Intelligent park equipment operation monitoring method and system based on artificial intelligence

Abstract

The invention discloses a smart park equipment operation monitoring method and system based on artificial intelligence, and belongs to the technical field of operation monitoring. The system comprises a data acquisition module, an intelligent analysis module, an operation monitoring module and a visualization module. The data acquisition module is used for acquiring a parameter comparison table and a maintenance log of the equipment, and acquiring parameter records of all parts under the equipment in real time through a sensor; the intelligent analysis module is used for analyzing the maintenance logs, the parameter records of all the parts and the influence relation between the parts, establishing a relation network according to the influence relation between the parts and setting an abnormal network; the operation monitoring module establishes a relational expression according to the relationship among the parts in the abnormal network, and analyzes the number of abnormal times of each part through the relational expression, so as to identify the abnormal parts and give an alarm; and the visualization module displays the alarm information through a large visualization screen and informs management personnel of occurrence of an abnormal event.

Description

Technical Field

[0001] The present invention relates to the field of operation monitoring technology, and specifically to an artificial intelligence-based smart park equipment operation monitoring method and system. Background Art

[0002] Industrial parks usually gather a large number of industrial equipment, which plays an important role in improving production efficiency and reducing operating costs. Abnormal vibration not only affects the operating efficiency of equipment, but also may cause production interruption and equipment damage. Therefore, studying the causes and manifestations of abnormality has important practical significance.

[0003] At present, for the abnormal monitoring of equipment, threshold comparison and manual inspection are usually used. The vibration value of each component of the equipment is collected in real time through vibration sensors. If the threshold is exceeded, manual on-site inspection is reminded. This method has certain problems. For example, the various components under the equipment are often closely connected and influence each other due to mechanical connections, and the thresholds are different due to the different working methods of different components. When a component vibrates abnormally, it may not meet its own threshold requirements, but it is possible to transmit this vibration to other components that are mechanically connected to it, so that other components passively trigger abnormal vibration thresholds, resulting in misjudgment of abnormal situations. Therefore, at this stage, an efficient and intelligent technical solution for abnormal equipment operation monitoring is needed to solve the above problems. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for monitoring the operation of smart park equipment based on artificial intelligence to solve the problems raised in the above-mentioned background technology.

[0005] In order to solve the above technical problems, the present invention provides a method for monitoring the operation of smart park equipment based on artificial intelligence, comprising the following steps:

[0006] S100, collect parameter comparison tables and maintenance logs of equipment in the park, and collect parameter records of various components of the equipment in real time through sensors.

[0007] S200, comprehensively analyzing the maintenance log and the parameter records of each component, as well as the influence relationship between the components, establishing a relationship network according to the influence relationship between the components and setting an abnormal network.

[0008] S300, analyzing the relationship between the components in the abnormal network and establishing a relational expression, analyzing the number of abnormalities of each component according to the relational expression, thereby identifying the abnormal components and issuing an alarm.

[0009] S400, displays alarm devices through a large visual screen and notifies management personnel of abnormal events.

[0010] In S100, the parameter comparison table includes the vibration thresholds of each component of the equipment. The maintenance log refers to the work order record of each maintenance of the equipment, and each work order record includes the component and the maintenance time period. The parameter record includes the vibration value of the component at different times, which is collected in real time by the vibration sensor installed on the component.

[0011] Vibration sensors are installed on various parts of the equipment and monitor their vibration in real time. The sensors convert vibration signals into electrical signals and process these signals into digital data. After filtering and frequency domain analysis, vibration values ​​are obtained and recorded in real time in the parameter records of the corresponding parts. Vibration values ​​at any time in history or at the current time can be obtained through parameter records.

[0012] The parameter comparison table is a document that records the maximum vibration values ​​allowed for all components of the equipment during normal operation, including the criteria for determining the vibration threshold. This table helps engineers and maintenance personnel evaluate the working status of components and quickly identify potential failure risks and abnormal conditions by comparing the actual measured vibration data with the vibration threshold, so that maintenance measures can be taken in a timely manner.

[0013] In S200, the specific steps are as follows:

[0014] S201, filter out the CAP parts in the maintenance log n All work order records are collected, and a time set is created. The maintenance time periods in these work order records are put into the time set. n Parameter records of other parts other than the original equipment, and mark the maintenance time periods within the time set in these parameter records.

[0015] S202, set the sampling time time, and calculate the average vibration value V of each parameter record within the time time before each maintenance time period. before , and the average vibration value V in the subsequent time period after .

