Water turbine operation monitoring system and method

Abstract

The invention discloses a water turbine operation monitoring system and method in the technical field of hydroelectric generation. The method comprises the steps that real-time operation data and real-time output data of a water turbine input externally are received; inputting real-time operation data of the water turbine into the water turbine operation analysis model, and calculating a prediction value of output data; comparing the output data predicted value with the real-time output data, and judging whether a data difference value between the two exceeds an abnormal threshold value or not; when the data difference value does not exceed the abnormal threshold value, it is judged that the water turbine runs abnormally; and when the data difference value exceeds the abnormal threshold value in response, judging that the water turbine operates abnormally, and reporting an abnormal condition to an operation management platform at the same time. According to the method for monitoring the operation of the water turbine, the operation state of the water turbine is monitored and evaluated, the abnormal operation condition of the water turbine is reported in time, operation and maintenance management personnel can conveniently conduct maintenance in time, and additional economic losses are avoided.

Description

Technical Field

[0001] The invention relates to a water turbine operation monitoring system and method, belonging to the technical field of hydropower generation. Background Art

[0002] A water turbine is a power machine that converts the energy of water flow into rotational mechanical energy. It belongs to the category of turbine machinery in fluid machinery. Most modern water turbines are installed in hydropower stations to drive generators to generate electricity. In a hydropower station, water from an upstream reservoir is led to the water turbine through a water diversion pipe, which drives the turbine runner to rotate and drives the generator to generate electricity. The water that has done work is discharged downstream through the tailwater pipe. The higher the water head and the greater the flow rate, the greater the output power of the water turbine.

[0003] During the operation of the turbine, some abnormal conditions that affect the turbine's operating output are difficult to detect in time through manual detection, and these abnormal conditions often cause the turbine output data to decrease, causing unnecessary economic losses to the hydropower station. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and to provide a turbine operation monitoring system and method.

[0005] To achieve the above object, the present invention is implemented by adopting the following technical solutions:

[0006] In a first aspect, the present invention provides a method for monitoring operation of a hydraulic turbine, comprising:

[0007] Receive real-time operation data and output data of the turbine input from the outside;

[0008] Input the real-time operation data of the turbine into the turbine operation analysis model to calculate the output data prediction value;

[0009] Compare the output data prediction value with the real-time output data to determine whether the data difference between the two exceeds the abnormal threshold;

[0010] In response to the data difference not exceeding the abnormal threshold, it is determined that the turbine operation is normal;

[0011] In response to the data difference exceeding the abnormal threshold, it is determined that the turbine is operating abnormally, and the abnormal situation is reported to the operation management platform.

[0012] Furthermore, the operating data includes working water head, flow rate, rotation speed, vibration swing, pressure pulsation and air gap, and the operating data is detected by sensors installed in the hydropower station.

[0013] Furthermore, the turbine operation analysis model is obtained by training a convolutional neural network model as the underlying model, including:

[0014] Collect historical operation data and output data during the operation of the turbine;

[0015] The historical operation data is divided into a training set and a validation set, and the training set is input into the convolutional neural network model for training to obtain a theoretical model of turbine operation;

[0016] The verification set is used to verify the turbine operation theory model to determine whether the turbine operation theory model has been trained;

[0017] In response to the fact that the training of the theoretical model of the operation of the hydraulic turbine is not completed, the model is continuously trained using historical operation data;

[0018] In response to the completion of the training of the water turbine operation theoretical model, the water turbine operation theoretical model is used as the water turbine operation analysis model.

[0019] Furthermore, the turbine operation theory model is verified using the verification set to determine whether the turbine operation theory model has been trained, including:

[0020] Input the validation set into the turbine operation theory model to calculate the theoretical output data;

[0021] Compare the theoretical output data with the historical output data to determine whether the data difference between the two exceeds a preset threshold;

[0022] In response to the data difference exceeding a preset threshold, determining that the training of the turbine operation theoretical model is incomplete;

[0023] In response to the data difference not exceeding the preset threshold, it is determined that the training of the turbine operation theoretical model is completed.

[0024] Furthermore, the turbine operation analysis model can analyze the change trend of turbine output performance according to the output data prediction value, and report the analysis results to the operation management platform.

