Online equipment monitoring system based on artificial intelligence

Abstract

The invention discloses an equipment on-line monitoring system based on artificial intelligence, relates to the technical field of equipment on-line monitoring, and aims to realize real-time monitoring and optimization of equipment operation through an intelligent means. Real-time operation parameters are obtained from the equipment monitoring module; and the parameter processing module determines the optimal operation parameter interval of the equipment by analyzing the historical operation parameters of the equipment of the same kind. And the tuning and notification module performs tuning operation on the equipment based on the real-time operation parameters, and notifies a worker when an abnormality is detected. And the central controller is responsible for coordinating data transmission among the modules and transmitting an adjusting and optimizing control signal to the adjusting and optimizing and notification module, and through cooperative work of the modules, the system can realize intelligent monitoring and optimization of the operation state of the equipment and improve the operation efficiency of the equipment.

Description

Technical Field

[0001] The present invention belongs to the technical field of on-line monitoring of equipment. Specifically, it relates to an on-line monitoring system for equipment based on artificial intelligence. Background Art

[0002] With the continuous development of technology, artificial intelligence technology has promoted the transformation of equipment monitoring solutions, enabling on-line monitoring of equipment. This monitoring method analyzes the operating state of equipment by collecting the operating data of the equipment and combining algorithms and models.

[0003] In the prior art, there is a common situation of only monitoring without processing during the on-line monitoring of equipment, that is, simply collecting the operating data of the equipment, judging whether it is abnormal, but lacking in-depth analysis of the data and subsequent processing measures. It is difficult to optimize the operating state of the equipment based on the monitoring data, resulting in high maintenance costs and long downtime of the equipment, affecting production efficiency and economic benefits. To solve this problem, the present invention proposes an on-line monitoring system for equipment based on artificial intelligence. This system can not only monitor the operating data of the equipment in real time, but also achieve abnormal diagnosis and dynamic optimization of the equipment through dynamic analysis and automated processing, thereby comprehensively improving the operating efficiency, energy efficiency and reliability of the equipment, reducing maintenance costs and increasing production benefits. Summary of the Invention

[0004] The purpose of the present invention is to provide an on-line monitoring system for equipment based on artificial intelligence, to solve the common situation of only monitoring without processing during the on-line monitoring of equipment in the prior art. The present invention analyzes the historical operating parameters of the equipment, obtains the optimal operating parameter range of the equipment, and performs real-time optimization operations on abnormal equipment.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] An on-line monitoring system for equipment based on artificial intelligence, the system includes:

[0007] An equipment parameter acquisition module, which acquires the historical operating parameters of the equipment from the cloud database and the real-time operating parameters of the equipment from the equipment monitoring module;

[0008] A parameter processing module, which acquires the historical operating parameters of several similar equipment and determines the optimal operating parameter range of this type of equipment through data analysis;

[0009] An optimization and notification module, which acquires the real-time operating parameters of the equipment, performs optimization operations on the equipment, and notifies the information of abnormal equipment to the staff;

[0010] A central controller, which processes and coordinates the data transmission between each module, acquires the optimization operation control signal and transmits it to the optimization and notification module.

[0011] As a further solution of the present invention, both the historical operation parameters and the real-time operation parameters involved in the device parameter acquisition module include: load rate and operation temperature.

[0012] As a further solution of the present invention, the system further includes:

[0013] A cloud database for storing the historical operation parameters, real-time operation parameters of the device, and the data analyzed by this system;

[0014] A device monitoring module for obtaining the real-time operation parameters of the device;

[0015] A regular update module for updating the data analyzed by this solution.

[0016] As a further solution of the present invention, the method for the parameter processing module to obtain the historical operation parameters of several similar devices and determine the optimal operation parameter range of this type of device through data analysis is as follows:

[0017] Within one operation cycle, obtain the historical operation parameters of several similar devices, and the energy efficiency at the same timestamp as the corresponding historical operation parameters. Let the energy efficiency E obtained at the i-th timestamp i and the load rate F associated with this timestamp i serve as the load rate sample feature S i ;

[0018] Let the energy efficiency E obtained at the i-th timestamp i and the operation temperature T associated with this timestamp i serve as the temperature sample feature W i ;

