Wind generating set fault monitoring system based on Internet of Things

Abstract

The invention discloses a wind generating set fault monitoring system based on the Internet of Things, relates to the technical field of fault monitoring, and solves the technical problems that in the aspect of fault reason analysis, deep mining and correlation analysis are lacked, and historical data cannot be effectively utilized to improve the monitoring accuracy and reliability. Mechanical and electrical data are comprehensively collected through the operation parameter acquisition unit, a rich information source is provided for accurate analysis, the operation state recognition unit performs preliminary judgment by combining real-time parameters and preset normal parameters, normal and abnormal states can be quickly distinguished, the normal state analysis unit establishes a risk evaluation standard by using historical data, and the risk evaluation standard is analyzed. According to the fault early warning system, normal operation parameters are continuously monitored and subjected to risk prediction, potential risks can be found in advance, preventive maintenance measures can be taken, a state abnormity analysis unit deeply analyzes fault reasons and related parameters, risk prediction is carried out in combination with historical data, and the accuracy and timeliness of fault early warning are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fault monitoring, and in particular to a wind turbine generator set fault monitoring system based on the Internet of Things. Background Art

[0002] As the global demand for clean energy continues to grow, wind power generation has been widely used as a sustainable energy production method, and the scale and number of wind turbines continue to increase. Wind turbines operate in a complex natural environment and face a variety of harsh conditions such as strong winds, dust, and lightning.

[0003] According to the publication number CN102235941A, a wind turbine generator set status monitoring and fault diagnosis system is disclosed, which sets an equipment identification unit on the wind turbine generator set to be monitored, and uses a portable inspection module to collect equipment operating parameter information and equipment identification information of the wind turbine generator set at the equipment site, and transmits the information to the server through the network. The monitoring and fault diagnosis system software installed in the server monitors the operating status of each wind turbine generator set according to the collected data information and pre-stored judgment conditions. If the operating data of the wind turbine generator set exceeds the judgment conditions, a fault alarm message is output.

[0004] However, traditional fault monitoring methods mostly rely on fixed thresholds and manual experience judgment. The analysis of operating parameters is relatively simple, making it difficult to accurately identify potential faults and provide early warnings. At the same time, in terms of fault cause analysis, there is a lack of in-depth mining and correlation analysis, and historical data cannot be effectively used to improve monitoring accuracy and reliability. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a wind turbine fault monitoring system based on the Internet of Things, which solves the problems of lack of in-depth mining and correlation analysis in fault cause analysis and inability to effectively utilize historical data to improve monitoring accuracy and reliability.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: A wind turbine generator fault monitoring system based on the Internet of Things, comprising:

[0007] An operation status identification unit is used to identify the operation status of the wind turbine generator set according to the operation parameters transmitted by the operation parameter acquisition unit, generate a normal state signal and an abnormal state signal, further transmit the generated abnormal state signal to the abnormal state analysis unit, and transmit the normal state signal to the normal state analysis unit;

[0008] A normal state analysis unit, which is used to analyze the acquired normal state signal, continuously monitor the operating parameters, and predict and analyze the overall operating risk, generate corresponding fault monitoring information, and transmit the fault monitoring information to the fault monitoring output unit;

[0009] A state abnormality analysis unit, which is used to analyze the acquired state abnormality signal, determine the cause of the fault according to the abnormal parameters, determine the corresponding associated parameters based on the cause of the fault, and perform risk prediction analysis according to the associated parameters to generate reduction warning information and secondary analysis signals, and then transmit the secondary analysis signals to the secondary identification analysis unit;

[0010] The secondary identification and analysis unit is used to process the acquired secondary analysis signal, make corresponding risk predictions by calculating the fault risk values ​​corresponding to the same type of fault causes, generate risk prediction information, and transmit the risk prediction information to the fault monitoring output unit.