[0016] Mark V in the same time period after Greater than V before maintenance time period, set the error number e, and record the corresponding component CAP when the number of maintenance time periods is greater than e m Considered to be subject to the CAP of the component n Impact of component CAP n As an active influencer, component CAP m As a passive influencer, CAP for components n CAP and parts m Build influential relationships.

[0017] The active influencing party is the object of influence in the influence relationship, and the passive influencing party is the object of influence in the influence relationship. After the corresponding parts of the active influencing party are replaced or repaired, they can exert positive influence on the vibration of the passive influencing party in its influence relationship, making the vibration value of the passive influencing party larger.

[0018] S203, analyzing whether each component has an influence relationship with other components, and classifying the components that have an influence relationship with each other into the same category. There is no influence relationship between any component in different categories and any component in other categories.

[0019] All components in each category are connected according to the influence relationship and a relationship network is established. The current vibration values ​​of all parts and the vibration thresholds of all components in the parameter comparison table are obtained, and the relationship network to which the components whose vibration values ​​are greater than the corresponding vibration thresholds belong is regarded as an abnormal network.

[0020] When a component does not belong to any relationship network and its vibration value is greater than the corresponding vibration threshold, it is directly regarded as an abnormal component and the equipment automatically enters the alarm state.

[0021] In S300, the specific steps are as follows:

[0022] S301. Analyze the impact relationship between components in the abnormal network and mark all components CAP x Parts that have an impact relationship and are the active influencing party. Get the CAP of the parts in the maintenance log x The maintenance time period td recorded in the most recent work order rec , and the current time T now and maintenance time period td rec End time T end .

[0023] S302, obtain component CAP x And all the parameter records of the marked parts, retrieve T end to T now The vibration value during this period is recorded by creating a vibration line graph for each parameter, with time as the horizontal axis and vibration value as the vertical axis. Set sampling point e, evenly set e time points in the horizontal axis direction of each vibration line graph, and analyze the vibration value at each time point.

[0024] S303: Establish a training set for each annotated component, and use the vibration value of the annotated component at the same time point as the independent variable, as well as the component CAP x The vibration value is used as the dependent variable, the independent variable and the dependent variable are packaged as samples, and these e samples are put into the training set of the corresponding labeled parts.

[0025] S304, establish a linear regression model, substitute the independent variable of each sample in the training set as the input value into the linear regression model expression, calculate the difference between the output result of each sample and the dependent variable, sum the differences of all samples in the training set as the gap coefficient, and minimize the gap coefficient by adjusting the intercept F and the regression coefficient β in the expression, and obtain the relationship expression corresponding to each training set respectively:

[0026] C=F+β×W

[0027] In the formula, C is the dependent variable and W is the independent variable.

[0028] The gap coefficient indicates the accuracy of the relationship expression. The smaller the gap coefficient, the more accurately the corresponding relationship expression can describe the vibration influence relationship between the component and its annotated component.

[0029] S305, obtain the current vibration value of each marked component, substitute it into the corresponding relationship expression to obtain the calculation result, and analyze each calculation result and the component CAP x The difference between the current vibration values. Set the error coefficient r, count the number of calculation results h corresponding to the difference greater than r, and use h as the component CAP x The number of exceptions.

[0030] S306, analyze the marked parts corresponding to each component and calculate the number of abnormalities, set the abnormal threshold p, and treat the parts with abnormal times greater than p as abnormal parts. When there are abnormal parts in the equipment, it automatically enters the alarm state, generates alarm information according to the current vibration value of the abnormal parts and transmits it to the monitoring center.

[0031] In S400, the monitoring center displays the alarm information uploaded by the devices in alarm status in real time through a large visual screen, and notifies the management personnel of the occurrence of abnormal events.

[0032] An artificial intelligence-based smart park equipment operation monitoring system includes a data acquisition module, an intelligent analysis module, an operation monitoring module and a visualization module.

[0033] The data acquisition module is used to collect the parameter comparison table and maintenance log of the equipment, and collect the parameter records of each component under the equipment in real time through sensors.

[0034] The intelligent analysis module is used to analyze the maintenance log and the parameter records of each component, as well as the influence relationship between the components, establish a relationship network according to the influence relationship between the components, and set the abnormal network.

[0035] The operation monitoring module establishes a relational expression based on the relationship between components in the abnormal network, and analyzes the abnormal number of each component through the relational expression, so as to identify the abnormal components and issue an alarm.

[0036] The visualization module displays alarm information through a large visualization screen.

[0037] The data collection module includes an operation record collection unit, a maintenance record collection unit and a table information collection unit.