[0019] where the energy efficiency E i and the load rate F i , and the operation temperature T i are all the means obtained after fitting the historical operation parameters of several similar devices;

[0020] where i represents the number of the timestamp, and its value range is from 1 to j. j represents the total number of timestamps taken at regular intervals within one operation cycle, and this fixed time is set by the staff;

[0021] Take the j load rate sample features as j data points and place them in a two-dimensional coordinate system, with the horizontal axis being the load rate and the vertical axis being the energy efficiency; connect adjacent data points to obtain the change line graph of the load rate sample feature;

[0022] Take the average energy efficiency E of this type of device within one operation cycle, and set the energy efficiency threshold e; use E i -E≥e to screen the load rate F associated with the qualified energy efficiency E i ​i And taking the maximum and minimum values of the preferred load rate as the preferred load rate, a preferred load rate range is formed;

[0023] Process the slope of the broken line graph of the change in the preferred load rate range to obtain the optimal load rate range.

[0024] As a further solution of the present invention, the method for the parameter processing module to process the slope of the broken line graph of the preferred load rate range is:

[0025] In the broken line graph of the change corresponding to the preferred load rate range, calculate the slope of two adjacent data points. If the slope is greater than 0, it indicates that the energy efficiency increases with the increase of the load rate;

[0026] If the slope is less than 0, it indicates that the energy efficiency decreases with the increase of the load rate;

[0027] Take the load rates corresponding to all adjacent data points with a positive slope, and form the optimal load rate range with the maximum and minimum values of the load rates;

[0028] Perform secondary processing on the slope.

[0029] As a further solution of the present invention, the method for the parameter processing module to perform secondary processing on the slope is:

[0030] Sort the slopes in descending order of value to obtain a slope sequence. The staff sets a slope threshold, and divides the high and low slope ranges from the sequence. The high slope range corresponds to the load rate range with low energy efficiency, and the low slope range corresponds to the load rate range with high energy efficiency, which are respectively recorded as the optimal high slope load rate range and the optimal low slope load rate range. The optimization and notification module further optimizes the device by combining these two ranges.

[0031] As a further solution of the present invention, for the temperature sample feature W i The processing is carried out according to the processing methods for obtaining the optimal load rate range, the optimal high slope frequency range, and the optimal low slope frequency range to obtain the optimal operating temperature range, the optimal high slope temperature range, and the optimal low slope temperature range.

[0032] As a further solution of the present invention, the method for the parameter processing module to obtain the optimal operating parameter range is:

[0033] Obtain the optimal load rate range and the optimal operating temperature range, and jointly use these two ranges as the optimal operating parameter range.

[0034] As a further solution of the present invention, the method for the optimization and notification module to obtain the real-time operating parameters of the device and perform optimization operations on the device is:

[0035] Continuously monitor the real-time load rate, real-time operating temperature, and real-time energy efficiency of the device. When the real-time energy efficiency of the device during normal operation is ≥ the average energy efficiency E, it is considered that the energy efficiency is normal and no processing is required;

[0036] If the real-time energy efficiency of the device during normal operation < the average energy efficiency E, then determine whether the real-time load rate and real-time operating temperature of the device are respectively within the preferred load rate range and the preferred operating temperature range. If either one is not within the preferred load rate range or the preferred operating temperature range, regard the device as an abnormal device and notify the staff;

[0037] The staff selects whether to perform a tuning operation on the abnormal device. If the answer is no, then no processing is done; if the answer is yes, obtain the tuning operation control signal and perform a basic tuning operation on the device;

[0038] The basic tuning operation is as follows:

[0039] S1. Obtain the real-time load rate and real-time operating temperature of the device during normal operation;

[0040] S2. Adjust the real-time load rate to the minimum value in the optimal load rate range, and adjust the real-time operating temperature to the minimum value in the optimal operating temperature range;

[0041] S3. Continuously monitor the real-time energy efficiency of the device. If the adjusted real-time energy efficiency ≥ the average energy efficiency E, the tuning operation ends;

[0042] S4. If the adjusted real-time energy efficiency is still lower than the average energy efficiency E, then fix the real-time load rate at the minimum value in the optimal load rate range, and gradually increase the real-time operating temperature according to the fixed value preset by the staff until the real-time operating temperature reaches the maximum value in the optimal operating temperature range;