[0011] As a further solution of the present invention: it also includes an operation parameter acquisition unit and a fault monitoring output unit;

[0012] An operating parameter acquisition unit, used to acquire the operating parameters of the wind turbine generator set and transmit the acquired operating parameters to the operating state identification unit;

[0013] The fault monitoring output unit is used to display the acquired fault monitoring information and risk prediction information to the corresponding operator.

[0014] As a further solution of the present invention: the specific manner in which the operating status identification unit identifies the operating status of the wind turbine generator set is:

[0015] Obtain the wind turbine operating parameters and the corresponding normal parameters and compare them. If the two are the same, a normal state signal is generated; if they are different, an abnormal state signal is generated;

[0016] The generated abnormal state signal is transmitted to the abnormal state analysis unit, and the normal state signal is transmitted to the normal state analysis unit.

[0017] As a further solution of the present invention: the specific manner in which the normal state analysis unit analyzes the normal state signal is:

[0018] Obtain abnormal records and their corresponding abnormal data in historical data, select one of them as analysis data and obtain related data, generate risk assessment criteria based on the failure risks of the two, classify abnormal records of the same type and generate corresponding criteria;

[0019] Obtain the changes in operating parameters within the time period t and compare them with the risk assessment criteria. If they match, a risk result is generated; if they do not match, a normal monitoring signal is generated.

[0020] As a further solution of the present invention: the specific manner in which the abnormal state analysis unit analyzes the acquired abnormal state signal is:

[0021] The operating parameters of the abnormal state signal are obtained as abnormal parameters, so as to determine the cause of the fault and obtain the associated parameters, and at the same time, the abnormal parameters and the associated parameters are combined to make risk prediction;

[0022] Obtain the change value of the abnormal parameter within the time period t1, and obtain the corresponding change value of the associated parameter at the same time, then compare the two sets of change values ​​obtained with the corresponding thresholds respectively. If both are less than the threshold, the risk threshold is reached to generate an abnormal signal; if any group is greater than the threshold, the risk threshold is reached to generate a warning signal, and then combine historical data to analyze the abnormal signal and warning signal.

[0023] As a further solution of the present invention: the specific method of analyzing abnormal signals and warning signals by the abnormal state analysis unit in combination with historical data is as follows:

[0024] Get the historical data within time T, and get the historical data with the same cause as the current fault, then get the frequency of the current fault cause in the historical data, record it as P, the mean time between failures G and the mean time to repair M, then substitute the obtained parameters into the formula Calculate the fault value corresponding to the current fault cause, where θ is the preset proportional coefficient and T is the time period;

[0025] The fault value S1 of the historical fault cause in the previous time period is calculated according to the formula, and the fault value S is compared with S1. If S is greater than S1, it indicates that the fault value has increased, and a secondary analysis signal is generated and transmitted to the secondary identification and analysis unit. If S is less than S1, the fault value has decreased, and a reduction warning information is generated and transmitted to the fault monitoring output unit.

[0026] As a further solution of the present invention: the specific manner in which the secondary identification and analysis unit processes the secondary analysis signal is:

[0027] According to historical data, obtain the abnormal parameters corresponding to the fault cause, and obtain the change value corresponding to the abnormal parameters. At the same time, calculate the mean of all change values ​​and record it as K. Then obtain the number of changes in the associated parameters corresponding to the abnormal parameters and record it as C, and obtain the change frequency corresponding to the abnormal parameters and record it as H. Then substitute the obtained parameters into the formula The fault risk value corresponding to the fault cause is calculated, and the fault risk value Z is compared with the preset value Zy.

[0028] As a further solution of the present invention: the specific manner in which the secondary identification and analysis unit compares the risk value Z with the preset value Zy is as follows:

[0029] If the fault risk value Z is greater than the preset value Zy, it means that the current fault cause is risky and an alarm signal is generated; if Z is less than the preset value Zy, it means that the current fault has not reached the risk threshold and a warning signal is generated. The alarm signal and the warning signal are then transmitted to the fault monitoring output unit.

