manner. In addition, the linkage between device management and production requirements is insufficient, resulting in low efficiency in resource adjustment and device operation optimization;

[0005] 3. There is a lack of effective dynamic connection between production prediction results and device regulation, making it difficult to flexibly adjust device operation according to actual needs, resulting in low utilization efficiency of production resources and affecting the operational efficiency of enterprises.

[0006] To solve the above defects, a technical solution is provided now. Summary of the Invention

[0007] The purpose of the present invention is to solve the deficiencies of the existing industrial Internet of Things platforms in the two-way intelligent management of production and devices, mainly aiming at the problems of low production prediction accuracy and poor linkage between device management and regulation, and to propose an industrial Internet of Things platform based on real-time monitoring of device data.

[0008] The purpose of the present invention can be achieved through the following technical solutions:

[0009] An industrial Internet of Things platform based on real-time monitoring of device data, comprising:

[0010] A data collection module, used to collect production status information at the production end and device status information at the device end;

[0011] A production analysis module, which is used to obtain the output data at the production end, calculate the mean of the output data at different time points every day as the daily data mean, determine the first trend coefficient of the daily output data according to the daily data mean, calculate the fluctuation of the output data at different time points every day, and determine the second trend coefficient of the daily output data according to the fluctuation;

[0012] A production prediction module, which is used to calculate the trend difference of the trend coefficients in the current monitoring time period and the historical monitoring time period, and use the time series and trend difference of the output data in the historical monitoring time period as input data to construct and train a model, thereby obtaining the future output value;

[0013] An equipment analysis module, which is used to monitor the equipment status information at the equipment end to obtain the temperature setting evaluation index, flow setting evaluation index, pressure setting evaluation index and vibration setting evaluation index, so as to determine the health value of the equipment end;

[0014] An equipment determination module, which is used to determine the equipment operation trend evaluation value based on the health value of the equipment end, and determine whether to generate an operation trend abnormal signal based on the equipment operation trend evaluation value;

[0015] A regulation and processing module, which is used to perform regulation and processing on the equipment at the equipment end. Specifically: determine the equipment regulation instruction based on the future output value, and determine the equipment re-regulation instruction based on the difference between the equipment operation trend evaluation value and the equipment operation trend evaluation threshold.

[0016] Furthermore, the specific processes for solving the first trend coefficient and the second trend coefficient are as follows:

[0017] Sort the daily data means of different days in chronological order to obtain a mean sequence, perform time series decomposition on the mean sequence based on the STL time series decomposition method, and obtain the slope of the trend term as the first trend coefficient;

[0018] Determine the maximum value of the absolute value of the difference between the output data at different time points in a day and the daily data mean as the maximum difference, calculate the product of the maximum difference and the preset difference coefficient as the fluctuation coefficient, sort the daily fluctuation coefficients in chronological order to obtain a fluctuation sequence, and perform time series decomposition on the fluctuation sequence based on the STL time series decomposition method to obtain the slope of the trend term as the second trend coefficient.

[0019] Furthermore, the specific process for solving the future output value is as follows:

[0020] Obtain the time series of the first trend coefficient T 1 hist and the second trend coefficient T 2 hist of the historical monitoring period, and the specific expression is: where t is the time point, that is, every day;

[0021] Obtain the first trend coefficient T for the current monitoring time period 1 real and the second trend coefficient T 2 real for the time series, specifically expressed as:

[0022] Obtain the time series of the production data for the historical monitoring period, specifically expressed as: {S hist (t)};

[0023] Calculate the trend difference of the trend coefficients between the current monitoring time period and the historical monitoring period, according to the formula:

[0024]

[0025] Obtain the trend difference (ΔT 1 (t), ΔT 2 (t));

[0026] Use the time series {S hist (t)} of the production data for the historical monitoring period and the trend difference (ΔT 1 (t), ΔT 2 (t)) as input data to construct and train an ARIMA model, and use the trained ARIMA model to predict the production data for a future period of time to obtain the future production value Spred(t + f), where f is the prediction time step size.