[0043] Then increase the real-time load rate according to the fixed value preset by the staff, only increase one unit of the fixed value, then adjust the real-time operating temperature to the minimum value, and gradually increase the real-time operating temperature according to the fixed value preset by the staff until the real-time operating temperature reaches the maximum value in the optimal operating temperature range;

[0044] S5. Repeat the method described in S4 until the adjusted real-time load rate and real-time operating temperature have reached the maximum values in the optimal load rate range and the optimal operating temperature range, and continuously monitor the real-time energy efficiency during the process of adjusting the real-time load rate and real-time operating temperature;

[0045] If the real-time energy efficiency of the device ≥ the average energy efficiency E appears for the first time during the tuning process, the tuning operation ends;

[0046] S6. If the real-time energy efficiency of the device is always lower than the average energy efficiency E during the tuning process, then it is considered that the current device has a fault, and notify the staff of the fault information;

[0047] If the staff is not satisfied with the result of the current basic tuning operation, then a further tuning operation is selected.

[0048] As a further solution of the present invention, the method for the tuning and notification module to further tune the device by combining the best interval of the high-slope load rate and the best interval of the low-slope load rate is as follows:

[0049] Obtain the real-time load rate and real-time operating temperature associated with all cases where the real-time energy efficiency of the device ≥ the average energy efficiency E during the tuning process, and use them as a set of associated features;

[0050] If there are multiple associated features during the tuning operation, obtain the slope values of the real-time load rate and real-time operating temperature corresponding to all associated features on their respective corresponding change line graphs;

[0051] Classify the associated features according to the combination method between the best interval of the low-slope load rate, the best interval of the high-slope load rate, the best interval of the low-slope temperature, and the best interval of the high-slope temperature: the optimal associated feature, the second-best associated feature, and the worst associated feature;

[0052] If the number of the same type of associated features corresponding to the device during the tuning process does not exceed 1, select the corresponding associated feature as the result of the tuning operation in the order of the optimal associated feature, the second-best associated feature, and the worst associated feature, and the tuning operation ends;

[0053] If the number of the same type of associated features is not less than 1, obtain the slope values of the real-time load rate and real-time operating temperature of any associated feature on their respective change line graphs, and sum them to obtain the associated feature slope;

[0054] Calculate the associated feature slopes of all associated features in this tuning operation, and select the set of associated features with the smallest associated feature slope as the result of the tuning operation, and the tuning operation ends.

[0055] The beneficial effects of the present invention:

[0056] (1) By combining the relationship between the load rate sample characteristics, temperature sample characteristics and energy efficiency, the present invention quantifies and determines a load rate value range and an operating temperature value range that satisfy the device in the best energy efficiency state, can flexibly adjust the best operating parameter range according to different device characteristics and different working conditions, and has good adaptability and universality; it is easy to implement and integrate in the existing device monitoring system, and there is no need to carry out large-scale transformation of the device;

[0057] (2) The present invention provides a method for performing basic optimization operations on a device. This method continuously monitors the real-time energy efficiency of the device. Once the energy efficiency reaches the average value, the optimization operation immediately ends, avoiding resource waste caused by over-optimization operations. This method realizes the automated operation of device optimization and reduces the frequency and complexity of manual intervention; the staff only needs to perform simple selection operations when necessary, greatly improving the management efficiency.

[0058] (3) The present invention provides a method for performing further optimization operations on a device. On the basis of the basic optimization operation, the associated features are classified according to the combination of the optimal interval of the low-slope load rate, the optimal interval of the high-slope load rate, the optimal interval of the low-slope temperature, and the optimal interval of the high-slope temperature, providing a clear priority for selecting the final optimization result, making the optimization result more scientific and reasonable, effectively improving the energy efficiency of the device, reducing the risk of device failure, and enhancing the stability and reliability of device operation. Description of the Drawings

[0059] The present invention will be further described below in conjunction with the drawings.

