Load prediction method for industrial Internet of Things service platform

Abstract

The invention belongs to the technical field of load prediction, and discloses a load prediction method for an industrial Internet of Things service platform, which comprises the following steps: acquiring equipment real-time data through an industrial Internet of Things equipment data acquisition module; when the industrial Internet of Things service platform generates a prediction instruction, the prediction instruction is output to a load prediction module, and real-time data of current equipment is analyzed and predicted based on the load prediction module; the equipment real-time data comprises equipment operation data and equipment sensor data, and the equipment operation data comprises equipment vibration data and equipment temperature data; a real-time data result analyzed based on the load prediction module is transmitted to a feedback response module, and an equipment fault type is output through the feedback response module; the equipment fault type comprises a high-load runnable stage and a high-load non-runnable stage.

Description

Technical Field

[0001] The present invention belongs to the technical field of load prediction, and specifically relates to a load prediction method for an industrial Internet of Things service platform. Background Technique

[0002] The industrial Internet of Things service platform mainly refers to an integrated platform that utilizes technologies such as the Internet, Internet of Things, big data, and Al to integrate various industrial devices, achieve functions such as device connection, data collection and analysis, and optimization of production processes.

[0003] In the industrial Internet of Things service platform, load prediction mainly refers to predicting whether industrial devices can operate under load in the future. Load prediction mainly predicts the utilization rate and performance of various devices and sensors in the industrial Internet of Things to ensure their normal operation under high load and avoid performance degradation or failures caused by overload.

[0004] Currently, during the normal operation of the industrial Internet of Things, due to the differences in device operation, different devices often exhibit abnormal temperature and vibration loads. However, to ensure the production efficiency of industrial products, when a load occurs, generally, the machine is shut down for inspection. When there are minor problems with the failure, the shutdown detection step will result in low work efficiency.

[0005] Therefore, the present invention provides a load prediction method for an industrial Internet of Things service platform. Summary of the Invention

[0006] In order to make up for the deficiencies of the prior art and solve at least one of the technical problems proposed in the background technique.

[0007] The technical solution adopted by the present invention to solve its technical problems is as follows: A load prediction method for an industrial Internet of Things service platform of the present invention includes the following steps:

[0008] S1. Obtain device real-time data through the industrial Internet of Things device data acquisition module;

[0009] S2. When the industrial Internet of Things service platform generates a prediction instruction, output the prediction instruction to the load prediction module, and analyze and predict the current device real-time data based on the load prediction module;

[0010] The device real-time data includes device operation data and device sensor data, and the device operation data includes device vibration data and device temperature data;

[0011] S3. Transmit the real-time data result analyzed by the load prediction module to the feedback response module, and output the device failure type through the feedback response module;

[0012] The device failure type includes a high-load operable stage and a high-load inoperable stage;

[0013] S4. Calculate the load qualification index of the current industrial equipment through the load prediction module.

[0014] If the load qualification index is equal to 1, generate a device standard normal signal.

[0015] If the load qualification index is not equal to 1, generate a load deviation signal and transmit the load deviation signal to the feedback response unit to analyze the device failure type through the feedback response unit.

[0016] Preferably, the load prediction module analyzes and predicts the current device real-time data specifically as follows:

[0017] Collect an operating duration of the industrial equipment and set it as the time threshold A. 1 , obtain the time threshold A. 1 Obtain the device operation data and device sensor data of the industrial equipment within the time threshold A, and calculate the device vibration evaluation index and the device temperature anomaly coefficient based on the device operation data and device sensor data, and mark the device vibration evaluation index and the device temperature anomaly coefficient as Kc 1 and Kc 2 , and calculate the side industrial equipment evaluation coefficient M through the formula. 1 ;

[0018] Compare the side industrial equipment evaluation coefficient M 1 with the side device evaluation coefficient M in the industrial Internet of Things database. 2 Conduct a comparative analysis to calculate the load qualification index N of the current device. 1 , the load qualification index N of the current device. 1 is the ratio of the side industrial equipment evaluation coefficient M 1 to the side device evaluation coefficient M in the industrial Internet of Things database. 2 Conduct a discriminant analysis on the load qualification index N 1 and transmit the analysis result to step S4.