[0025] Furthermore, when the turbine operation analysis model determines that the turbine operation is abnormal, the real-time operation data and the historical operation data are compared and analyzed to infer the source of the turbine operation abnormality and obtain the abnormal analysis result, and the abnormal analysis result is reported to the turbine degradation feature knowledge base of the operation management platform.

[0026] Furthermore, the expression of the abnormal analysis result is:

[0027] S=φ*a+β*b+χ*c+δ*d

[0028] Where: S is the anomaly analysis result, φ is the anomaly code at position one, a is the anomaly probability at position one, β is the anomaly code at position two, b is the anomaly probability at position two, χ is the anomaly code at position three, c is the anomaly probability at position three, δ is the anomaly code at position four, d is the anomaly probability at position four, and a>b>c>d.

[0029] In a second aspect, the present invention provides a turbine operation monitoring system, comprising:

[0030] Data receiving module: receiving real-time operation data and output data of the turbine inputted from the outside;

[0031] Output prediction module: input the real-time operation data of the turbine into the turbine operation analysis model to calculate the output data prediction value;

[0032] Data comparison module: compares the output data prediction value with the real-time output data to determine whether the data difference between the two exceeds the abnormal threshold;

[0033] Abnormal judgment module: when the data difference does not exceed the abnormal threshold, it is judged that the turbine operation is normal; when the data difference exceeds the abnormal threshold, it is judged that the turbine operation is abnormal, and the abnormal situation is reported to the operation management platform.

[0034] In a third aspect, the present invention provides a turbine operation monitoring device, including a processor and a storage medium;

[0035] The storage medium is used to store instructions;

[0036] The processor is used to operate according to the instructions to execute the steps of any of the methods described above.

[0037] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of any of the above methods when executed by a processor.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] 1. This scheme uses neural network technology to establish a turbine operation analysis model, realizes the monitoring and analysis of turbine operation conditions, and automatically identifies the operation conditions and verifies the output data during the normal operation of the turbine, simulating the manual monitoring method, and automatically evaluates the operation status, so that the operation and maintenance management personnel can find abnormal conditions in time, carry out operation and maintenance in time, and avoid additional economic losses;

[0040] 2. The present invention uses a convolutional neural network as the underlying logical model. Compared with other types of models, it has good fault tolerance, parallel processing capabilities and self-learning capabilities. It can handle problems with complex environmental information, unclear background knowledge, and unclear reasoning rules. It allows samples to have large defects and distortions, has a fast running speed, good adaptive performance, and has a higher resolution. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings:

[0042] Figure 1 A schematic flow chart of a method for monitoring operation of a water turbine provided in Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0043] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present invention can be combined with each other without conflict.

[0044] The following detailed description is an exemplary description, which is intended to provide further detailed description of the present invention. Unless otherwise specified, all technical terms used in the present invention have the same meaning as those generally understood by those skilled in the art to which the present invention belongs. The terms used in the present invention are only for describing specific embodiments, and are not intended to limit the exemplary embodiments according to the present invention.

[0045] Embodiment 1:

[0046] When modifying traditional turbine monitoring equipment, studying operating characteristics, and evaluating status, it is necessary to conduct surveys, inspections, and calibrations of turbine on-site information collection systems, equipment, and facilities, and form inspection and analysis reports. It is necessary to fully consider factors such as signal anti-interference, transmission attenuation, turbine structure, water flow stability, and sensor characteristics, select sensors at different positions of turbines, arrange signal transmission hardware, and design signal conditioners, and optimize, adjust, and supplement data collection of on-site turbine equipment to improve the reliability of turbine monitoring signals and monitoring systems. It is necessary to combine on-site test procedures and expert experience to identify operating conditions and evaluate the operating status of turbines.

[0047] With the development of neural network technology, more and more industries use neural network technology to replace the original manual methods for improvement and innovation. This solution designs a method for monitoring the operation of turbines, monitors and evaluates the operation status of turbines, and reports abnormal operation of turbines in time, so as to facilitate timely maintenance by operation and maintenance managers and avoid additional economic losses. The details are as follows:

[0048] 1. Use sensors to collect historical operating data and output data during the operation of the turbine. The operating data includes working head, flow, speed, vibration swing, pressure pulsation and air gap, among which:

[0049] 1. Working head. Water flows from high to low, which is the objective law of water flow. When a hydropower station is built on a certain river section and a turbine is installed, water flows from the turbine inlet to the outlet through the turbine. This means that there is an energy difference between the turbine inlet and outlet, and its size can be determined according to the law of water flow energy conversion. The working head of the turbine refers to the difference in the total specific energy of the water flow at the turbine inlet and outlet. The working head of the turbine is an important working parameter of the turbine, and its size indicates the amount of unit energy of the water flow used by the turbine.