[0060] Figure 1 is a schematic structural diagram of the system of the present invention;

[0061] Figure 2 is a schematic flow diagram of the method described in Embodiment 2;

[0062] Figure 3 is a schematic flow diagram of the method described in Embodiment 4. Detailed Embodiments

[0063] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0064] Embodiment 1

[0065] An on-line monitoring system for devices based on artificial intelligence, as Figure 1 shown, this system includes:

[0066] A device parameter acquisition module, which acquires the historical operation parameters of the corresponding device from the cloud database and the real-time operation parameters of the corresponding device from the device monitoring module;

[0067] Both the historical operation parameters and the real-time operation parameters include the load rate and the operating temperature.

[0068] The parameter processing module obtains the historical operation parameters of several similar devices, conducts data analysis on the historical operation parameters of the similar devices, and determines the optimal operation parameter range when the devices of this type are at the best energy efficiency;

[0069] The optimal operation parameter range includes: the optimal load rate range and the optimal operation temperature range.

[0070] The tuning and notification module is responsible for performing tuning operations on the devices and notifying the implementation information to the staff. The real-time information includes the information of abnormal devices and the data obtained from the analysis in this solution.

[0071] The central controller, as the core hub in the entire system, precisely processes and coordinates the data transmission between each module; obtains the tuning operation control signal set by the staff and transmits the tuning operation control signal to the tuning and notification module for performing the tuning operation;

[0072] The cloud database stores the real-time operation parameters, historical operation parameters of the devices, and the data obtained from the analysis in this solution.

[0073] The device monitoring module includes sensors that monitor the devices and obtain the device operation data in real time, and transmits the obtained device operation data to the cloud database.

[0074] The regular update module regularly executes the update policy, updates the data obtained from the analysis of this solution, and transmits it to the cloud database for storage.

[0075] Embodiment 2

[0076] This embodiment discloses a method for obtaining the preferred load rate range and the preferred operation temperature range, as Figure 2 shown, specifically including the following:

[0077] Taking the start time to the end time of a normal working process of a device as an operation cycle, for several similar devices, determining the operation cycles at the same time, obtaining the historical operation parameters of the several similar devices, and the energy efficiency at the same time stamp. The same time stamp refers to the same moment, and the value range of the time stamp is included in the taken operation cycle at the same time;

[0078] The calculation method of energy efficiency can be simplified as the working duration of the device in the normal working state divided by the power consumption of the corresponding device. For several similar devices, when the working duration is the same, the lower the power consumption, the higher the energy efficiency is regarded as;

[0079] The energy efficiency E of several devices obtained at the i-th time stamp i and the load rate F associated with this time stamp iAs a group, it is denoted as the load rate sample feature S i , where the energy efficiency E i and the load rate F i are both the means obtained by fitting the historical operation parameters of a number of similar devices;

[0080] Take the energy efficiency E of the device obtained at the i-th timestamp i and the operating temperature T associated with this timestamp i as a group, and denote it as the temperature sample feature W i , where the energy efficiency E i and the operating temperature T i are both the means obtained by fitting the historical operation parameters of a number of similar devices;

[0081] where i represents the number of a certain timestamp, and its value range is from 1 to j, and j represents the total number of timestamps taken at regular intervals within an operating cycle;

[0082] For the j timestamps, a total of j load rate sample features are obtained. Place these j load rate sample features in a two-dimensional coordinate system, with the horizontal axis corresponding to the load rate in the load rate sample feature and the vertical axis corresponding to the energy efficiency in the load rate sample feature;

[0083] Regard the j load rate sample features obtained as j data points and represent them in the two-dimensional coordinate system; starting from the first data point, connect them with short lines in sequence until the j-th data point is connected to obtain the change line graph of the load rate sample feature in the two-dimensional coordinate system;

[0084] Obtain the average energy efficiency E of all similar devices within an operating cycle, set the energy efficiency threshold e such that E i - E ≥ e, and take the load rate F i associated with the energy efficiency E that meets the threshold screening condition i as the preferred load rate, and form the preferred load rate interval by taking the maximum and minimum values in the preferred load rate. Similarly, obtain the preferred operating temperature interval.