[0027] Furthermore, the specific process of solving the temperature setting evaluation index is as follows:

[0028] Extract the temperature setting values at each monitoring time point from the device status information corresponding to the current monitoring period, denoted as sw i , where i represents the number of each monitoring time point in the current monitoring period, and i = 1, 2, 3... m, and m represents the total number of the numbers of each monitoring time point in the current monitoring period;

[0029] Extract the ambient temperature corresponding to the current monitoring period at the device end, and at the same time match the ambient temperature corresponding to the current monitoring period at the device end with the reference temperature setting values corresponding to each set ambient temperature to obtain the reference temperature setting value sw * ;

[0030] Calculate the temperature setting evaluation index SB1 corresponding to the current monitoring period at the device end through the following formula;

[0031] Specific formula: where sw i-1Let \(t_{i - 1}\) denote the set temperature value at the \((i - 1)\)-th monitoring time point in the current monitoring period, \(e\) denote the natural constant, \(\Delta s_w\) denote the set reference temperature difference, and \(a_1\), \(a_2\), \(a_3\) denote the influence factors corresponding to the average set temperature difference between adjacent monitoring time points, the set temperature difference between adjacent monitoring time points, and the reference set temperature difference, respectively.

[0032] Furthermore, the specific process for solving the set flow evaluation index is as follows:

[0033] Extract the set flow value at each monitoring time point from the device status information corresponding to the current monitoring period at the device end, denoted as \(s_l\). i ; and use the set flow value at the first monitoring time point in the current monitoring period at the device end as the reference set flow value, denoted as \(s_{l0}\). * ;

[0034] Calculate the set flow evaluation index \(SB_2\) at the device end corresponding to the current monitoring period through the following formula:

[0035] Specific formula: where \(s_{lz}\) i denotes the set flow fluctuation value at each monitoring time point, \(p\) denotes the natural constant, and \(a_4\), \(a_5\) denote the influence factors corresponding to the set flow value increase degree and set flow value decrease degree, respectively.

[0036] Furthermore, the specific process for solving the set pressure evaluation index is as follows:

[0037] Extract the set pressure value at each monitoring time point from the device status information corresponding to the current monitoring period at the device end, denoted as \(s_y\). i ; and use the set pressure value at the first monitoring time point in the current monitoring period at the device end as the reference set pressure value, denoted as \(s_{y0}\). * ;

[0038] Calculate the set pressure evaluation index \(SB_3\) at the device end corresponding to the current monitoring period through the following formula:

[0039] Specific formula: where \(s_{yz}\) i denotes the set pressure fluctuation value at each monitoring time point, \(q\) denotes the natural constant, and \(a_6\), \(a_7\) denote the influence factors corresponding to the set pressure value increase degree and set pressure value decrease degree, respectively.

[0040] Furthermore, the specific process for solving the set vibration evaluation index is as follows:

[0041] Extract the set vibration signal from the device status information corresponding to the current monitoring period at the device end, and generate a set vibration waveform diagram at the device end corresponding to the current monitoring period based on the specified software.

[0042] Extract the reference vibration waveform diagram corresponding to the device end. At the same time, overlap and compare the vibration waveform diagram of the device end during the current monitoring period with the reference vibration waveform diagram to obtain the vibration waveform overlap diagram of the device end during the current monitoring period, and extract the vibration deviation area from it, denoted as sz;

[0043] Calculate the vibration evaluation index SB4 of the device end during the current monitoring period through the following formula;

[0044] Specific formula: Among them, sz * represents the set reference vibration deviation area, Δsz represents the set vibration deviation area difference, and a8 represents the influence factor corresponding to the set vibration deviation area degree.

[0045] Furthermore, the specific process of solving the health value is as follows:

[0046] Extract the values of the temperature evaluation index SB1, flow evaluation index SB2, pressure evaluation index SB3, and vibration evaluation index SB4 of the device end during the current monitoring period for normalization processing, according to the formula: Obtain the health value JKZ of the device end. Among them, β1, β2, β3, and β4 respectively represent the evaluation factors corresponding to the set temperature evaluation index, flow evaluation index, pressure evaluation index, and vibration evaluation index.