[0030] The present invention provides a wind turbine fault monitoring system based on the Internet of Things. Compared with the prior art, it has the following beneficial effects:

[0031] The present invention comprehensively collects mechanical and electrical data through the operation parameter acquisition unit, providing a rich information source for accurate analysis; the operation status identification unit makes a preliminary judgment based on the real-time parameters and the preset normal parameters, can quickly distinguish between normal and abnormal states, and improve monitoring efficiency; the normal state analysis unit uses historical data to establish risk assessment standards, continuously monitors and predicts risks of normal operating parameters, can discover potential risks in advance, and is helpful to take preventive maintenance measures to reduce the probability of failure; the abnormal state analysis unit deeply analyzes the cause of the failure and related parameters, and makes risk prediction based on historical data; the secondary identification and analysis unit further accurately calculates the failure risk value, and improves the accuracy and timeliness of failure warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a block diagram of the system principle of the present invention;

[0033] Figure 2 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0034] 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.

[0035] For example, see Figure 1 and Figure 2 The present application provides a wind turbine fault monitoring system based on the Internet of Things, including an operation parameter acquisition unit, an operation status identification unit, an abnormal status analysis unit, a normal status analysis unit, a secondary identification analysis unit and a fault monitoring output unit. Figure 1 It can be known that the above functional units are electrically connected in a unidirectional manner.

[0036] An operating parameter acquisition unit is used to acquire the operating parameters of the wind turbine generator set and transmit the acquired operating parameters to the operating status identification unit, wherein the operating parameters include mechanical data and electrical data, and the mechanical data includes: rotation speed, vibration, torque and pitch angle, and the electrical data includes output power, voltage, current and frequency.

[0037] The operating state identification unit is used to identify the operating state of the wind turbine generator set according to the acquired operating parameters, and the specific identification method is as follows:

[0038] The operating parameters are obtained, and the operating parameters here are real-time data. At the same time, the normal parameters corresponding to the wind turbine are obtained, and the normal parameter preset values ​​here are set by the operator, and then the operating parameters are compared with the normal parameters; it is assumed that the normal parameters of the wind turbine specify that the generator output voltage should be within a specific range, such as 380V±5%.

[0039] If the operating parameters are the same as the normal parameters, it means that the current wind turbine generator set is in normal operating state, and a normal state signal is generated. Conversely, if the operating parameters are different from the normal parameters, it means that the current wind turbine generator set is in abnormal operating state, and an abnormal state signal is generated.

[0040] When the system obtains the actual output voltage of the generator as 375V in real time, it compares it with the normal parameters and finds it is within the specified range. It can be judged that the current wind turbine generator set is operating normally and generates a normal status signal. If the output voltage obtained is 360V, which is different from the normal parameters, it means that the current wind turbine generator set is operating abnormally and generates an abnormal status signal.

[0041] For example, the gearbox oil temperature is set to be lower than a specific temperature value, such as 80°C, in the normal parameters. If the real-time monitoring of the gearbox oil temperature is 75°C, which is consistent with the normal parameters, it means that the operation is normal; if the oil temperature is monitored to be 85°C, which is beyond the normal parameter range, it indicates that the operation is abnormal.

[0042] The generated abnormal state signal is further transmitted to the abnormal state analysis unit, and the normal state signal is transmitted to the normal state analysis unit.

[0043] The normal state analysis unit is used to analyze the acquired normal state signal, continuously monitor the operating parameters, and predict and analyze the overall operating risk to generate corresponding fault monitoring information. The specific fault monitoring information includes risk results and normal monitoring signals. The specific method of generating fault monitoring information is as follows:

[0044] Obtain historical data, obtain abnormal records in the historical data, and the abnormal records here represent records of corresponding faults of wind turbines, and at the same time obtain abnormal data corresponding to the abnormal records, obtain any one of the abnormal data as analysis data, and then obtain the associated data of the analysis data, and the associated data represents a corresponding change relationship with the abnormal data. For example, if the analysis data is voltage data, it is found through analysis that when the voltage changes, the corresponding temperature will also change. Therefore, the temperature data is marked as associated data, and a corresponding risk assessment standard is generated according to the fault risks corresponding to the analysis data and the associated data. Similarly, the abnormal records of the same type are classified, and the corresponding risk assessment standard is generated;