[0019] Preferably, the specific calculation method of the side industrial equipment evaluation coefficient M 1 is as follows:

[0020] where ω 1 and ω 2 are respectively the preset proportional factor coefficients of the vibration evaluation index and the device temperature anomaly coefficient. The proportional factor coefficient is used to correct the deviation that appears in the formula calculation of each parameter, so as to make the calculation result more accurate. ω 3 is the preset correction coefficient.

[0021] Preferably, the side device evaluation coefficient M in the industrial Internet of Things database 2The specific establishment method is as follows:

[0022] Establish a database, which includes historical data. The historical data includes equipment historical temperature data and equipment historical vibration data;

[0023] Calculate the temperature anomaly index α in the historical data 1 and the equipment vibration time index α 2 , and the side equipment evaluation coefficient M in the industrial Internet of Things database 2 is

[0024] where ω 4 and ω 5 are respectively the preset proportional factor coefficients of the vibration evaluation index and the equipment temperature anomaly coefficient. The proportional factor coefficients are used to correct the deviations that occur in the formula calculation of each parameter, so as to make the calculation result more accurate. ω 6 is the preset correction coefficient.

[0025] Preferably, the temperature anomaly index α in the historical data 1 and the equipment vibration time index α 2 The specific calculation method is as follows:

[0026] In the historical data, the temperature anomaly index α 1 includes the temperature first anomaly time point Te 1 , the time point Te from the first temperature anomaly to the end 2 , the time point Te when the equipment operation ends 3 ; the equipment vibration time index α 2 includes the equipment vibration first anomaly time point Tf 1 , the time point Tf from the first vibration anomaly of the equipment to the end 2 , the time point Tf when the equipment operation ends 3 ; (It should be noted that Te 3 and Tf 3 have the same time point)

[0027] where the temperature anomaly index

[0028] the equipment vibration time index

[0029] Preferably, the temperature first anomaly time point Te 1 , the time point Te from the first temperature anomaly to the end 2 , the equipment vibration first anomaly time point Tf 1 , the time point Tf from the first vibration anomaly of the equipment to the end 2 The specific selection method is as follows:

[0030] Periodically obtain the starting temperature to the ending temperature during the operation of multiple identical devices, and the starting vibration level to the ending vibration level during the operation of the devices. Establish multiple temperature sets and vibration level sets, and respectively establish Temperature-Time Curve 1, Temperature-Time Curve 2, Temperature-Time Curve 3, …, Temperature-Time Curve n based on the multiple temperature sets and vibration level sets 1 ,

[0031] Device Vibration-Time Curve 1, Device Vibration-Time Curve 2, Device Vibration-Time Curve 3, …, Device Vibration-Time Curve n 2 ;

[0032] And establish a device quantity set X = [x 1 , x 2 , …, x n based on multiple identical devices. Among them, x 1 is Device 1, x 2 is Device 2, x n is Device n. Device 1, Device 2, and Device n are identical devices. Calculate the mean value J of all the identical device quantities in the device quantity set X 1 ;

[0033] Obtain the initial time point when the temperature in the device quantity set X changes with time and tends to be stable, and mark the initial time point when it tends to be stable as ta 1 , and the initial time point when the device vibration changes with time and tends to be stable, and mark the initial time point when it tends to be stable as ta 2 ;

[0034] Respectively obtain the quantities of the initial time points ta 1 and ta 2 when the devices in the device quantity set X tend to be stable in the same way. If the quantities of the initial time points ta 1 and ta 2 when the devices in the device quantity set X tend to be stable in the same way are respectively greater than the mean value J 1 of all the identical device quantities in the device quantity set X, then ta 1 and ta 2 will be respectively the normal operating stable temperature and the normal operating stable vibration level of the device. Mark one of the Temperature-Time Curves as the Normal Device Temperature-Time Curve, and mark the corresponding device as the Temperature-Normal Device P 1 ;

[0035] Mark one of the Device Vibration-Time Curves as the Normal Device Vibration-Time Curve, and mark the corresponding device as the Vibration-Normal Device P 2 ;

[0036] Based on the normal temperature-time curve, compare the initial time point ta when the number of devices in the set X of devices tends to be stable with the normal temperature-time curve 1 Compare the devices that are different from the initial time point ta when tending to be stable 1 Mark the devices different from the normal temperature-time curve as temperature anomaly index devices, and mark the initial different temperature time point from the normal temperature-time curve as the initial anomaly time point Te 1 When the temperature of the normal temperature-time curve is the same from the initial different temperature time point, that is, mark the end as the time point Te 2 ,