[0038] The operation record collection unit is used to collect parameter records of each component. The parameter records include the vibration values ​​of the components at different times, which are collected in real time by the vibration sensors installed on the components.

[0039] The maintenance record collection unit is used to collect the maintenance log of the equipment. The maintenance log refers to the work order record of each maintenance of the equipment. Each work order record includes parts and maintenance time period.

[0040] The table information collection unit is used to collect a parameter comparison table, which includes the vibration thresholds of various components under the equipment.

[0041] The intelligent analysis module includes an impact analysis unit and a network construction unit.

[0042] The impact analysis unit is used to analyze the impact relationship between components.

[0043] First, select the component CAP n All work order records are created, and the maintenance time periods in these work order records are put into the time set. n In the parameter records of other parts, mark the maintenance time periods within the time set.

[0044] Secondly, set the sampling time time, and calculate the average vibration value V in each parameter record within the time length before each maintenance time period. before , and the average vibration value V in the subsequent time period after ; Mark V in the same time period after Greater than V before maintenance time period.

[0045] Finally, set the error times e, and record the corresponding parts CAP of the parameter with the maintenance time period times greater than e m Considered to be subject to the CAP of the component n Impact of component CAP n As an active influencer, component CAP m As a passive influencer, CAP for components n CAP and parts m Build influential relationships.

[0046] The network building unit is used to build a relationship network.

[0047] First, analyze whether each component has an impact relationship with other components, and classify the components that have an impact relationship with each other into the same category. There is no impact relationship between any component in different categories and any component in other categories.

[0048] Then, all the components in each category are connected according to the influence relationship and a relationship network is established to obtain the current vibration values ​​of all parts and the vibration thresholds of all components in the parameter comparison table. The relationship network to which the components with vibration values ​​greater than the corresponding vibration threshold belong is regarded as an abnormal network.

[0049] The operation monitoring module includes an operation analysis unit and an abnormal alarm unit.

[0050] The operation analysis unit is used to establish a relational expression based on the relationship between components in the abnormal network.

[0051] First, mark all the abnormal networks with component CAP x For parts that have an impact relationship and are the active influencing party, obtain the CAP of the parts in the maintenance log x The maintenance time period td recorded in the most recent work order rec , and the current time T now and maintenance time period td rec End time T end .

[0052] Second, obtain the parts CAP x And all the parameter records of the marked parts, retrieve T end to T now The vibration value during this period is recorded by creating a vibration line graph for each parameter, with time as the horizontal axis and vibration value as the vertical axis. Set sampling point e, evenly set e time points in the horizontal axis direction of each vibration line graph, and analyze the vibration value at each time point.

[0053] Then, a training set is established for each annotated component, and the vibration value of the annotated component at the same time point is used as the independent variable, as well as the component CAP x The vibration value is used as the dependent variable, the independent variable and the dependent variable are packaged as samples, and these e samples are put into the training set of the corresponding labeled parts.

[0054] Finally, a linear regression model is established, and the independent variables of each sample in the training set are substituted into the linear regression model expression as input values. The difference between the output result of each sample and the dependent variable is calculated, and the difference of all samples in the training set is summed as the gap coefficient. The gap coefficient is minimized by adjusting the intercept F and the regression coefficient β in the expression, and the corresponding relationship expression for each training set is obtained: C=F+β×W, where C is the dependent variable and W is the independent variable.

[0055] The abnormal alarm unit is used to identify abnormal parts and issue an alarm.

[0056] First, obtain the current vibration value of each marked component, substitute it into the corresponding relationship expression to obtain the calculation result, and analyze each calculation result and the component CAP x The difference between the current vibration values.

[0057] Secondly, set the error coefficient r, count the number of calculation results h corresponding to the difference greater than r, and use h as the component CAP x Analyze the anomaly times of each component and calculate the anomaly times.

[0058] Finally, an abnormal threshold p is set, and parts with abnormal times greater than p are regarded as abnormal parts. When there are abnormal parts in the equipment, it will automatically enter the alarm state, generate alarm information based on the vibration value of the abnormal parts and transmit it to the monitoring center.

[0059] The visualization module displays the alarm information uploaded by the alarm status equipment in real time through the visualization large screen of the monitoring center.