[0045] In a specific example, suppose that after analyzing multiple sets of similar data, it is found that when the current is 20% higher than the normal range, the bearing temperature will rise by 15°C within 10 minutes. If the temperature rises by more than 20°C and continues to rise, the probability of a serious failure of the unit (such as bearing burnout) increases significantly. Based on this, a risk assessment standard can be set: when the current increases and the bearing temperature rises within a certain range (such as 10-20°C), the risk level is medium; when the temperature rise exceeds 20°C, the risk level is high.

[0046] The operating parameters are compared with the risk assessment criteria, and the specific comparison method is: the changes in the operating parameters within the time period t are obtained, and the specific values ​​of the time period t are set by the operator, and the changes in the operating parameters are matched with the risk assessment criteria. If the two can match, the corresponding risk results are generated, and here the two can match, it means that there are corresponding abnormal data, further indicating the existence of risk situations, if the two cannot match, a normal monitoring signal is generated, and the generated risk results and normal monitoring signals are further transmitted to the fault monitoring output unit.

[0047] For example, the time period t is set to 1 hour. The risk assessment standard stipulates that if the generator speed continues to rise within 1 hour and the increase exceeds 30%, there is a risk of equipment overload. During the monitoring process, it is obtained that the generator speed has increased from 1200 revolutions per minute to 1600 revolutions per minute within this hour, with an increase of 33.3%. At this time, the change in operating parameters matches the risk assessment standard, and the result of equipment overload risk will be generated.

[0048] Fault monitoring output unit, which is used to display the acquired risk results and normal monitoring signals to the corresponding operators.

[0049] Embodiment 2, as Embodiment 2 of the present invention, is implemented on the basis of Embodiment 1, and differs from Embodiment 1 in that:

[0050] The state abnormality analysis unit is used to analyze the acquired state abnormality signal, determine the cause of the fault according to the abnormal parameters, determine the corresponding associated parameters based on the cause of the fault, and perform risk prediction analysis according to the associated parameters. The specific risk prediction analysis method is as follows:

[0051] The operating parameters corresponding to the abnormal state signal are obtained and recorded as abnormal parameters. At the same time, the corresponding fault cause is determined according to the abnormal parameters. Then, the associated parameters of the abnormal parameters are obtained according to the fault cause, and the risk prediction is performed in combination with the abnormal parameters and the associated parameters. The specific risk prediction method is as follows:

[0052] Obtain the change value of the abnormal parameter within the time period t1, and obtain the change value corresponding to the associated parameter at the same time, and then compare the two sets of change values ​​obtained with the corresponding thresholds respectively. For example, if the abnormal parameter is voltage data, then compare the voltage change value with the corresponding voltage threshold. At the same time, the associated parameter of the voltage data is temperature data, and further compare the temperature data with the temperature threshold. If both sets of change values ​​are less than the corresponding threshold, it means that the risk threshold is reached, and an abnormal signal is generated at the same time. On the contrary, if any of the two sets of change values ​​is greater than the threshold, it means that the risk threshold is reached, and an early warning signal is generated. Further analysis is performed on the generated abnormal signal and early warning signal in combination with historical data;

[0053] Get the historical data within time T, and get the historical data with the same cause as the current fault, then get the frequency of the current fault cause in the historical data and record it as P, the average time between failures G, and the average time between failures refers to the average working time of a repairable product between two adjacent failures. The calculation formula is: MTBF = total operating time / number of failures. For example, the total operating time of a wind turbine generator set in a period of time is 10,000 hours, and 5 failures occurred during this period. Then the MTBF of the wind turbine generator set is 10,000 / 5 = 2,000 hours, and the average repair time M. The average repair time refers to the average time required to repair a repairable device after a failure occurs. The calculation formula is: MTTR = total fault repair time / number of failures. Then substitute the obtained parameters into the formula Calculate the fault value corresponding to the current fault cause, where θ is the preset proportional coefficient and T is the time period;