[0047] Furthermore, determine whether to generate an abnormal operation trend signal. The specific process is as follows:

[0048] By obtaining the health values of the device end over a period of time, construct a two-dimensional health dynamic coordinate system;

[0049] Obtain the slope value between the health value of the device end over a period of time and the origin on the two-dimensional health dynamic coordinate system, thereby obtaining the health slope value. Set the health reference slope threshold, and compare and analyze the health slope value with the health reference slope threshold. When the health slope value is greater than or equal to the health reference slope threshold, the health state of the device end is determined to be an abnormal state. Count the number of times determined to be in an abnormal state, and calculate the proportion with the total number of determinations of the health state of the device end over a period of time to obtain the device operation trend evaluation value;

[0050] Compare and analyze the device operation trend evaluation value with the preset device operation trend evaluation threshold. When the device operation trend evaluation value is greater than or equal to the preset device operation trend evaluation threshold, generate an abnormal operation trend signal.

[0051] Furthermore, perform regulation and control processing on the device of the device end. The specific process is as follows:

[0052] Retrieve the future production value, match and analyze the future production value with the future production judgment table to obtain the future production level. At the same time, match it with the equipment operation parameters corresponding to the future production level to obtain the equipment operation parameters and generate equipment control instructions.

[0053] If an abnormal operation trend signal is captured, retrieve the equipment operation trend evaluation value at the equipment end, calculate the difference between the equipment operation trend evaluation value at the equipment end and the equipment operation trend evaluation threshold to obtain the equipment operation trend difference at the equipment end.

[0054] Substitute the equipment operation trend difference at the equipment end into the corresponding preset value range. Set different value ranges to correspond to an equipment operation parameter respectively and generate an equipment re-control instruction.

[0055] The technical solution provided by the present invention has the following beneficial effects compared with the known prior art:

[0056] 1. By obtaining key data such as production in real time, calculating the production data at different time points of each day, respectively extracting the mean value and fluctuation of the daily production data, and analyzing it using the STL time series decomposition method, the first trend coefficient representing the production trend change and the second trend coefficient representing the production fluctuation trend change are obtained, so as to comprehensively analyze the production trend. On this basis, compare the first trend coefficient and the second trend coefficient of the historical monitoring period and the current monitoring period in a time series, calculate the trend difference, use the trend difference and the historical production time series as input data, construct and train a prediction model using the ARIMA model, and finally accurately predict the production value in the future period of time, thus providing a scientific basis and data support for enterprise production and production decision-making.

[0057] 2. By monitoring and deeply evaluating the multi-dimensional state information (including set temperature, set flow, set pressure and set vibration) at the equipment end in real time, extracting the state data at the equipment end, calculating the set temperature evaluation index, set flow evaluation index, set pressure evaluation index and set vibration evaluation index, and combining normalization processing, a health value comprehensively reflecting the equipment operation state is generated, and a dynamic health coordinate system of the equipment is constructed using the health value. By analyzing the dynamic change trend of the health value over a period of time, abnormal signal capture is realized, thus effectively improving the prediction ability of equipment operation risks and providing reliable data support for the intelligent management and optimization of equipment.

[0058] 3. The present invention generates device control instructions based on the future production prediction results and transmits them to the device side through the cloud database interface to automatically adjust the device operation parameters, ensuring the dynamic matching of device operation and production requirements. At the same time, for the captured abnormal signals of the operation trend, it can automatically obtain the device operation trend evaluation value, calculate the difference between it and the threshold, and match the corresponding device operation parameters according to the difference for re-regulation, thereby effectively improving the adaptability and flexibility of the production system, ensuring that the device is always in the best operating state, meeting production requirements while reducing operation risks, and optimizing the overall production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for use in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0060] Figure 1 It is the overall module block diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0061] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0062] As Figure 1 shown, the industrial Internet of Things platform based on real-time monitoring of device data includes: a cloud database, which is connected to a data collection module, a production analysis module, a production prediction module, a device analysis module, a device determination module, a regulation and processing module, and a display terminal;

[0063] The data collection module is used to collect the production status information of the production side and the device status information of the device side, and send the production status information of the production side and the device status information of the device side to the cloud database for storage;

[0064] It should be noted that during the data collection process, by calling the production system API interface and receiving the sensor data transmitted by the Internet of Things gateway, the real-time and accurate collection of the information of the production side and the device side is realized, providing a solid foundation for subsequent data analysis and decision-making;