[0085] Embodiment 3

[0086] This embodiment discloses a method for obtaining the optimal operating parameter interval, which specifically includes the following steps:

[0087] According to the method described in Embodiment 2, obtain the preferred load rate interval and the corresponding change line graph of the preferred load rate interval, and calculate the slopes of any two adjacent data points in the change line graph respectively;

[0088] If the calculated slope is greater than 0, the relationship between energy efficiency and load rate is a positive relationship at this time: the energy efficiency increases with the increase of the load rate. Within the load rate range corresponding to this slope, increasing the load rate has a positive impact on the energy efficiency, and the energy efficiency increases with the increase of the load rate;

[0089] If the slope is less than 0, it indicates that within the load rate range corresponding to this slope, the relationship between energy efficiency and load rate is a negative relationship: the energy efficiency decreases with the increase of the load rate. Within this load rate range, increasing the load rate is not conducive to the improvement of energy efficiency, but will instead reduce the energy efficiency;

[0090] Take all the load rate ranges with a positive slope, and use the maximum and minimum values of the load rate as the maximum and minimum values of the optimal load rate range respectively, to jointly form the optimal load rate range;

[0091] Obtain the optimal operating temperature range in the same way, and use the optimal load rate range and the optimal operating temperature range together as the optimal operating parameter range.

[0092] Embodiment 4

[0093] This embodiment discloses a method for optimizing the operation of equipment with abnormal energy efficiency, as Figure 3 shown, specifically including the following:

[0094] Continuously monitor the real-time load rate, real-time operating temperature, and real-time energy efficiency of the equipment in the normal working state. If the real-time energy efficiency of a certain equipment in the normal working state is higher than or equal to the average energy efficiency E of this type of equipment, then the real-time energy efficiency of this equipment is regarded as normal, and this equipment is regarded as a normal equipment, and no treatment is performed on this equipment;

[0095] If the real-time energy efficiency of a certain equipment in the normal working state is lower than the average energy efficiency E of this type of equipment, then further determine whether the real-time load rate and real-time operating temperature of this equipment are respectively within the preferred load rate range and the preferred operating temperature range. If any one of the real-time load rate and real-time operating temperature of this equipment is not within the above-mentioned preferred load rate range or preferred operating temperature range, then this equipment is regarded as an abnormal equipment, and the abnormal equipment information is notified to the staff;

[0096] The staff can choose whether to perform an optimization operation on the abnormal equipment. If the choice is no, then no treatment is performed on this abnormal equipment. If the choice is yes, then obtain the optimization operation control signal determined by the staff, and the optimization and notification module performs a basic optimization operation on this equipment;

[0097] The specific steps of the above basic optimization operation are as follows:

[0098] Step 1: Obtain the real-time load rate and real-time operating temperature of this abnormal equipment in the normal working state.

[0099] Step 2: Adjust the real-time load rate to the minimum value within the optimal load rate range for this type of device, and adjust the real-time operating temperature to the minimum value within the optimal operating temperature range for this type of device.

[0100] Step 3: After the preliminary adjustment, continuously monitor the real-time energy efficiency of the abnormal device. If the real-time energy efficiency after the preliminary adjustment is higher than or equal to the average energy efficiency E of this type of device, it is considered that the current abnormal device has returned to the normal state after the optimization operation, and this optimization operation ends.

[0101] Step 4: If the real-time energy efficiency after the preliminary adjustment is still lower than the average energy efficiency E of this type of device, fix the real-time load rate at the minimum value within the optimal load rate range, and increase the real-time operating temperature within the optimal operating temperature range according to the fixed value preset by the staff until the real-time operating temperature reaches the maximum value within the optimal operating temperature range.

[0102] Step 5: Then increase the real-time load rate according to the fixed value preset by the staff, only increase by one unit of the fixed value, then adjust the real-time operating temperature to the minimum value within the optimal operating temperature range, and increase the real-time operating temperature within the optimal operating temperature range according to the fixed value preset by the staff until the real-time operating temperature reaches the maximum value within the optimal operating temperature range.

[0103] Step 6: Repeat the method described in Step 5 until the adjusted real-time load rate reaches the maximum value within the optimal load rate range;

[0104] If during the optimization process, for the first time, the real-time energy efficiency of the abnormal device after adjusting the real-time load rate and real-time operating temperature is higher than or equal to the average energy efficiency E, it is considered that the current abnormal device has returned to the normal state after the optimization operation, and this optimization operation ends;

[0105] If the staff is not satisfied with the result of the current basic optimization operation, they can choose to perform a further optimization operation.