[0054] Then, the fault value S1 of the historical fault cause corresponding to the previous time period is calculated according to the above formula, and the fault value S is compared with the fault value S1. If the fault value S is greater than the fault value S1, it means that the fault value is increasing, and a secondary analysis signal is generated, and the secondary analysis signal is transmitted to the secondary identification and analysis unit. If the fault value S is less than the fault value S1, it means that the fault value is decreasing, and a reduction warning information is generated, and the reduction warning information is transmitted to the fault monitoring output unit.

[0055] The secondary identification and analysis unit is used to process the acquired secondary analysis signal, perform corresponding risk prediction by calculating the fault risk value corresponding to the same type of fault cause, and generate risk prediction information, wherein the risk prediction information includes an alarm signal and a warning signal, and the specific calculation method of the fault risk value is:

[0056] According to the historical data, obtain the abnormal parameters corresponding to the fault cause, and obtain the change values ​​corresponding to the abnormal parameters, and calculate the mean of all change values, record it as K, then obtain the number of changes in the associated parameters corresponding to the abnormal parameters, record it as C, and the number of changes in the associated parameters represents the difference between the associated parameters corresponding to the current fault cause and the number of associated parameters in the historical data, and the number in the historical data is the mean, and obtain the change frequency corresponding to the abnormal parameters, record it as H, and the change frequency here is the mean of all change frequencies, which is specifically calculated according to the formula change frequency = change value / change time, and then substitute the obtained parameters into the formula The fault risk value corresponding to the fault cause is calculated, and the fault risk value Z is compared with a preset value Zy, and the specific value of the preset value Zy is set by the operator;

[0057] When the fault risk value Z is greater than the preset value Zy, it means that there is a risk in the current fault cause and an alarm signal is generated. Conversely, when the fault risk value Z is less than the preset value Zy, it means that the current fault cause has not reached the risk threshold and a warning signal is generated. The generated alarm signal and warning signal are then transmitted to the fault monitoring output unit.

[0058] A fault monitoring output unit is used to display the acquired reduction warning information, alarm signal and warning signal to the corresponding operator.

[0059] Embodiment 3, as the embodiment 3 of the present invention, focuses on combining the implementation processes of embodiment 1 and embodiment 2 for implementation.

[0060] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0061] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A wind turbine fault monitoring system based on the Internet of Things, characterized in that: include: An operation status identification unit is used to identify the operation status of the wind turbine generator set according to the operation parameters transmitted by the operation parameter acquisition unit, generate a normal state signal and an abnormal state signal, further transmit the generated abnormal state signal to the abnormal state analysis unit, and transmit the normal state signal to the normal state analysis unit; A normal state analysis unit, which is used to analyze the acquired normal state signal, continuously monitor the operating parameters, and predict and analyze the overall operating risk, generate corresponding fault monitoring information, and transmit the fault monitoring information to the fault monitoring output unit; A state abnormality analysis unit, which is used to analyze the acquired state abnormality signal, determine the cause of the fault according to the abnormal parameters, determine the corresponding associated parameters based on the cause of the fault, and perform risk prediction analysis according to the associated parameters to generate reduction warning information and secondary analysis signals, and then transmit the secondary analysis signals to the secondary identification analysis unit; The secondary identification and analysis unit is used to process the acquired secondary analysis signal, make corresponding risk predictions by calculating the fault risk values ​​corresponding to the same type of fault causes, generate risk prediction information, and transmit the risk prediction information to the fault monitoring output unit.

2. The wind turbine generator fault monitoring system based on the Internet of Things according to claim 1 is characterized in that: It also includes an operating parameter acquisition unit and a fault monitoring output unit; An operating parameter acquisition unit, used to acquire the operating parameters of the wind turbine generator set and transmit the acquired operating parameters to the operating state identification unit; The fault monitoring output unit is used to display the acquired fault monitoring information and risk prediction information to the corresponding operator.