[0037] Based on the normal device vibration-time curve, compare the initial time point ta when the number of devices in the set X of devices tends to be stable with the vibration normal device P 2 The initial time point ta when the device vibration-time curve tends to be stable 2 Compare the devices that are different from the initial time point ta when tending to be stable 2 Mark the devices different from the normal device vibration-time curve as vibration anomaly index devices, and mark the initial different vibration time point from the normal device vibration-time curve as the initial anomaly time point Tf 1 When the vibration of the normal device vibration-time curve is the same from the initial different vibration time point, that is, mark the end as the time point Tf 2 .

[0038] Preferably, based on step S4, if the device load qualification index at the current time is not equal to 1, a load deviation signal is generated, and the feedback response unit analyzes the specific device failure types as follows:

[0039] The industrial Internet of Things service platform includes a repeated monitoring unit, which is used to monitor the number of times of device temperature and vibration anomalies;

[0040] Calculate the temperature time normal time index μ 1 And the device vibration time normal index μ 2 ;

[0041] If the temperature time normal time index μ 1 Is less than the temperature anomaly index α 1 , or the device vibration time normal index μ 2 Is less than the device vibration time index α 2 , then a signal is generated and transmitted to the repeated monitoring unit;

[0042] If the temperature time normal time index μ 1 Is greater than the temperature anomaly index α 1 , or the device vibration time normal index μ 2 Is greater than the device vibration time index α 2 , then the feedback response unit outputs that the current device is in a high-load non-operable stage.

[0043] Preferably, the temperature-time normal time index μ 1 and the equipment vibration time normal index μ 2 The specific calculation method is as follows:

[0044] Obtain the equipment operation data at the current time, compare the equipment operation data at the current time with the normal temperature-time curve in the historical data. If the equipment operation data parameters at the current time are the same as those in the historical data, it is determined that the equipment operation data at the current time is in the normal period. If the equipment operation data parameters at the current time are different from those in the historical data, it is determined that the equipment operation data at the current time is in the abnormal period;

[0045] Obtain the initial temperature abnormal time point g of the abnormal period 2 , the vibration abnormal time point g 3 and the initial equipment operation time point g 1 ;

[0046] Then the temperature-time normal time index

[0047] The equipment vibration time normal index

[0048] Based on when the repeated monitoring unit receives the signal, that is, the temperature-time normal time index μ 1 is less than the temperature abnormal index α 1 , or the equipment vibration time normal index μ 2 is less than the equipment vibration time index α 2 ;

[0049] Obtain the number of temperature anomalies and the number of vibration anomalies in the abnormal period;

[0050] If the number of temperature anomalies in the abnormal period is greater than 1 and the number of vibration anomalies is greater than 1, the feedback response unit outputs that the current equipment is in the high-load non-operable stage;

[0051] If the number of temperature anomalies in the abnormal period is less than 1 or the number of vibration anomalies is less than 1, the feedback response unit outputs that the current equipment is in the high-load operable stage;

[0052] Specifically, when in the high-load operable stage, if the number of temperature anomalies and the number of vibration anomalies are too many, it may cause the equipment to continue to operate incorrectly under high load, that is, it is necessary to continue to judge the number of temperature anomalies and the number of vibration anomalies, that is, to monitor through the repeated monitoring unit.

[0053] Preferably, the industrial Internet of Things service platform includes an industrial Internet of Things device data acquisition module, a load prediction module, a feedback response module, and a repeated monitoring unit.

[0054] The beneficial effects of the present invention are as follows:

[0055] A load prediction method for an industrial Internet of Things service platform of the present invention analyzes the current data of industrial equipment through a load prediction module to determine whether production work can be carried out subsequently, that is, determines whether the equipment stops operating based on the type of equipment failure, improving equipment productivity and reducing unexpected downtime. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The present invention will be further described below with reference to the accompanying drawings.

[0057] Figure 1 is the flowchart of the method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0058] In order to make the technical means, creative features, achieved purposes and functions of the present invention easy to understand, the present invention will be further described below in conjunction with specific embodiments.