[0050] 2. Flow rate. The volume of water flowing through a given flow section of a turbine per unit time is called the flow rate of the turbine, and its unit is m3 / s. The flow rate of the turbine is another important working parameter of the turbine, and its size also indicates how much energy the turbine utilizes.

[0051] 3. Speed. The speed of the turbine is the number of times the turbine wheel rotates per unit time, and the unit of speed is r / min.

[0052] 4. Vibration swing. The vibration swing of a turbine generator refers to the degree of fluctuation of the rotor during operation, which is usually divided into radial swing and axial swing. Excessive vibration swing can cause contact between the rotor and metal parts, causing damage to the equipment and even affecting the normal operation of the hydropower station.

[0053] 5. Pressure pulsation. Pressure pulsation refers to the alternating random changes of water pressure around its average value in the flow channel of the turbine. It is one of the main reasons for the vibration of the turbine generator set. This type of vibration is often random and related to the operating conditions of the turbine.

[0054] Second, the historical operation data is divided into a training set and a validation set, and the convolutional neural network model is selected as the underlying model. The training set and the underlying model are combined for training to obtain a theoretical model of turbine operation. Convolutional Neural Networks (CNN) is a type of feedforward neural network that contains convolution operations and has a deep structure. The overall architecture includes input layer, convolution layer, pooling layer, fully connected layer and output layer. Convolutional neural networks have some advantages that traditional technologies do not have:

[0055] (1) Good fault tolerance, parallel processing and self-learning capabilities, which can handle problems with complex environmental information, unclear background knowledge and unclear reasoning rules;

[0056] (2) It allows samples to have larger defects and distortions, has fast running speed, good adaptive performance, and high resolution;

[0057] (3) It integrates the feature extraction function into the multi-layer perceptron by restructuring the structure and reducing the weights, thus omitting the complex image feature extraction process before recognition.

[0058] 3. Use the previously classified verification set to verify the trained turbine operation theoretical model to determine whether the data difference between the theoretical output data calculated according to the turbine operation theoretical model and the historical output data actually collected exceeds the preset threshold. The preset threshold is set by the operation and maintenance personnel according to the actual situation.

[0059] 4. If the data difference between the theoretical output data calculated by the turbine operation theoretical model and the historical output data actually collected exceeds the preset threshold, the turbine operation theoretical model has not been trained yet, and it is necessary to continue to use the historical operation data to carry out the next round of training, verification and judgment of the turbine operation theoretical model. Here, the turbine operation theoretical model is iteratively trained using historical operation data to reduce the error between the theoretical output data and the historical output data, so that the turbine operation theoretical model is closer to the actual operation of the turbine.

[0060] 5. If the data difference between the theoretical output data calculated by the turbine operation theoretical model and the historical output data actually collected does not exceed the preset threshold, the turbine operation theoretical model is used as the final turbine operation analysis model to facilitate the subsequent steps. The turbine operation analysis model obtained after multiple iterations of training is close to the actual operation of the turbine and can be used to estimate the turbine output data.

[0061] 6. Use sensors to collect real-time operation data and real-time output data of the turbine, combine the real-time operation data with the turbine operation analysis model, predict the output data, and obtain the output data prediction value;

[0062] 7. Analyze the change trend of turbine output performance according to the output data prediction value, and report the analysis results to the operation management platform. The management personnel can predict the subsequent output data changes through the data trend analysis results on the platform, so as to make countermeasures in advance;

[0063] 8. Compare and analyze the output data prediction value with the real-time output data to determine whether the data difference between the output data prediction value and the real-time output data exceeds the abnormal threshold. The abnormal threshold here is summarized by the management personnel based on experience analysis;