[0060] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0061] This application first establishes a time set by analyzing the work order records of the parts, analyzes whether there is an influence relationship between the parts based on the time set, and defines active influencers and passive influencers. Secondly, a relationship network is established based on the influence relationship and the abnormal network is identified, the relationship between the parts under the abnormal network is analyzed, multiple labeled parts are set for each part, and a vibration line graph and a relationship expression are established based on the parameter records of the labeled parts. Finally, the current vibration value of the part is substituted into the relationship expression, the calculated result is compared with the actual value and the number of abnormalities of the part is marked, and the abnormal parts are identified according to the number of abnormalities and an alarm is issued. Compared with the monitoring method of the prior art that simply performs threshold judgment and manual inspection, the present application can detect anomalies more quickly and efficiently, and the positioning and identification of abnormal parts are also more intelligent.

[0062] In summary, compared with traditional technologies, the present invention has the advantages of rapid anomaly screening and intelligent anomaly positioning, and can improve the efficiency of anomaly detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0064] Figure 1 It is a flow chart of a method for monitoring the operation of smart park equipment based on artificial intelligence of the present invention;

[0065] Figure 2 It is a structural schematic diagram of an artificial intelligence-based smart park equipment operation monitoring system of the present invention. DETAILED DESCRIPTION

[0066] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] See also Figure 1 The present invention provides a method for monitoring the operation of smart park equipment based on artificial intelligence, comprising the following steps:

[0068] S100, collect parameter comparison tables and maintenance logs of equipment in the park, and collect parameter records of various components of the equipment in real time through sensors.

[0069] S200, comprehensively analyzing the maintenance log and the parameter records of each component, as well as the influence relationship between the components, establishing a relationship network according to the influence relationship between the components and setting an abnormal network.

[0070] S300, analyzing the relationship between the components in the abnormal network and establishing a relational expression, analyzing the number of abnormalities of each component according to the relational expression, thereby identifying the abnormal components and issuing an alarm.

[0071] S400, displays alarm devices through a large visual screen and notifies management personnel of abnormal events.

[0072] In S100, the parameter comparison table includes the vibration thresholds of each component of the equipment. The maintenance log refers to the work order record of each maintenance of the equipment, and each work order record includes the component and the maintenance time period. The parameter record includes the vibration value of the component at different times, which is collected in real time by the vibration sensor installed on the component.

[0073] Vibration sensors are installed on various parts of the equipment and monitor their vibration in real time. The sensors convert vibration signals into electrical signals and process these signals into digital data. After filtering and frequency domain analysis, vibration values ​​are obtained and recorded in real time in the parameter records of the corresponding parts. Vibration values ​​at any time in history or at the current time can be obtained through parameter records.

[0074] The parameter comparison table is a document that records the maximum vibration values ​​allowed for all components of the equipment during normal operation, including the criteria for determining the vibration threshold. This table helps engineers and maintenance personnel evaluate the working status of components and quickly identify potential failure risks and abnormal conditions by comparing the actual measured vibration data with the vibration threshold, so that maintenance measures can be taken in a timely manner.

[0075] In S200, the specific steps are as follows:

[0076] S201, filter out the CAP parts in the maintenance log n All work order records are collected, and a time set is created. The maintenance time periods in these work order records are put into the time set. n Parameter records of other parts other than the original equipment, and mark the maintenance time periods within the time set in these parameter records.

[0077] S202, set the sampling time time, and calculate the average vibration value V of each parameter record within the time time before each maintenance time period. before , and the average vibration value V in the subsequent time period after .

[0078] Mark V in the same time period after Greater than V before maintenance time period, set the error number e, and record the corresponding component CAP when the number of maintenance time periods is greater than e m Considered to be subject to the CAP of the component n Impact of component CAP n As an active influencer, component CAP m As a passive influencer, CAP for components n CAP and parts m Build influential relationships.

[0079] The active influencing party is the object of influence in the influence relationship, and the passive influencing party is the object of influence in the influence relationship. After the corresponding parts of the active influencing party are replaced or repaired, they can exert positive influence on the vibration of the passive influencing party in its influence relationship, making the vibration value of the passive influencing party larger.

[0080] S203, analyzing whether each component has an influence relationship with other components, and classifying the components that have an influence relationship with each other into the same category. There is no influence relationship between any component in different categories and any component in other categories.

[0081] All components in each category are connected according to the influence relationship and a relationship network is established. The current vibration values ​​of all parts and the vibration thresholds of all components in the parameter comparison table are obtained, and the relationship network to which the components whose vibration values ​​are greater than the corresponding vibration thresholds belong is regarded as an abnormal network.

[0082] When a component does not belong to any relationship network and its vibration value is greater than the corresponding vibration threshold, it is directly regarded as an abnormal component and the equipment automatically enters the alarm state.