[0059] As Figure 1 shown, in an industrial Internet of Things service platform, load prediction mainly refers to predicting whether industrial equipment can perform load operation in a future period of time. Load prediction mainly predicts the usage rate and performance of various devices and sensors in the industrial Internet of Things to ensure that they can operate normally under high load and avoid performance degradation or failures caused by overload;

[0060] The industrial Internet of Things service platform mainly refers to an integrated platform that uses technologies such as the Internet, Internet of Things, big data, and Al to integrate various industrial equipment, realizing functions such as device connection, data collection and analysis, and optimization of production processes.

[0061] A load prediction method for an industrial Internet of Things service platform according to an embodiment of the present invention includes the following steps:

[0062] S1. Obtain device real-time data through an industrial Internet of Things device data acquisition module;

[0063] S2. When the industrial Internet of Things service platform generates a prediction instruction, output the prediction instruction to the load prediction module, and analyze and predict the current device real-time data based on the load prediction module;

[0064] The device real-time data includes device operation data and device sensor data, and the device operation data includes device vibration data and device temperature data;

[0065] S3. Transmit the real-time data result analyzed by the load prediction module to the feedback response module, and output the device failure type through the feedback response module;

[0066] The types of equipment failures include the high-load operable stage and the high-load inoperable stage. (During the high-load operable stage, the equipment can operate normally; during the high-load inoperable stage, the equipment cannot operate normally, that is, the equipment needs to be shut down.)

[0067] S4. Calculate the load qualification index of the current industrial equipment through the load prediction module.

[0068] If the load qualification index is equal to 1, generate a standard normal signal for the equipment.

[0069] If the load qualification index is not equal to 1, generate a load deviation signal and transmit the load deviation signal to the feedback response unit to analyze the type of equipment failure through the feedback response unit.

[0070] The load prediction module analyzes and predicts the real-time data of the current equipment specifically as follows:

[0071] Collect the operating duration of an industrial equipment for a period of time and set it as the time threshold A. 1 , obtain the time threshold A. 1 Obtain the equipment operation data and equipment sensor data of the industrial equipment within the time threshold A, and calculate the equipment vibration evaluation index and the equipment temperature anomaly coefficient based on the equipment operation data and equipment sensor data, and mark the equipment vibration evaluation index and the equipment temperature anomaly coefficient as Kc 1 and Kc 2 , and calculate the evaluation coefficient M of the side industrial equipment through a formula. 1 ;

[0072] Compare the evaluation coefficient M of the side industrial equipment 1 with the evaluation coefficient M of the side equipment in the industrial Internet of Things database 2 for comparative analysis, and calculate the load qualification index N of the current equipment 1 , the load qualification index N of the current equipment 1 is the ratio of the evaluation coefficient M of the side industrial equipment 1 to the evaluation coefficient M of the side equipment in the industrial Internet of Things database 2 , and perform discriminant analysis on the load qualification index N 1 , and transmit the analysis result to step S4.

[0073] By analyzing the historical and real-time data of the equipment, predict the possible failures of the equipment, perform maintenance in advance, reduce unexpected shutdowns, which failures can operate normally under high load, and which failures cannot operate normally under high load, and avoid performance degradation or failures caused by overload.

[0074] The specific calculation method of the evaluation coefficient M of the side industrial equipment 1 is as follows:

[0075] where ω 1 and ω 2 are respectively the preset scale factor coefficients of the vibration evaluation index and the equipment temperature anomaly coefficient. The scale factor coefficient is used to correct the deviation of each parameter in the formula calculation process, so as to make the calculation result more accurate. ω 3 is the preset correction coefficient;

[0076] The side device evaluation coefficient M in the industrial Internet of Things database 2 The specific establishment method is as follows:

[0077] Establish a database, the database includes historical data, and the historical data includes equipment historical temperature data and equipment historical vibration data;

[0078] Calculate the temperature anomaly index α 1 and the equipment vibration time index α 2 in the historical data. The side device evaluation coefficient M in the industrial Internet of Things database 2 is

[0079] where ω 4 and ω 5 are respectively the preset scale factor coefficients of the vibration evaluation index and the equipment temperature anomaly coefficient. The scale factor coefficient is used to correct the deviation of each parameter in the formula calculation process, so as to make the calculation result more accurate. ω 6 is the preset correction coefficient;

[0080] The temperature anomaly index α 1 and the equipment vibration time index α 2 in the historical data. The specific calculation method is as follows:

[0081] In the historical data, the temperature anomaly index α 1 includes the temperature first anomaly time point Te 1 , the time point Te 2 from the first temperature anomaly to the end, and the time point Te 3 when the equipment operation ends; the equipment vibration time index α 2 includes the equipment vibration first anomaly time point Tf 1 , the time point Tf 2 from the first vibration anomaly of the equipment to the end, and the time point Tf 3 when the equipment operation ends; (it should be noted that Te 3 and Tf 3 have the same time point).