[0106] Step 7: If both the adjusted real-time load rate and real-time operating temperature have reached the maximum values within the optimal load rate range and optimal operating temperature range, and the real-time energy efficiency of this device has been lower than the average energy efficiency E during this optimization process, it is considered that there is a fault with the abnormal device, and the fault information is notified to the staff.

[0107] Example 4

[0108] This example discloses a method for further optimizing a device, including the following steps:

[0109] First, in Example 3, sort the slopes of the optimal load rate range and the optimal operating temperature range in descending order of value to obtain a slope sequence. The slope threshold is set by the staff, and a high-slope range and a low-slope range are divided from the slope sequence;

[0110] The high-slope range corresponds to a load rate range with relatively low energy efficiency, and the low-slope range corresponds to a load rate range with relatively high energy efficiency, which are respectively denoted as the optimal high-slope load rate range, the optimal low-slope load rate range, the optimal high-slope temperature range, and the optimal low-slope temperature range;

[0111] The tuning and notification module combines these two ranges to perform further tuning operations on the device;

[0112] In step six of the basic tuning operation, when the real-time energy efficiency of the abnormal device first appears to be higher than or equal to the average energy efficiency E after adjusting the real-time load rate and the real-time operating temperature, the tuning operation continues until both the adjusted real-time load rate and the real-time operating temperature have reached the maximum values in the optimal load rate range and the optimal operating temperature range;

[0113] Obtain the adjusted real-time load rate and the real-time operating temperature associated with all cases where the real-time energy efficiency of the abnormal device is higher than or equal to the average energy efficiency E during the tuning process, and use the adjusted real-time load rate and the real-time operating temperature as a set of associated features. A set of associated features corresponds to a real-time load rate and a real-time operating temperature;

[0114] If, during the tuning operation of a device, there are more than one set of associated features, obtain the slope values of all associated features and their corresponding real-time load rates and real-time operating temperatures on their respective change line graphs;

[0115] Classify the associated features according to the combination method between the optimal low-slope load rate range, the optimal high-slope load rate range, the optimal low-slope temperature range, and the optimal high-slope temperature range:

[0116] Among them, the associated features corresponding to the optimal low-slope load rate range and the optimal low-slope temperature range are the optimal associated features;

[0117] The two sets of associated features of the optimal high-slope load rate range and the optimal low-slope temperature range, and the optimal low-slope load rate range and the optimal high-slope temperature range are the second-best associated features;

[0118] The associated features corresponding to the optimal high-slope load rate range and the optimal high-slope temperature range are the worst associated features;

[0119] If the number of the same type of associated features corresponding to the device during the optimization process does not exceed 1, then starting from the optimal associated feature, select the corresponding associated feature in the order of the optimal associated feature, the second-best associated feature, and the worst associated feature. If there is no optimal associated feature, select the second-best associated feature. If there is still none, select the worst associated feature;

[0120] Take the real-time load rate and real-time operating temperature corresponding to the selected associated feature as the result of this optimization operation, and the optimization operation ends;

[0121] If the number of the same type of associated features is not less than 1, then obtain the slope values of the real-time load rate and real-time operating temperature of any associated feature on their respective change line graphs, and sum them to obtain the associated feature slope;

[0122] Compare the associated feature slopes of all associated features in this optimization operation, and take the real-time load rate and real-time operating temperature corresponding to the group of associated features with the smallest associated feature slope as the result of the final optimization operation, and the optimization operation ends;

[0123] This embodiment involves adjusting the real-time load rate and real-time operating temperature. Among them, the real-time operating temperature can be adjusted by a thermometer and a heat exchange device or a cooling device for the operating temperature of the device; adjusting the real-time load rate can be achieved through existing technologies, such as load balancing technology, reactive power compensation devices, etc., which can adjust the load rate.

[0124] The above content is only an example and explanation of the present invention. Those skilled in the art of this technology can make various modifications or supplements to the described specific embodiments or use similar methods to replace them. As long as they do not deviate from the invention or exceed the scope defined by this claim book, they should all fall within the protection scope of the present invention.

[0125] It should be stated that: all user data collected in this application is collected with the consent and authorization of the user. And the uses of user data are all legal and compliant, and the use and processing of user data comply with the relevant laws, regulations and standards of the relevant regions.