[0065] The cloud database is used to store the production status information of the production side, the equipment status information of the equipment side, the future output value, the future output determination table, the ambient temperature of the equipment side corresponding to each monitoring period, and the reference vibration waveform diagram corresponding to the equipment side;

[0066] The production analysis module monitors the production status information of the production side, thereby analyzing the production status of the production side. The specific analysis process is as follows:

[0067] Obtain the output data of the production side, calculate the average value of the output data at different time points every day as the daily data average value, determine the first trend coefficient of the daily output data according to the daily data average value, calculate the fluctuation of the output data at different time points every day, and determine the second trend coefficient of the daily output data according to the fluctuation;

[0068] Among them, the output data is the sum of the output data in each production area corresponding to industrial enterprises;

[0069] In the embodiment of the present invention, the average value of the output data at different time points every day can be calculated as the daily data average value;

[0070] Optionally, in the embodiment of the present invention, determining the first trend coefficient of the daily output data according to the daily data average value includes: sorting the daily data average values of different days in chronological order to obtain an average value sequence, performing time series decomposition on the average value sequence based on the STL time series decomposition method, and obtaining the slope of the trend term as the first trend coefficient;

[0071] Among them, the time series decomposition (Seasonal and Trend decomposition using Loess, STL) method is a time series decomposition method with robust locally weighted regression as the smoothing method, which decomposes the time series into corresponding trend terms, seasonal terms and residual terms. The STL time series decomposition method is a well-known time series decomposition method in the art and will not be elaborated here;

[0072] In the embodiment of the present invention, by performing STL time series decomposition on the average value sequence, the corresponding trend term curve is obtained. Subsequently, the slope of the trend term is used as the first trend coefficient. Since the trend term can represent the overall trend of the corresponding time series, the slope of the trend term can represent the change trend of the daily data average values corresponding to different days. That is to say, the first trend coefficient can represent the change trend of the daily data average values corresponding to different days;

[0073] Optionally, in the embodiments of the present invention, calculating the fluctuation of the production data at different time points every day, and determining the second trend coefficient of each day according to the fluctuation includes: determining the maximum value of the absolute value of the difference between the production data at different time points in a day and the daily data mean as the maximum difference, calculating the product of the maximum difference and the preset difference coefficient as the fluctuation coefficient, sorting the fluctuation coefficients of each day in chronological order to obtain a fluctuation sequence, and performing time series decomposition on the fluctuation sequence based on the STL time series decomposition method to obtain the slope of the trend term as the second trend coefficient;

[0074] In the embodiments of the present invention, sorting the fluctuation coefficients of each day in chronological order to obtain a fluctuation sequence, and performing time series decomposition on the fluctuation sequence based on the STL time series decomposition method to obtain the slope of the trend term as the second trend coefficient. This process of time series decomposition of the fluctuation sequence is similar to the STL time series decomposition method of the mean sequence in the above text, and will not be elaborated here. Thus, the second trend coefficient is obtained;

[0075] The production prediction module is used to predict and analyze the production volume at the production end for a period of time in the future. The specific analysis process is as follows:

[0076] Obtain the first trend coefficient T of each day in the historical monitoring period 1 hist and the second trend coefficient T 2 hist Thus, the first trend coefficient T of the historical monitoring period is obtained 1 hist and the second trend coefficient T 2 hist The time series is specifically expressed as: where t is the time point, that is, each day;

[0077] Obtain the first trend coefficient T of each day in the current monitoring period 1 real and the second trend coefficient T 2 real Thus, the first trend coefficient T of the current monitoring period is obtained 1 real and the second trend coefficient T 2 real The time series is specifically expressed as:

[0078] Obtain the production data S of each day in the historical monitoring period hist (t), thus obtaining the time series of the production data in the historical monitoring period, specifically expressed as: {S hist (t)};

[0079] Calculate the trend difference of the trend coefficients between the current monitoring period and the historical monitoring period, according to the formula:

[0080]

[0081] Obtain the trend difference (ΔT 1 (t), ΔT 2 (t));