[0082] where the temperature anomaly index

[0083] the equipment vibration time index

[0084] Initial abnormal temperature time point Te 1 and the time point Te when the temperature returns to normal from the initial abnormal temperature 2 Initial abnormal vibration time point Tf of the equipment 1 and the time point Tf when the equipment returns to normal from the initial abnormal vibration 2 The specific selection method is as follows:

[0085] Periodically obtain the temperatures from the start of operation to the end of operation of multiple identical devices, and the vibration levels from the start of operation to the end of operation of the devices, and establish multiple temperature sets and vibration level sets, and respectively establish Temperature-Time Curve 1, Temperature-Time Curve 2, Temperature-Time Curve 3, …, Temperature-Time Curve n based on the multiple temperature sets and vibration level sets 1 ,

[0086] Equipment Vibration-Time Curve 1, Equipment Vibration-Time Curve 2, Equipment Vibration-Time Curve 3, …, Equipment Vibration-Time Curve n 2

[0087] And establish an equipment quantity set X = [x 1 , x 2 , …, x n based on multiple identical devices, where x 1 is Device 1, x 2 is Device 2, x n is Device n, and Device 1, Device 2, …, Device n are identical devices. Calculate the mean value J of all the identical device quantities in the equipment quantity set X 1 ;

[0088] Obtain the initial time point when the temperature in the equipment quantity set X changes with time and tends to be stable, and mark the initial time point when it tends to be stable as ta 1 , and the initial time point when the equipment vibration changes with time and tends to be stable, and mark the initial time point when it tends to be stable as ta 2

[0089] Respectively obtain the quantities of the same initial time points ta 1 and ta 2 when they tend to be stable in the equipment quantity set X. If the quantities of the same initial time points ta 1 and ta 2 when they tend to be stable in the equipment quantity set X are respectively greater than the mean value J of all the identical device quantities in the equipment quantity set X 1 , then ta 1 and ta 2They are respectively the stable temperature and stable vibration level during the normal operation of the device. One of the temperature-time curves is marked as the normal device temperature-time curve, and the corresponding device is marked as the temperature-normal device P 1 ;

[0090] One of the device vibration-time curves is marked as the normal device vibration-time curve, and the corresponding device is marked as the vibration-normal device P 2

[0091] Based on the normal temperature-time curve, compare the initial time point ta when the temperature-time curve of the device in the device quantity set X tends to be stable 1 with the initial time point ta when it tends to be stable 1 Devices that are different are marked as temperature-abnormal-index devices, and the first different temperature-time point from the normal temperature-time curve is marked as the initial abnormal time point Te 1 , and from the first different temperature-time point to the time when the temperature is the same as that of the normal temperature-time curve, that is, at the end, it is marked as the time point Te 2 ,

[0092] Based on the normal device vibration-time curve, compare the initial time point ta when the device vibration-time curve of the device in the device quantity set X tends to be stable with the device P 2 of the device vibration-time curve when it tends to be stable 2 Devices that are different are marked as vibration-abnormal-index devices, and the first different vibration time point from the normal device vibration-time curve is marked as the initial abnormal time point Tf 2 , and from the first different vibration time point to the time when the vibration is the same as that of the normal device vibration-time curve, that is, at the end, it is marked as the time point Tf 1 , 2 ;

[0093] Fault types that can operate normally under high load:

[0094] Intermittent faults: These faults refer to the situation where the device loses some functions in the short term, but can be restored with a little repair or debugging without the need to replace parts. For example, temporary performance degradation caused by operation errors or environmental factors can be solved by simple adjustment or restarting the device

[0095] Software faults: Temporary failure of device functions caused by program errors or system bugs, usually can be restored by software updates or resets without physical intervention or hardware replacement

[0096] Fault types that cannot operate normally under high load:

[0097] Permanent faults: Such faults refer to the damage of some components of the equipment, which need to be replaced or overhauled to resume operation. For example, bearing wear or motor damage, etc. These faults will cause the equipment to malfunction under high load and require timely maintenance and replacement of components; Temperature rise: When the operating temperature of the equipment continues to rise and exceeds the normal range, the equipment should be immediately stopped, and the internal ventilation and heat dissipation facilities of the equipment should be checked and processed in a timely manner; Abnormal sound: When abnormal noise occurs during the operation of the equipment, it may be due to the fracture or wear of internal components of the equipment. At this time, the equipment needs to be immediately stopped for inspection and repaired.