[0083] In S300, the specific steps are as follows:

[0084] S301. Analyze the impact relationship between components in the abnormal network and mark all components CAP x Parts that have an impact relationship and are the active influencing party. Get the CAP of the parts in the maintenance log x The maintenance time period td recorded in the most recent work order rec , and the current time T now and maintenance time period td rec End time T end .

[0085] S302, obtain component CAP x And all the parameter records of the marked parts, retrieve T end to T now The vibration value during this period is recorded by creating a vibration line graph for each parameter, with time as the horizontal axis and vibration value as the vertical axis. Set sampling point e, evenly set e time points in the horizontal axis direction of each vibration line graph, and analyze the vibration value at each time point.

[0086] S303: Establish a training set for each annotated component, and use the vibration value of the annotated component at the same time point as the independent variable, as well as the component CAP x The vibration value is used as the dependent variable, the independent variable and the dependent variable are packaged as samples, and these e samples are put into the training set of the corresponding labeled parts.

[0087] S304, establish a linear regression model, substitute the independent variable of each sample in the training set as the input value into the linear regression model expression, calculate the difference between the output result of each sample and the dependent variable, sum the differences of all samples in the training set as the gap coefficient, and minimize the gap coefficient by adjusting the intercept F and the regression coefficient β in the expression, and obtain the relationship expression corresponding to each training set respectively:

[0088] C=F+β×W

[0089] In the formula, C is the dependent variable and W is the independent variable.

[0090] The gap coefficient indicates the accuracy of the relationship expression. The smaller the gap coefficient, the more accurately the corresponding relationship expression can describe the vibration influence relationship between the component and its annotated component.

[0091] S305, obtain the current vibration value of each marked component, substitute it into the corresponding relationship expression to obtain the calculation result, and analyze each calculation result and the component CAP x The difference between the current vibration values. Set the error coefficient r, count the number of calculation results h corresponding to the difference greater than r, and use h as the component CAP x The number of exceptions.

[0092] S306, analyze the marked parts corresponding to each component and calculate the number of abnormalities, set the abnormal threshold p, and treat the parts with abnormal times greater than p as abnormal parts. When there are abnormal parts in the equipment, it automatically enters the alarm state, generates alarm information according to the current vibration value of the abnormal parts and transmits it to the monitoring center.

[0093] In S400, the monitoring center displays the alarm information uploaded by the devices in alarm status in real time through a large visual screen, and notifies the management personnel of the occurrence of abnormal events.

[0094] See also Figure 2 The present invention provides an artificial intelligence-based smart park equipment operation monitoring system, including a data acquisition module, an intelligent analysis module, an operation monitoring module and a visualization module.

[0095] The data acquisition module is used to collect the parameter comparison table and maintenance log of the equipment, and collect the parameter records of each component under the equipment in real time through sensors.

[0096] The intelligent analysis module is used to analyze the maintenance log and the parameter records of each component, as well as the influence relationship between the components, establish a relationship network according to the influence relationship between the components, and set the abnormal network.

[0097] The operation monitoring module establishes a relational expression based on the relationship between components in the abnormal network, and analyzes the abnormal number of each component through the relational expression, so as to identify the abnormal components and issue an alarm.

[0098] The visualization module displays alarm information through a large visualization screen.

[0099] The data collection module includes an operation record collection unit, a maintenance record collection unit and a table information collection unit.

[0100] The operation record collection unit is used to collect parameter records of each component. The parameter records include the vibration values ​​of the components at different times, which are collected in real time by the vibration sensors installed on the components.

[0101] The maintenance record collection unit is used to collect the maintenance log of the equipment. The maintenance log refers to the work order record of each maintenance of the equipment. Each work order record includes parts and maintenance time period.

[0102] The table information collection unit is used to collect a parameter comparison table, which includes the vibration thresholds of various components under the equipment.

[0103] The intelligent analysis module includes an impact analysis unit and a network construction unit.

[0104] The impact analysis unit is used to analyze the impact relationship between components.

[0105] First, select the component CAP n All work order records are created, and the maintenance time periods in these work order records are put into the time set. n In the parameter records of other parts, mark the maintenance time periods within the time set.

[0106] Secondly, set the sampling time time, and calculate the average vibration value V in each parameter record within the time length before each maintenance time period. before , and the average vibration value V in the subsequent time period after ; Mark V in the same time period after Greater than V before maintenance time period.

[0107] Finally, set the error times e, and record the corresponding parts CAP of the parameter with the maintenance time period times greater than e m Considered to be subject to the CAP of the component n Impact of component CAP n As an active influencer, component CAP m As a passive influencer, CAP for components n CAP and parts m Build influential relationships.