[0064] 9. If the data difference between the output data prediction value and the real-time output data does not exceed the abnormal threshold, it is determined that the turbine operation is normal; if the data difference between the output data prediction value and the real-time output data exceeds the abnormal threshold, it is determined that the turbine operation is abnormal. At this time, the real-time operation data and the historical operation data are compared and analyzed to infer the source of the turbine operation abnormality and obtain the abnormal analysis results. At the same time, the abnormal analysis results are reported to the turbine degradation feature knowledge base of the operation management platform, so that the management personnel can check or repair the abnormal location in time. The expression of the abnormal analysis result is:

[0065] S=φ*a+β*b+χ*c+δ*d

[0066] In the formula: S is the abnormality analysis result, φ is the abnormality code of position 1, a is the abnormality probability of position 1, β is the abnormality code of position 2, b is the abnormality probability of position 2, χ is the abnormality code of position 3, c is the abnormality probability of position 3, δ is the abnormality code of position 4, d is the abnormality probability of position 4, and a>b>c>d. Managers can understand each possible abnormal position through the abnormality analysis results, and each abnormal position is arranged from front to back according to the size of the abnormality probability, which is more convenient and intuitive.

[0067] It should be noted that this solution, by identifying anomalies and establishing a comprehensive knowledge base of turbine degradation characteristics, collects and analyzes the monitoring and monitoring information of the turbine's vibration swing, pressure pulsation, air gap, etc. in real time, timely identifies the status of the turbine, finds early signs of failure, and analyzes the cause of the failure, the severity of the failure, and the development trend of the failure. In addition, managers can also use the turbine operation analysis model, combined with simulation calculations, field tests, mechanism analysis and other verification methods, to test and evaluate the accuracy of various evaluation results of the turbine monitoring equipment modification, operation characteristics research and status assessment system, to ensure that the overall accuracy rate is higher than 95%, and provide a verification report.

[0068] Embodiment 2:

[0069] A water turbine operation monitoring system can implement a water turbine operation monitoring method described in Embodiment 1, comprising:

[0070] Data receiving module: receiving real-time operation data and output data of the turbine inputted from the outside;

[0071] Output prediction module: input the real-time operation data of the turbine into the turbine operation analysis model to calculate the output data prediction value;

[0072] Data comparison module: compares the output data prediction value with the real-time output data to determine whether the data difference between the two exceeds the abnormal threshold;

[0073] Abnormal judgment module: when the data difference does not exceed the abnormal threshold, it is judged that the turbine operation is normal; when the data difference exceeds the abnormal threshold, it is judged that the turbine operation is abnormal, and the abnormal situation is reported to the operation management platform.

[0074] Embodiment three:

[0075] The embodiment of the present invention further provides a water turbine operation monitoring device, which can implement a water turbine operation monitoring method described in the first embodiment, including a processor and a storage medium;

[0076] The storage medium is used to store instructions;

[0077] The processor is used to operate according to the instructions to perform the steps of the following method:

[0078] Receive real-time operation data and output data of the turbine input from the outside;

[0079] Input the real-time operation data of the turbine into the turbine operation analysis model to calculate the output data prediction value;

[0080] Compare the output data prediction value with the real-time output data to determine whether the data difference between the two exceeds the abnormal threshold;

[0081] In response to the data difference not exceeding the abnormal threshold, it is determined that the turbine operation is normal;

[0082] In response to the data difference exceeding the abnormal threshold, it is determined that the turbine is operating abnormally, and the abnormal situation is reported to the operation management platform.

[0083] Embodiment 4:

[0084] The embodiment of the present invention further provides a computer-readable storage medium, which can implement the turbine operation monitoring method described in the first embodiment, and stores a computer program thereon, which implements the steps of the following method when the program is executed by a processor:

[0085] Receive real-time operation data and output data of the turbine input from the outside;

[0086] Input the real-time operation data of the turbine into the turbine operation analysis model to calculate the output data prediction value;

[0087] Compare the output data prediction value with the real-time output data to determine whether the data difference between the two exceeds the abnormal threshold;

[0088] In response to the data difference not exceeding the abnormal threshold, it is determined that the turbine operation is normal;

[0089] In response to the data difference exceeding the abnormal threshold, it is determined that the turbine is operating abnormally, and the abnormal situation is reported to the operation management platform.