[0098] Based on step S4, if the equipment load qualification index at the current time is not equal to 1, a load deviation signal is generated, and the feedback response unit analyzes the specific types of equipment faults as follows:

[0099] The industrial Internet of Things service platform includes a repeated monitoring unit, which is used to monitor the number of abnormal times of equipment temperature and vibration;

[0100] Calculate the temperature-time normal time index μ 1 and the equipment vibration time normal index μ 2 ;

[0101] If the temperature-time normal time index μ 1 is less than the temperature anomaly index α 1 , or the equipment vibration time normal index μ 2 is less than the equipment vibration time index α 2 , a signal is generated and transmitted to the repeated monitoring unit;

[0102] If the temperature-time normal time index μ 1 is greater than the temperature anomaly index α 1 , or the equipment vibration time normal index μ 2 is greater than the equipment vibration time index α 2 , the feedback response unit outputs that the current equipment is in the non-operable stage under high load;

[0103] The temperature-time normal time index μ 1 and the equipment vibration time normal index μ 2 The specific calculation method is as follows:

[0104] Obtain the equipment operation data at the current time, compare the equipment operation data at the current time with the normal temperature-time curve in the historical data. If the equipment operation data parameters at the current time are the same as those in the historical data, it is judged that the equipment operation data at the current time is in the normal period. If the equipment operation data parameters at the current time are different from those in the historical data, it is judged that the equipment operation data at the current time is in the abnormal period;

[0105] Obtain the initial temperature anomaly time point g during the abnormal period 2 , the vibration anomaly time point g 3 and the initial device operation time point g 1 ;

[0106] Then the temperature time normal time index

[0107] The device vibration time normal index

[0108] Based on when the repeated monitoring unit receives the signal, that is, the temperature time normal time index μ 1 is less than the temperature anomaly index α 1 , or the device vibration time normal index μ 2 is less than the device vibration time index α 2 ;

[0109] Obtain the number of temperature anomalies and the number of vibration anomalies during the abnormal period;

[0110] If the number of temperature anomalies during the abnormal period is greater than 1 and the number of vibration anomalies is greater than 1, then the feedback response unit outputs that the current device is in a high-load non-operable stage;

[0111] If the number of temperature anomalies during the abnormal period is less than 1 or the number of vibration anomalies is less than 1, then the feedback response unit outputs that the current device is in a high-load operable stage;

[0112] Specifically, when in the high-load operable stage, if the number of temperature anomalies and the number of vibration anomalies are too many, it may cause the device to continue to operate incorrectly under high load, that is, it is necessary to continue to judge the number of temperature anomalies and the number of vibration anomalies, that is, monitor through the repeated monitoring unit.

[0113] The industrial Internet of Things service platform includes an industrial Internet of Things device data acquisition module, a load prediction module, a feedback response module, and a repeated monitoring unit.

[0114] The setting of the size of the threshold is for the convenience of comparison. Regarding the size of the threshold, it depends on the amount of sample data and the base quantity set by those skilled in the art for each group of sample data; as long as it does not affect the proportional relationship between the parameters and the quantified values.

[0115] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution of the present invention and its inventive concept, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. A load prediction method for an industrial Internet of Things service platform, characterized in that: The following steps are involved: S1. Obtain real-time data of equipment through the industrial Internet of Things equipment data acquisition module; S2. When the industrial Internet of Things service platform generates a prediction instruction, the prediction instruction is output to the load prediction module, and the real-time data of the current device is predicted based on the load prediction module analysis; The real-time data of the equipment includes the equipment operation data and the equipment sensor data. The equipment operation data includes the equipment vibration data and the equipment temperature data. S3, the real-time data results analyzed by the load prediction module are transmitted to the feedback response module, and the equipment fault type is output through the feedback response module; Equipment failure types include high-load operable phase and high-load inoperable phase; S4. Calculate the load qualification index of the current industrial equipment through the load prediction module. If the load qualification index is equal to 1, a device standard normal signal is generated; If the load qualification index is not equal to 1, a load deviation signal is generated and transmitted to the feedback response unit, and the equipment fault type is analyzed by the feedback response unit.