[0082] Take the time series {S hist (t)} of the production data in the historical monitoring period and the trend difference (ΔT 1 (t), ΔT 2 (t)) as input data, construct and train an ARIMA model, predict the production data for a period of time in the future through the trained ARIMA model, obtain the future production value Spred(t + f), where f is the prediction time step, and send the future production value to the cloud database for storage;

[0083] The regulation processing module is used to regulate and process the devices at the device end. The specific process is as follows:

[0084] Retrieve the future production value, match and analyze the future production value with the future production judgment table stored in the cloud database, thereby obtaining the future production level. Each future production value corresponds to a future production level. At the same time, match it with the device operation parameters corresponding to the future production level, thereby obtaining the device operation parameters, generate a device regulation instruction, send the device regulation instruction to the device end through the device interface of the cloud database, and display a notification on the display terminal, where the device operation parameters include operating power, output, operating cycle, and operating environment;

[0085] The device analysis module is used to monitor the device status information of the devices at the device end, thereby analyzing the device status at the device end. The specific analysis process is as follows:

[0086] By obtaining the device status information corresponding to the current monitoring period at the device end, the device status information includes set temperature value, set flow value, set pressure value, and set vibration signal;

[0087] Optionally, in the embodiment of the present invention, collect the data at the collection points arranged on the collection device, such as device temperature, device current, device voltage, and device vibration signal. Subsequently, calculate the average value of these data respectively to obtain the average temperature, average current, average voltage, and average vibration signal of the device, and use these average values as the device status information;

[0088] Extract the set temperature values at each monitoring time point from the device status information corresponding to the current monitoring period at the device end, denoted as sw i , where i represents the number of each monitoring time point in the current monitoring period, and i = 1, 2, 3... m, and m represents the total number of the numbers of each monitoring time point in the current monitoring period;

[0089] Extract the environmental temperature of the device side corresponding to each monitoring period from the cloud database, and extract the environmental temperature of the device side corresponding to the current monitoring period from them. At the same time, match the environmental temperature of the device side corresponding to the current monitoring period with the reference temperature setting values corresponding to the set environmental temperatures to obtain the reference temperature setting value sw of the device side corresponding to the current monitoring period * ;

[0090] Calculate the temperature setting evaluation index SB1 of the device side corresponding to the current monitoring period through the following formula;

[0091] Specific formula: Among them, sw i-1 represents the temperature setting value at the (i - 1)-th monitoring time point in the current monitoring period, e represents the natural constant, Δsw represents the set reference temperature difference, and a1, a2, and a3 respectively represent the influence factors corresponding to the average temperature difference between adjacent monitoring time points, the temperature difference between adjacent monitoring time points, and the reference temperature difference;

[0092] Extract the current flow values of each monitoring time point from the device status information of the device side corresponding to the current monitoring period, denoted as sl i ; and use the current flow value at the first monitoring time point of the device side corresponding to the current monitoring period as the reference current flow value, denoted as sl * ;

[0093] Calculate the current flow evaluation index SB2 of the device side corresponding to the current monitoring period through the following formula;

[0094] Specific formula: Among them, slz i represents the current flow fluctuation value of each monitoring time point, p represents the natural constant, and a4 and a5 respectively represent the influence factors corresponding to the set increase degree of the current flow value and the set decrease degree of the current flow value;

[0095] Extract the current pressure values of each monitoring time point from the device status information of the device side corresponding to the current monitoring period, denoted as sy i ; and use the current pressure value at the first monitoring time point of the device side corresponding to the current monitoring period as the reference current pressure value, denoted as sy * ;

[0096] Calculate the current pressure evaluation index SB3 of the device side corresponding to the current monitoring period through the following formula;

[0097] Specific formula: Among them, syz i represents the current pressure fluctuation value of each monitoring time point, q represents the natural constant, and a6 and a7 respectively represent the influence factors corresponding to the set increase degree of the current pressure value and the set decrease degree of the current pressure value;

[0098] Extract the vibration signal from the device status information corresponding to the current monitoring period at the device end, and generate a vibration waveform graph corresponding to the current monitoring period at the device end based on the specified software, where the specified software is specifically vibration analysis software;

[0099] Extract the reference vibration waveform graph corresponding to the device end from the cloud database, and at the same time, overlap and compare the vibration waveform graph corresponding to the current monitoring period at the device end with the reference vibration waveform graph to obtain the vibration waveform overlap graph corresponding to the current monitoring period at the device end, and extract the vibration deviation area therefrom, denoted as sz;