Claims

1. An on-line monitoring system for devices based on artificial intelligence, characterized in that, The system includes: A device parameter acquisition module, which acquires the historical operation parameters of the device from the cloud database and the real-time operation parameters of the device from the device monitoring module; A parameter processing module, which acquires the historical operation parameters of several similar devices and determines the optimal operation parameter range of this type of device through data analysis; An optimization and notification module, which acquires the real-time operation parameters of the device, performs optimization operations on the device, and notifies the information of abnormal devices to the staff; A central controller, which processes and coordinates the data transmission between modules, acquires the optimization operation control signal and transmits it to the optimization and notification module.

2. The on-line monitoring system for devices based on artificial intelligence according to claim 1, characterized in that, Both the historical operation parameters and the real-time operation parameters involved in the device parameter acquisition module include: load rate and operation temperature.

3. The on-line monitoring system for devices based on artificial intelligence according to claim 1, characterized in that The system further includes: A cloud database, which stores the historical operation parameters, real-time operation parameters of the device, and the data analyzed by this system; A device monitoring module, which acquires the real-time operation parameters of the device; A periodic update module, which updates the data analyzed by this solution.

4. The on-line monitoring system for devices based on artificial intelligence according to claim 1, characterized in that, The method by which the parameter processing module acquires the historical operation parameters of several similar devices and determines the optimal operation parameter range of this type of device through data analysis is as follows: During one operating cycle, obtain the historical operating parameters of several devices of the same type, and the energy efficiency at the same timestamp as the corresponding historical operating parameters. Take the energy efficiency E i obtained at the i-th timestamp and the load factor F i associated with this timestamp as the load factor sample feature S i ; Take the energy efficiency E obtained at the i-th timestamp i and the operating temperature T associated with this timestamp i as the temperature sample feature W i ; Among them, the energy efficiency E i and the load factor F i , and the operating temperature T i are all the means obtained by fitting the historical operating parameters of several similar devices; Where i represents the serial number of the timestamp, and the value range is from 1 to j. j represents the total number of timestamps taken at regular intervals within one operation cycle. This fixed time is set by the staff; Taking the j load rate sample features obtained as j data points and placing them in a two-dimensional coordinate system, with the horizontal axis being the load rate and the vertical axis being the energy efficiency; connecting adjacent data points to obtain the change line graph of the load rate sample features; Take the average energy efficiency E of this type of device within one operating cycle and set the energy efficiency threshold e; use E i - E ≥ e to screen the eligible energy efficiency E i The associated load factor F i And use it as the preferred load factor. Take the maximum and minimum values of the preferred load factor to form the preferred load factor range; Processing the slope of the change line graph of the preferred load rate range to obtain the optimal load rate range.

5. The on-line monitoring system for devices based on artificial intelligence according to claim 4, characterized in that The method by which the parameter processing module processes the slope of the line graph of the preferred load rate range is as follows: In the change line graph corresponding to the preferred load rate range, calculate the slope between two adjacent data points. If the slope is greater than 0, it means that the energy efficiency increases with the increase of the load rate; If the slope is less than 0, it means that the energy efficiency decreases with the increase of the load rate; Taking the load rates corresponding to all adjacent data points with a positive slope, and forming the optimal load rate range with the maximum and minimum values in the load rates; Performing secondary processing on the slope.

6. The on-line monitoring system for devices based on artificial intelligence according to claim 5, characterized in that, The method by which the parameter processing module performs secondary processing on the slope is as follows: Sorting the slopes from large to small by value to obtain a slope sequence. The staff sets a slope threshold, and divides the high and low slope ranges from the sequence. The high slope range corresponds to the load rate range with low energy efficiency, and the low slope range corresponds to the load rate range with high energy efficiency, which are respectively recorded as the optimal high slope load rate range and the optimal low slope load rate range. The optimization and notification module combines these two ranges to perform further optimization operations on the device.

7. The on-line monitoring system for an artificial-intelligence-based device according to claim 6, wherein The parameter processing module processes the temperature sample feature W i in accordance with the processing methods for obtaining the optimal load rate interval, the optimal high slope frequency interval, and the optimal low slope frequency interval to obtain the optimal operating temperature interval, the optimal high slope temperature interval, and the optimal low slope temperature interval.