[0108] The network building unit is used to build a relationship network.

[0109] First, analyze whether each component has an impact relationship with other components, and classify the components that have an impact relationship with each other into the same category. There is no impact relationship between any component in different categories and any component in other categories.

[0110] Then, all the components in each category are connected according to the influence relationship and a relationship network is established to obtain the current vibration values ​​of all parts and the vibration thresholds of all components in the parameter comparison table. The relationship network to which the components with vibration values ​​greater than the corresponding vibration threshold belong is regarded as an abnormal network.

[0111] The operation monitoring module includes an operation analysis unit and an abnormal alarm unit.

[0112] The operation analysis unit is used to establish a relational expression based on the relationship between components in the abnormal network.

[0113] First, mark all the abnormal networks with component CAP x For parts that have an impact relationship and are the active influencing party, obtain the CAP of the parts in the maintenance log x The maintenance time period td recorded in the most recent work order rec , and the current time T now and maintenance time period td re The end time of c is T end .

[0114] Second, obtain the parts CAP x And all the parameter records of the marked parts, retrieve T end to T now The vibration value during this period is recorded by creating a vibration line graph for each parameter, with time as the horizontal axis and vibration value as the vertical axis. Set sampling point e, evenly set e time points in the horizontal axis direction of each vibration line graph, and analyze the vibration value at each time point.

[0115] Then, a training set is established for each annotated component, and the vibration value of the annotated component at the same time point is used as the independent variable, as well as the component CAP x The vibration value is used as the dependent variable, the independent variable and the dependent variable are packaged as samples, and these e samples are put into the training set of the corresponding labeled parts.

[0116] Finally, a linear regression model is established, and the independent variables of each sample in the training set are substituted into the linear regression model expression as input values. The difference between the output result of each sample and the dependent variable is calculated, and the difference of all samples in the training set is summed as the gap coefficient. The gap coefficient is minimized by adjusting the intercept F and the regression coefficient β in the expression, and the corresponding relationship expression for each training set is obtained: C=F+β×W, where C is the dependent variable and W is the independent variable.

[0117] The abnormal alarm unit is used to identify abnormal parts and issue an alarm.

[0118] First, obtain the current vibration value of each marked component, substitute it into the corresponding relationship expression to obtain the calculation result, and analyze each calculation result and the component CAP x The difference between the current vibration values.

[0119] Secondly, set the error coefficient r, count the number of calculation results h corresponding to the difference greater than r, and use h as the component CAP xAnalyze the anomaly times of each component and calculate the anomaly times.

[0120] Finally, an abnormal threshold p is set, and parts with abnormal times greater than p are regarded as abnormal parts. When there are abnormal parts in the equipment, it will automatically enter the alarm state, generate alarm information based on the vibration value of the abnormal parts and transmit it to the monitoring center.

[0121] The visualization module displays the alarm information uploaded by the alarm status equipment in real time through the visualization large screen of the monitoring center.

[0122] Embodiment 1:

[0123] Assume that there are two samples A1 and A2 in the training set Z, their independent variables are 2 and 3, and their dependent variables are 5 and 6, respectively. The initial relationship expression is C = F + β × W;

[0124] Substitute the A1 independent variable and dependent variable into the formula, and adjust the intercept F and regression coefficient β to obtain a further relationship expression:

[0125] C=1+2×W

[0126] Substitute the A2 independent variable into the above formula and we get:

[0127] 1+2×3=7

[0128] Then the gap coefficient is 1. By adjusting the intercept F and regression coefficient β in the expression to minimize the gap coefficient to 0, the final relationship expression is obtained:

[0129] C=3+1×W

[0130] Then the relational expression corresponding to the training set Z is C=3+1×W.

[0131] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0132] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for monitoring the operation of smart park equipment based on artificial intelligence, characterized in that: The method comprises the following steps: S100, collect parameter comparison tables and maintenance logs of equipment in the park, and collect parameter records of various components of the equipment in real time through sensors; S200, comprehensively analyzing the maintenance log and the parameter records of each component, as well as the influence relationship between the components, establishing a relationship network according to the influence relationship between the components and setting an abnormal network; S300, analyzing the relationship between the components in the abnormal network and establishing a relational expression, analyzing the number of abnormalities of each component according to the relational expression, thereby identifying the abnormal components and issuing an alarm; S400, displays alarm devices through a large visual screen and notifies management personnel of abnormal events.