2. A load prediction method for an industrial Internet of Things service platform according to claim 1, characterized in that: The load prediction module analyzes and predicts the current real-time data of the equipment as follows: Collect a period of operation time of industrial equipment and set it as time threshold A1, obtain equipment operation data and equipment sensor data of industrial equipment within time threshold A1, calculate equipment vibration evaluation index and equipment temperature anomaly coefficient based on equipment operation data and equipment sensor data, mark equipment vibration evaluation index and equipment temperature anomaly coefficient as Kc1 and Kc2 respectively, and calculate side industrial equipment evaluation coefficient M1 through formula; The side industrial equipment evaluation coefficient M1 is compared and analyzed with the side equipment evaluation coefficient M2 in the industrial Internet of Things database, and the load qualification index N1 of the current equipment is calculated. The load qualification index N1 of the current equipment is the ratio of the side industrial equipment evaluation coefficient M1 to the side equipment evaluation coefficient M2 in the industrial Internet of Things database. The load qualification index N1 is discriminated and analyzed, and the analysis results are transmitted to step S4.

3. A load prediction method for an industrial Internet of Things service platform according to claim 2, characterized in that: The specific calculation method of the side industrial equipment assessment coefficient M1 is: Where ω1 and ω2 are the preset proportional factor coefficients of the vibration evaluation index and the equipment temperature anomaly coefficient respectively, and ω3 is the preset correction coefficient.

4. A load prediction method for an industrial Internet of Things service platform according to claim 2, characterized in that: The specific method for establishing the side equipment evaluation coefficient M2 in the industrial Internet of Things database is as follows: Establish a database, the database includes historical data, the historical data includes equipment historical temperature data and equipment historical vibration data; Calculate the temperature anomaly index α1 and equipment vibration time index α2 in the historical data, and the side equipment evaluation coefficient M2 in the industrial Internet of Things database is Wherein ω4 and ω5 are the preset proportional factor coefficients of the vibration evaluation index and the equipment temperature anomaly coefficient respectively, and ω6 is the preset correction coefficient.

5. A load prediction method for an industrial Internet of Things service platform according to claim 4, characterized in that: The specific calculation method of the temperature anomaly index α1 and the equipment vibration time index α2 in the historical data is: In the historical data, the temperature anomaly index α1 includes the time point Te1 when the temperature first becomes abnormal, the time point Te2 when the temperature changes from the first abnormality to the end, and the time point Te3 when the equipment ends; the equipment vibration time index α2 includes the time point Tf1 when the equipment first becomes abnormal, the time point Tf2 when the equipment changes from the first abnormal vibration to the end, and the time point Tf3 when the equipment ends; (It should be noted that the time points Te3 and Tf3 are the same) The temperature anomaly index Equipment Vibration Time Index 6. A load prediction method for an industrial Internet of Things service platform according to claim 5, characterized in that: The specific selection method of the time point Te1 when the temperature is first abnormal, the time point Te2 when the temperature is from the first abnormal to the end, the time point Tf1 when the equipment vibration is first abnormal, and the time point Tf2 when the equipment is from the first abnormal vibration to the end is: Periodically obtain the start temperature to the end temperature of multiple identical devices, the start vibration degree to the end vibration degree of the devices, and establish multiple temperature sets and vibration degree sets, and respectively establish temperature-time curve 1, temperature-time curve 2, temperature-time curve 3, ┄, temperature-time curve n1 based on the multiple temperature sets and vibration degree sets; Equipment vibration-time curve 1, equipment vibration-time curve 2, equipment vibration-time curve 3, ┄, equipment vibration-time curve n2; And based on multiple identical devices, establish a device quantity set X = [x1, x2, ..., x n ], where x1 is device 1, x2 is device 2, and x n For device n, where device 1, device 2, and device n are identical devices, calculate the mean J1 of all identical device numbers in the device number set X; Obtain the first time point when the temperature in the equipment quantity set X changes with time and tends to be stable, and mark the first time point when it tends to be stable as ta1, and the first time point when the equipment vibration changes with time and tends to be stable, and mark the first time point when it tends to be stable as ta2; The numbers of the same initial time points ta1 and ta2 that tend to be stable in the equipment quantity set X are obtained respectively. If the numbers of the same initial time points ta1 and ta2 that tend to be stable in the equipment quantity set X are respectively greater than the mean value J1 of all the same equipment quantities in the equipment quantity set X, then ta1 and ta2 are the normal operating stable temperature and normal operating stable vibration degree of the equipment respectively, and one of the temperature-time curves is marked as the normal equipment temperature-time curve, and the corresponding equipment is marked as the normal temperature equipment P1; One of the equipment vibration-time curves is marked as a normal equipment vibration-time curve, and the corresponding equipment is marked as a normal vibration equipment P2; Based on the normal temperature-time curve, the equipment quantity set X is compared with the initial time point ta1 when the normal temperature-time curve tends to be stable, and the equipment that is different from the initial time point ta1 when it tends to be stable is marked as the temperature abnormality index equipment, and the time point when the temperature is different from the normal temperature-time curve for the first time is marked as the initial abnormal time point Te1, and the time point when the temperature is different from the normal temperature-time curve from the initial different temperature time point to the same temperature, that is, the end is marked as time point Te2; Based on the normal equipment vibration-time curve, the first time point ta2 when the vibration-time curve of the equipment P2 with normal vibration tends to be stable is compared in the equipment quantity set X, and the equipment that is different from the first time point ta2 when it tends to be stable is marked as the vibration abnormality index equipment, and the first vibration time point that is different from the normal equipment vibration-time curve is marked as the first abnormal time point Tf1, and the time point from the first different vibration time point to the same vibration time point as the normal equipment vibration-time curve, that is, the end is marked as time point Tf2.