8. The online monitoring system for devices based on artificial intelligence according to claim 7, characterized in that, The method by which the parameter processing module acquires the optimal operation parameter range is as follows: Acquiring the optimal load rate range and the optimal operation temperature range, and taking the two ranges together as the optimal operation parameter range.

9. The on-line monitoring system for devices based on artificial intelligence according to claim 8, characterized in that, The method by which the optimization and notification module acquires the real-time operation parameters of the device and performs optimization operations on the device is as follows: Continuously monitoring the real-time load rate, real-time operation temperature, and real-time energy efficiency of the device. When the real-time energy efficiency of the device during normal operation ≥ the average energy efficiency E, it is considered that the energy efficiency is normal and no processing is performed; If the real-time energy efficiency of the device during normal operation < the average energy efficiency E, determine whether the real-time load rate and real-time operating temperature of the device are respectively within the preferred load rate range and the preferred operating temperature range. If either is not within the preferred load rate range or the preferred operating temperature range, regard the device as an abnormal device and notify the staff; The staff selects whether to perform a tuning operation on the abnormal device. If the answer is no, do not process it; if the answer is yes, obtain the tuning operation control signal and perform a basic tuning operation on the device; The basic tuning operation is as follows: S1. Obtain the real-time load rate and real-time operating temperature of the device during normal operation; S2. Adjust the real-time load rate to the minimum value in the optimal load rate range, and adjust the real-time operating temperature to the minimum value in the optimal operating temperature range; S3. Continuously monitor the real-time energy efficiency of the device. If the adjusted real-time energy efficiency ≥ the average energy efficiency E, the tuning operation ends; S4. If the adjusted real-time energy efficiency is still lower than the average energy efficiency E, fix the real-time load rate at the minimum value in the optimal load rate range, and gradually increase the real-time operating temperature according to the fixed value preset by the staff until the real-time operating temperature reaches the maximum value in the optimal operating temperature range; Then increase the real-time load rate according to the fixed value preset by the staff, only increase one unit of the fixed value, then adjust the real-time operating temperature to the minimum value, and gradually increase the real-time operating temperature according to the fixed value preset by the staff until the real-time operating temperature reaches the maximum value in the optimal operating temperature range; S5. Repeat the method described in S4 until the adjusted real-time load rate and real-time operating temperature have reached the maximum values in the optimal load rate range and the optimal operating temperature range, and continuously monitor the real-time energy efficiency during the process of adjusting the real-time load rate and real-time operating temperature; If the real-time energy efficiency of the device ≥ the average energy efficiency E appears for the first time during the tuning process, the tuning operation ends; S6. If the real-time energy efficiency of the device is always lower than the average energy efficiency E during the tuning process, regard the current device as having a fault and notify the fault information to the staff; If the staff is not satisfied with the result of the current basic tuning operation, they choose to perform a further tuning operation.

10. The online monitoring system for devices based on artificial intelligence according to claim 9, characterized in that, The method for the tuning and notification module to perform a further tuning operation on the device by combining the optimal high-slope load rate range and the optimal low-slope load rate range is as follows: Obtain the real-time load rate and real-time operating temperature associated with all cases where the real-time energy efficiency of the device ≥ the average energy efficiency E during the tuning process, and use them as a set of associated features; If there are multiple associated features during the tuning operation, obtain the slope values of the real-time load rate and real-time operating temperature corresponding to all associated features on their respective corresponding change line graphs; Classify the associated features according to the combination method between the optimal low-slope load rate range, the optimal high-slope load rate range, the optimal low-slope temperature range, and the optimal high-slope temperature range: the optimal associated feature, the second-best associated feature, and the worst associated feature; If the number of the same type of associated features corresponding to the device during the tuning process does not exceed 1, select the corresponding associated feature as the result of the tuning operation in the order of the optimal associated feature, the second-best associated feature, and the worst associated feature, and the tuning operation ends; If the number of the same type of associated features is not less than 1, obtain the slope values of any associated feature's real-time load rate and real-time operating temperature on their respective change line charts, and sum them to obtain the associated feature slope; Calculate the associated feature slopes of all associated features in this tuning operation, and take the group of associated features with the smallest associated feature slope as the result of the tuning operation, and the tuning operation ends.

Images