3. The wind turbine generator fault monitoring system based on the Internet of Things according to claim 1 is characterized in that: The specific method for the operating state identification unit to identify the operating state of the wind turbine generator set is: Obtain the wind turbine operating parameters and the corresponding normal parameters and compare them. If the two are the same, a normal state signal is generated; if they are different, an abnormal state signal is generated; The generated abnormal state signal is transmitted to the abnormal state analysis unit, and the normal state signal is transmitted to the normal state analysis unit.

4. The wind turbine generator fault monitoring system based on the Internet of Things according to claim 1 is characterized in that: The specific manner in which the normal state analysis unit analyzes the normal state signal is: Obtain abnormal records and their corresponding abnormal data in historical data, select one of them as analysis data and obtain related data, generate risk assessment criteria based on the failure risks of the two, classify abnormal records of the same type and generate corresponding criteria; Obtain the changes in operating parameters within the time period t and compare them with the risk assessment criteria. If they match, a risk result is generated; if they do not match, a normal monitoring signal is generated.

5. The wind turbine generator fault monitoring system based on the Internet of Things according to claim 1 is characterized in that: The specific method in which the abnormal state analysis unit analyzes the acquired abnormal state signal is as follows: The operating parameters of the abnormal state signal are obtained as abnormal parameters, so as to determine the cause of the fault and obtain the associated parameters, and at the same time, the abnormal parameters and the associated parameters are combined to make risk prediction; Obtain the change value of the abnormal parameter within the time period t1, and obtain the corresponding change value of the associated parameter at the same time, then compare the two sets of change values ​​obtained with the corresponding thresholds respectively. If both are less than the threshold, the risk threshold is reached to generate an abnormal signal; if any group is greater than the threshold, the risk threshold is reached to generate a warning signal, and then combine historical data to analyze the abnormal signal and warning signal.

6. The wind turbine generator fault monitoring system based on the Internet of Things according to claim 5 is characterized in that: The specific method of analyzing abnormal signals and warning signals by the state abnormality analysis unit in combination with historical data is as follows: Get the historical data within time T, and get the historical data with the same cause as the current fault, then get the frequency of the current fault cause in the historical data, record it as P, the mean time between failures G and the mean time to repair M, then substitute the obtained parameters into the formula Calculate the fault value corresponding to the current fault cause, where θ is the preset proportional coefficient and T is the time period; The fault value S1 of the historical fault cause in the previous time period is calculated according to the formula, and the fault value S is compared with S1. If S is greater than S1, it indicates that the fault value has increased, and a secondary analysis signal is generated and transmitted to the secondary identification and analysis unit. If S is less than S1, the fault value has decreased, and a reduction warning information is generated and transmitted to the fault monitoring output unit.

7. The wind turbine generator fault monitoring system based on the Internet of Things according to claim 1, characterized in that: The specific method in which the secondary identification and analysis unit processes the secondary analysis signal is as follows: According to historical data, obtain the abnormal parameters corresponding to the fault cause, and obtain the change value corresponding to the abnormal parameters. At the same time, calculate the mean of all change values ​​and record it as K. Then obtain the number of changes in the associated parameters corresponding to the abnormal parameters and record it as C, and obtain the change frequency corresponding to the abnormal parameters and record it as H. Then substitute the obtained parameters into the formula The fault risk value corresponding to the fault cause is calculated, and the fault risk value Z is compared with the preset value Zy.

8. The wind turbine generator fault monitoring system based on the Internet of Things according to claim 7 is characterized in that: The specific manner in which the secondary identification and analysis unit compares the risk value Z with the preset value Zy is as follows: If the fault risk value Z is greater than the preset value Zy, it means that the current fault cause is risky and an alarm signal is generated; if Z is less than the preset value Zy, it means that the current fault has not reached the risk threshold and a warning signal is generated. The alarm signal and the warning signal are then transmitted to the fault monitoring output unit.

Images