7. A load prediction method for an industrial Internet of Things service platform according to claim 6, characterized in that: Based on step S4, if the equipment load qualification index at the current time is not equal to 1, a load deviation signal is generated, and the feedback response unit analyzes the equipment fault type as follows: The industrial Internet of Things service platform includes a repeated monitoring unit, which is used to monitor the temperature and vibration abnormality of the equipment; Calculate the temperature time normal time index μ1 and the equipment vibration time normal index μ2; If the temperature time normal time index μ1 is less than the temperature abnormal time index α1, or the equipment vibration time normal index μ2 is less than the equipment vibration time index α2, a signal is generated and transmitted to the repeated monitoring unit; If the normal temperature time index μ1 is greater than the abnormal temperature index α1, or the normal equipment vibration time index μ2 is greater than the equipment vibration time index α2, the feedback response unit outputs that the current equipment is in a high-load and non-operable stage.

8. A load prediction method for an industrial Internet of Things service platform according to claim 7, characterized in that: The specific calculation method of temperature time normal time index μ1 and equipment vibration time normal index μ2 is: Obtain the equipment operation data at the current time, and compare the equipment operation data at the current time with the normal temperature-time curve in the historical data. If the equipment operation data at the current time and the equipment operation data parameters in the historical data are the same, then the equipment operation data at the current time is judged to be in a normal period; if the equipment operation data at the current time and the equipment operation data parameters in the historical data are different, then the equipment operation data at the current time is judged to be in an abnormal period; Obtain the temperature anomaly initial time point g2, the vibration anomaly time point g3 and the equipment initial operation time point g1 during the abnormal period; Temperature time normal time index Equipment vibration time normal index 9. A load prediction method for an industrial Internet of Things service platform according to claim 8, characterized in that: Based on the signal received by the repeated monitoring unit, that is, the temperature time normal time index μ1 is less than the temperature abnormality index α1, or the equipment vibration time normal index μ2 is less than the equipment vibration time index α2; Obtain the number of abnormal temperature and abnormal vibration during the abnormal period; If the number of temperature anomalies during the abnormal period is greater than 1 and the number of vibration anomalies is greater than 1, the feedback response unit outputs that the current device is in a high-load and non-operable stage; If the number of temperature anomalies during the abnormal period is less than 1 or the number of vibration anomalies is less than 1, the feedback response unit outputs that the current device is in a high-load operable stage.

10. A load prediction method for an industrial Internet of Things service platform according to claim 1, characterized in that: The industrial Internet of Things service platform includes an industrial Internet of Things device data acquisition module, a load prediction module, a feedback response module and a repeated monitoring unit.

Images