[0100] Calculate the vibration evaluation index SB4 corresponding to the current monitoring period at the device end through the following formula;

[0101] Specific formula: where sz * represents the set reference vibration deviation area, Δsz represents the set vibration deviation area difference, and a8 represents the influence factor corresponding to the set vibration deviation area degree;

[0102] Extract the values of the temperature evaluation index SB1, current evaluation index SB2, pressure evaluation index SB3, and vibration evaluation index SB4 corresponding to the current monitoring period at the device end for normalization processing, according to the formula: Obtain the health value JKZ of the device end, where β1, β2, β3, and β4 respectively represent the evaluation factors corresponding to the set temperature evaluation index, current evaluation index, pressure evaluation index, and vibration evaluation index;

[0103] The device determination module is used to determine and analyze the device operation state of the device end. The specific analysis process is as follows:

[0104] By obtaining the health values of the device end within a period of time, using the period of time as the abscissa and the health value as the ordinate, thereby constructing a health dynamic coordinate system of the device end, and plotting the health values of the device end within a period of time on the two-dimensional health dynamic coordinate system by means of point plotting;

[0105] Obtain the slope value between the health value of the device end within a period of time and the origin on the two-dimensional health dynamic coordinate system, thereby obtaining the health slope value, set the health reference slope threshold, and compare and analyze the health slope value with the health reference slope threshold. When the health slope value is greater than or equal to the health reference slope threshold, then determine the health state of the device end as an abnormal state, count the number of devices determined to be in an abnormal state, and calculate the proportion with the total number of health state determinations of the device end within a period of time, according to the formula: Obtain the device operation trend evaluation value SQZ, where k 异 represents the number of devices determined to be in an abnormal state, k 总Let \(\alpha\) represent the total number of judgments. \(\gamma_1\) and \(\gamma_2\) respectively represent the set weight coefficients, and \(\gamma_1>\gamma_2\). The weight coefficients are used to balance the proportion weights of various data in the formula calculation, so as to promote the accuracy of the calculation results. The specific setting of the weight coefficients is reasonably set by those skilled in the art according to the actual situation;

[0106] Compare and analyze the device operation trend evaluation value with the preset device operation trend evaluation threshold. When the device operation trend evaluation value is greater than or equal to the preset device operation trend evaluation threshold, a running trend abnormal signal is generated. When the device operation trend evaluation value is less than the preset device operation trend evaluation threshold, a running trend normal signal is generated;

[0107] And send the generated running trend abnormal signal to the regulation and control processing module through the cloud database;

[0108] The regulation and control processing module is used to receive the running trend abnormal signal, and thus perform secondary regulation and control processing on the devices at the device end. The specific process is as follows:

[0109] If a running trend abnormal signal is captured, the device operation trend evaluation value at the device end is retrieved, and the difference between the device operation trend evaluation value at the device end and the device operation trend evaluation threshold is calculated to obtain the device operation trend difference at the device end;

[0110] Substitute the device operation trend difference at the device end into the corresponding preset value range. Different value ranges respectively correspond to a device operation parameter, and a device secondary regulation and control instruction is generated. According to the device secondary regulation and control instruction, the device operation parameters at the device end are regulated again, and a display notification is made on the display terminal.

[0111] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the present invention to only the specific implementation manners. Obviously, many modifications and changes can be made according to the content of this specification. These embodiments are selected and specifically described in this specification to better explain the principles and practical applications of the present invention, so that those skilled in the relevant technical fields can understand and utilize the present invention well. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. An industrial Internet of Things platform based on real-time monitoring of equipment data, including a data acquisition module for collecting production status information at the production end and equipment status information at the equipment end, characterized in that: Also includes: The production analysis module is used to obtain the production data of the production end, calculate the average of the production data at different time points every day as the daily data average, determine the first trend coefficient of the daily production data according to the daily data average, calculate the fluctuation of the production data at different time points every day, and determine the second trend coefficient of the daily production data according to the fluctuation; The production forecasting module is used to calculate the trend difference between the trend coefficients in the current monitoring period and the historical monitoring period, and use the time series and trend difference of the production data in the historical monitoring period as input data to build and train the model, thereby obtaining the future production value; The device analysis module is used to monitor the device status information of the device end, obtain the set temperature assessment index, the set flow assessment index, the set pressure assessment index and the set vibration assessment index, so as to determine the health value of the device end; The device determination module is used to determine the device operation trend evaluation value based on the health value of the device end, and determine whether to generate an operation trend abnormal signal; The control processing module is used to control the equipment at the equipment end, specifically: determine the equipment control instructions based on the future output value, and determine the equipment re-control instructions based on the difference between the equipment operation trend evaluation value and the equipment operation trend evaluation threshold.