[0090] It is known from common technical knowledge that the present invention can be implemented by other embodiments that do not deviate from its spirit or essential features. Therefore, the above disclosed embodiments are only illustrative in all respects and are not exclusive. All changes within the scope of the present invention or within the scope equivalent to the present invention are included in the present invention.

[0091] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0092] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the process in the flowchart. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0093] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0094] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0095] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for monitoring the operation of a hydraulic turbine, characterized in that: include: Receive real-time operation data and output data of the turbine input from the outside; Input the real-time operation data of the turbine into the turbine operation analysis model to calculate the output data prediction value; Compare the output data prediction value with the real-time output data to determine whether the data difference between the two exceeds the abnormal threshold; In response to the data difference not exceeding the abnormal threshold, it is determined that the turbine operation is normal; In response to the data difference exceeding the abnormal threshold, it is determined that the turbine is operating abnormally, and the abnormal situation is reported to the operation management platform.

2. The method for monitoring the operation of a water turbine according to claim 1, characterized in that: The operation data include working water head, flow rate, rotation speed, vibration swing, pressure pulsation and air gap, and the operation data is detected by sensors installed in the hydropower station.

3. The method for monitoring the operation of a water turbine according to claim 1, characterized in that: The turbine operation analysis model is obtained by training a convolutional neural network model as the underlying model, and includes: Collect historical operation data and output data during the operation of the turbine; The historical operation data is divided into a training set and a validation set, and the training set is input into the convolutional neural network model for training to obtain a theoretical model of turbine operation; The verification set is used to verify the turbine operation theory model to determine whether the turbine operation theory model has been trained; In response to the fact that the training of the theoretical model of the operation of the hydraulic turbine is not completed, the model is continuously trained using historical operation data; In response to the completion of the training of the water turbine operation theoretical model, the water turbine operation theoretical model is used as the water turbine operation analysis model.

4. The method for monitoring the operation of a water turbine according to claim 3, characterized in that: The verification set is used to verify the turbine operation theoretical model to determine whether the turbine operation theoretical model has been trained, including: Input the validation set into the turbine operation theory model to calculate the theoretical output data; Compare the theoretical output data with the historical output data to determine whether the data difference between the two exceeds a preset threshold; In response to the data difference exceeding a preset threshold, determining that the training of the turbine operation theoretical model is incomplete; In response to the data difference not exceeding the preset threshold, it is determined that the training of the turbine operation theoretical model is completed.

5. The method for monitoring the operation of a water turbine according to claim 1, characterized in that: The turbine operation analysis model can analyze the change trend of turbine output performance according to the output data prediction value, and report the analysis results to the operation management platform.

6. The method for monitoring the operation of a water turbine according to claim 1, characterized in that: When the turbine operation analysis model determines that the turbine operation is abnormal, the real-time operation data and the historical operation data are compared and analyzed to infer the source of the turbine operation abnormality and obtain the abnormal analysis result, and the abnormal analysis result is reported to the turbine degradation feature knowledge base of the operation management platform.

7. The method for monitoring the operation of a water turbine according to claim 6, characterized in that: The expression of the abnormal analysis result is: S=φ*a+β*b+χ*c+δ*d Where: S is the anomaly analysis result, φ is the anomaly code at position one, a is the anomaly probability at position one, β is the anomaly code at position two, b is the anomaly probability at position two, χ is the anomaly code at position three, c is the anomaly probability at position three, δ is the anomaly code at position four, d is the anomaly probability at position four, and a>b>c>d.

8. A turbine operation monitoring system, characterized in that: include: Data receiving module: receiving real-time operation data and output data of the turbine inputted from the outside; Output prediction module: input the real-time operation data of the turbine into the turbine operation analysis model to calculate the output data prediction value; Data comparison module: compares the output data prediction value with the real-time output data to determine whether the data difference between the two exceeds the abnormal threshold; Abnormality determination module: in response to the data difference not exceeding the abnormality threshold, it is determined that the turbine operation is normal; In response to the data difference exceeding the abnormal threshold, it is determined that the turbine is operating abnormally, and the abnormal situation is reported to the operation management platform.

9. A turbine operation monitoring device, characterized in that: including processor and storage medium; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images