2. According to the method for monitoring the operation of smart park equipment based on artificial intelligence in claim 1, it is characterized by: In S100, the parameter comparison table includes the vibration thresholds of each component of the equipment; the maintenance log refers to the work order record of each maintenance of the equipment, and each work order record includes the component and the maintenance time period; the parameter record includes the vibration values ​​of the components at different times, which are collected in real time by the vibration sensors installed on the components.

3. According to the method for monitoring the operation of smart park equipment based on artificial intelligence in claim 2, it is characterized by: In S200, the specific steps are as follows: S201, filter out the CAP parts in the maintenance log n All work order records are collected, and a time set is established. The maintenance time periods in these work order records are put into the time set. n Parameter records of other parts and components, marking each maintenance time period within the time set in these parameter records; S202, set the sampling time time, and calculate the average vibration value V in each parameter record within the time time before each maintenance time period. before , and the average vibration value V in the subsequent time period after ; Mark V in the same time period after Greater than V before maintenance time period, set the error number e, and record the corresponding component CAP when the number of maintenance time periods is greater than e m Considered to be subject to component CAP n Impact of CAP on components n As an active influencer, component CAP m As a passive influencer, CAP for components n CPA with parts m Build influence relationships; S203, analyzing whether each component has an influence relationship with other components, and classifying the components that have an influence relationship with each other into the same category, and there is no influence relationship between any component in different categories and any component in other categories; connecting all components in each category according to the influence relationship and establishing a relationship network, obtaining the current vibration values ​​of all parts and the vibration thresholds of all components in the parameter comparison table, and treating the relationship network to which the components whose vibration values ​​are greater than the corresponding vibration thresholds belong as abnormal networks.

4. According to the method for monitoring the operation of smart park equipment based on artificial intelligence in claim 3, it is characterized by: In S300, the specific steps are as follows: S301. Analyze the impact relationship between components in the abnormal network and mark all components CAP x Parts that have an impact relationship and are the active influencing party; obtain the CAP of the parts in the maintenance log x The maintenance time period td recorded in the most recent work order rec , and the current time T now and maintenance time period td rec End time T end ; S302, obtain component CAP x And all the parameter records of the marked parts, retrieve T end to T now The vibration value during this period is recorded by establishing a vibration line graph for each parameter with time as the horizontal axis and the vibration value as the vertical axis; setting a sampling point e, evenly setting e time points in the horizontal axis direction of each vibration line graph, and analyzing the vibration value at each time point; S303: Establish a training set for each annotated component, and use the vibration value of the annotated component at the same time point as the independent variable, as well as the component CAP x The vibration value is used as the dependent variable, and the independent variable and the dependent variable are packaged as samples, and these e samples are put into the training set of the corresponding annotated parts; S304, establish a linear regression model, substitute the independent variable of each sample in the training set as the input value into the linear regression model expression, calculate the difference between the output result of each sample and the dependent variable, sum the differences of all samples in the training set as the gap coefficient, and minimize the gap coefficient by adjusting the intercept F and the regression coefficient β in the expression, and obtain the relationship expression corresponding to each training set respectively: C=F+β×W In the formula, C is the dependent variable and W is the independent variable; S305, obtain the current vibration value of each marked component, substitute it into the corresponding relationship expression to obtain the calculation result, and analyze each calculation result and the component CAP x The difference between the current vibration values; set the error coefficient r, count the number of calculation results h corresponding to the difference greater than r, and use h as the component CAP x The number of abnormalities; S306. Analyze the labeled parts corresponding to each component and calculate the number of abnormalities, set the abnormal threshold p, and regard the parts with abnormal times greater than p as abnormal parts; when there are abnormal parts in the equipment, it automatically enters the alarm state, generates alarm information according to the current vibration value of the abnormal parts and transmits it to the monitoring center.

5. According to the method for monitoring the operation of smart park equipment based on artificial intelligence according to claim 4, it is characterized in that: In S400, the monitoring center displays the alarm information uploaded by the devices in alarm status in real time through a large visual screen, and notifies the management personnel of the occurrence of abnormal events.

6. An artificial intelligence-based smart park equipment operation monitoring system, characterized by: The system includes data acquisition module, intelligent analysis module, operation monitoring module and visualization module; The data acquisition module is used to collect the parameter comparison table and maintenance log of the equipment, and collect the parameter records of each component of the equipment in real time through sensors; the intelligent analysis module is used to analyze the maintenance log and the parameter records of each component, as well as the influence relationship between the components, establish a relationship network according to the influence relationship between the components, and set the abnormal network; The operation monitoring module establishes a relational expression based on the relationship between components in the abnormal network, and analyzes the number of abnormalities of each component through the relational expression, so as to identify the abnormal components and issue an alarm; the visualization module displays the alarm information through a large visualization screen.