2. The industrial Internet of Things platform based on real-time monitoring of equipment data according to claim 1 is characterized in that: The specific process of solving the first trend coefficient and the second trend coefficient is as follows: The daily data means of different days are sorted in time series order to obtain the mean sequence, and the mean sequence is decomposed based on the STL time series decomposition method to obtain the slope of the trend term as the first trend coefficient; Determine the maximum absolute value of the difference between the production data at different time points in a day and the daily data mean as the maximum difference, calculate the product of the maximum difference and the preset difference coefficient as the fluctuation coefficient, sort the daily fluctuation coefficients in chronological order to obtain the fluctuation sequence, and perform time series decomposition on the fluctuation sequence based on the STL time series decomposition method to obtain the slope of the trend item as the second trend coefficient.

3. The industrial Internet of Things platform based on real-time monitoring of equipment data according to claim 1 is characterized in that: The specific process of solving the future output value is as follows: Get the first trend coefficient T1 of the historical monitoring period hist and the second trend coefficient T2 hist The time series is expressed as: Among them, t is the time point, that is, daily; Get the first trend coefficient T1 of the current monitoring time period real and the second trend coefficient T2 real The time series is expressed as: Obtain the time series of production data during the historical monitoring period, specifically expressed as: {S hist (t)}; Calculate the trend difference between the trend coefficient in the current monitoring period and the historical monitoring period according to the formula: Obtain trend differences (ΔT1(t), ΔT2(t)); The time series of production data during the historical monitoring period {S hist (t)} and trend difference (ΔT1(t), ΔT2(t)) are used as input data to build and train the ARIMA model. The trained ARIMA model is used to predict the production data for a period of time in the future and the future production value Spred(t+f) is obtained, where f is the prediction time step.

4. The industrial Internet of Things platform based on real-time monitoring of equipment data according to claim 1 is characterized in that: The specific process of solving the temperature setting evaluation index is as follows: Extract the set temperature value at each monitoring time point from the device status information corresponding to the current monitoring period on the device side, denoted as sw i , i represents the number of each monitoring time point in the current monitoring period, and i=1, 2, 3…m, m represents the total number of monitoring time point numbers in the current monitoring period; Extract the ambient temperature of the device corresponding to the current monitoring period, and match the ambient temperature of the device corresponding to the current monitoring period with the reference temperature value corresponding to each set ambient temperature to obtain the reference temperature value sw of the device corresponding to the current monitoring period * ; The temperature setting assessment index SB1 of the device corresponding to the current monitoring period is calculated by the following formula; Specific formula: Among them, sw i-1 It represents the set temperature value at the i-1th monitoring time point in the current monitoring period, e represents the natural constant, Δsw represents the set reference set temperature difference, a1, a2, and a3 represent the influence factors corresponding to the average set temperature difference corresponding to adjacent monitoring time points, the set temperature difference corresponding to adjacent monitoring time points, and the reference set temperature difference, respectively.

5. The industrial Internet of Things platform based on real-time monitoring of equipment data according to claim 1 is characterized in that: The specific process of solving the assumption flow evaluation index is as follows: Extract the set flow value at each monitoring time point from the device status information corresponding to the current monitoring period on the device side, denoted as sl i ; and the set flow value of the device corresponding to the first monitoring time point of the current monitoring period is used as the reference set flow value, recorded as sl * ; The flow evaluation index SB2 of the device corresponding to the current monitoring period is calculated by the following formula; Specific formula: Among them, slz i It is represented as the set flow fluctuation value at each monitoring time point, p is represented as a natural constant, a4 and a5 are represented as the influencing factors corresponding to the degree of increase and decrease of the set flow value, respectively.