7. The artificial intelligence-based smart park equipment operation monitoring system according to claim 6 is characterized by: The data collection module includes an operation record collection unit, a maintenance record collection unit and a table information collection unit; The operation record collection unit is used to collect parameter records of each component. The parameter records include the vibration values ​​of the components at different times, which are collected in real time by the vibration sensors installed on the components. The maintenance record collection unit is used to collect the maintenance log of the equipment. The maintenance log refers to the work order record of each maintenance of the equipment. Each work order record includes parts and maintenance time period; The table information collection unit is used to collect a parameter comparison table, which includes the vibration thresholds of various components under the equipment.

8. The artificial intelligence-based smart park equipment operation monitoring system according to claim 7 is characterized by: The intelligent analysis module includes an impact analysis unit and a network construction unit; The impact analysis unit is used to analyze the impact relationship between components; First, select the component CAP n All work order records are created, and the maintenance time periods in these work order records are put into the time set. n In the parameter records of other parts, mark each maintenance time period within the time set; Secondly, set the sampling time time, and calculate the average vibration value V in each parameter record within the time length before each maintenance time period. before , and the average vibration value V in the subsequent time period after ; Mark V in the same time period after Greater than V before Maintenance time period; Finally, set the error times e, and record the corresponding parts CAP of the parameter with the maintenance time period times greater than e m Considered to be subject to component CAP n Impact of CAP on components n As an active influencer, component CAP m As a passive influencer, CAP for components n CAP and parts m Build influence relationships; The network building unit is used to build a relationship network; First, analyze whether each component has an impact relationship with other components, and classify the components that have an impact relationship with each other into the same category. There is no impact relationship between any component in different categories and any component in other categories. Then, all the components in each category are connected according to the influence relationship and a relationship network is established to obtain the current vibration values ​​of all parts and the vibration thresholds of all components in the parameter comparison table. The relationship network to which the components with vibration values ​​greater than the corresponding vibration threshold belong is regarded as an abnormal network.

9. The artificial intelligence-based smart park equipment operation monitoring system according to claim 8 is characterized by: The operation monitoring module includes an operation analysis unit and an abnormal alarm unit; The operation analysis unit is used to establish a relational expression according to the relationship between the components in the abnormal network; First, mark all the abnormal networks with component CAP x For parts that have an impact relationship and are the active influencing party, obtain the CAP of the parts in the maintenance log x The maintenance time period td recorded in the most recent work order rec , and the current time T now and maintenance time period td rec End time T end ; Second, obtain the parts CAP x And all the parameter records of the marked parts, retrieve T end to T now The vibration value during this period is recorded by establishing a vibration line graph for each parameter with time as the horizontal axis and the vibration value as the vertical axis; setting a sampling point e, evenly setting e time points in the horizontal axis direction of each vibration line graph, and analyzing the vibration value at each time point; Then, a training set is established for each annotated component, and the vibration value of the annotated component at the same time point is used as the independent variable, as well as the component CAP x The vibration value is used as the dependent variable, and the independent variable and the dependent variable are packaged as samples, and these e samples are put into the training set of the corresponding annotated parts; Finally, a linear regression model is established. The independent variables of each sample in the training set are substituted into the linear regression model expression as input values. The difference between the output result of each sample and the dependent variable is calculated. The difference of all samples in the training set is summed as the gap coefficient. The gap coefficient is minimized by adjusting the intercept F and the regression coefficient β in the expression. The corresponding relationship expression for each training set is obtained: C = F + β × W, where C is the dependent variable and W is the independent variable. The abnormal alarm unit is used to identify abnormal parts and issue an alarm; First, obtain the current vibration value of each marked component, substitute it into the corresponding relationship expression to obtain the calculation result, and analyze each calculation result and the component CAP x The difference between the current vibration values; Secondly, set the error coefficient r, count the number of calculation results h corresponding to the difference greater than r, and use h as the component CAP x Analyze the anomaly number of each component and calculate the anomaly number; Finally, an abnormal threshold p is set, and components with abnormal times greater than p are regarded as abnormal components; when there are abnormal components in the equipment, it automatically enters the alarm state, generates alarm information based on the vibration value of the abnormal components and transmits it to the monitoring center.

10. The artificial intelligence-based smart park equipment operation monitoring system according to claim 9 is characterized in that: The visualization module displays the alarm information uploaded by the alarm status equipment in real time through the visualization large screen of the monitoring center.

Images