6. The industrial Internet of Things platform based on real-time monitoring of equipment data according to claim 1 is characterized in that: The specific process of solving the set pressure assessment index is as follows: Extract the set pressure value at each monitoring time point from the device status information corresponding to the current monitoring period on the device side, and record it as sy i ; and the set pressure value of the device corresponding to the first monitoring time point of the current monitoring period is used as the reference set pressure value, recorded as sy * ; The pressure assessment index SB3 of the device corresponding to the current monitoring period is calculated by the following formula; Specific formula: Among them, syz i It is represented as the set pressure fluctuation value at each monitoring time point, q is represented as a natural constant, a6 and a7 are represented as the influencing factors corresponding to the degree of increase and decrease of the set pressure value, respectively.

7. The industrial Internet of Things platform based on real-time monitoring of equipment data according to claim 1 is characterized in that: The specific process of solving the vibration evaluation index is as follows: Extracting a vibration setting signal from the device status information corresponding to the current monitoring period at the device end, and generating a vibration setting waveform diagram corresponding to the current monitoring period at the device end; Extract the reference vibration waveform corresponding to the device end, and at the same time, overlap and compare the vibration waveform corresponding to the current monitoring period of the device end with the reference vibration waveform to obtain the vibration waveform overlap diagram corresponding to the current monitoring period of the device end, and extract the vibration deviation area from it, recorded as sz; The vibration assessment index SB4 of the device corresponding to the current monitoring period is calculated by the following formula; Specific formula: Among them, sz * represents the set reference vibration deviation area, Δsz represents the set vibration deviation area difference, and a8 represents the influence factor corresponding to the set vibration deviation area degree.

8. The industrial Internet of Things platform based on real-time monitoring of equipment data according to claim 1 is characterized in that: The specific process of solving the health value is as follows: The values ​​of the set temperature assessment index SB1, set flow assessment index SB2, set pressure assessment index SB3 and set vibration assessment index SB4 corresponding to the current monitoring period on the device side are extracted and normalized according to the formula: The health value JKZ of the equipment side is obtained, where β1, β2, β3 and β4 respectively represent the evaluation factors corresponding to the set temperature evaluation index, flow evaluation index, pressure evaluation index and vibration evaluation index.

9. The industrial Internet of Things platform based on real-time monitoring of equipment data according to claim 1 is characterized in that: Determine whether to generate an abnormal operation trend signal. The specific process is as follows: By obtaining the health value of the device over a period of time, a two-dimensional health dynamic coordinate system is constructed; Obtain the slope value between the health value of the device over a period of time and the origin in the two-dimensional health dynamic coordinate system, thereby obtaining the health slope value, setting the health reference slope threshold, and comparing and analyzing the health slope value with the health reference slope threshold. When the health slope value is greater than or equal to the health reference slope threshold, the health status of the device is determined to be abnormal, and the number of abnormal states is counted, and the ratio is calculated with the total number of health status judgments of the device over a period of time to obtain the device operation trend evaluation value; The equipment operation trend evaluation value is compared and analyzed with a preset equipment operation trend evaluation threshold value. When the equipment operation trend evaluation value is greater than or equal to the preset equipment operation trend evaluation threshold value, an operation trend abnormality signal is generated.

10. The industrial Internet of Things platform based on real-time monitoring of equipment data according to claim 1, characterized in that: The device at the device end is regulated and processed. The specific process is as follows: Retrieving the future production value, matching and analyzing the future production value with the future production determination table, thereby obtaining the future production level, and matching it with the equipment operating parameters corresponding to the future production level, thereby obtaining the equipment operating parameters, and generating equipment control instructions; If an abnormal operation trend signal is captured, the device operation trend evaluation value of the device side is retrieved, and the difference between the device operation trend evaluation value of the device side and the device operation trend evaluation threshold is calculated to obtain the device operation trend difference of the device side; Substitute the equipment operation trend difference on the device side into the corresponding preset value range, set different value ranges to correspond to a device operation parameter respectively, and generate a device re-adjustment instruction.

Images