A device health state prediction method, electronic device and system
By using historical power generation prediction and neural network models, the problem of assessing the health status of power generation equipment has been solved, enabling precise judgment of the timing of maintenance for generator sets.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- WANAN HYDROPOWER PLANT OF GUODIAN JIANGXI ELECTRIC POWER CO LTD
- Filing Date
- 2024-11-19
- Publication Date
- 2026-05-22
AI Technical Summary
Existing technologies make it difficult to accurately assess the health status of power generation equipment, leading to difficulties in determining when maintenance is needed.
By predicting load consumption based on the historical power generation of generator sets, and using a combination of long short-term memory network and convolutional neural network models, the health consumption value of generator sets is predicted, and maintenance timing is determined by combining health thresholds.
It enables accurate assessment of the health status of generator sets, helping staff to accurately determine maintenance timing and avoid too many or too few maintenance tasks.
Smart Images
Figure CN119853266B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power plant management technology, and in particular to a method, electronic device and system for predicting equipment health status. Background Technology
[0002] As a core component of the power system, the stable and efficient operation of power generation equipment is the cornerstone of ensuring a safe and reliable power supply. Therefore, regular and professional maintenance of power generation equipment is of paramount importance.
[0003] From a safety perspective, regularly maintaining power generation equipment to ensure it is in good working order is an important measure to ensure production safety and prevent accidents.
[0004] Furthermore, from an environmental perspective, maintaining power generation equipment also helps reduce pollutant emissions. Regular maintenance ensures the normal operation of these devices, reduces pollutant emissions, and protects the ecological environment.
[0005] In conclusion, maintaining power generation equipment is not only necessary to ensure the stable operation of the power system and reduce economic costs, but also essential for ensuring personnel safety and protecting the ecological environment. However, in actual power generation equipment maintenance work, staff often find it difficult to assess the health status of the equipment, thus making it difficult to accurately determine the appropriate time for maintenance.
[0006] In view of this, the present invention is proposed. Summary of the Invention
[0007] The purpose of this invention is to provide a method, electronic device and system for predicting the health status of equipment. This solution can predict the consumption of generator sets by predicting the load and the status of the generator sets, and determine whether maintenance is required. It can pre-assess the health status of power generation equipment and help staff accurately grasp the timing of maintenance.
[0008] This invention provides a method for predicting the health status of equipment, the method comprising the following steps:
[0009] The power generation of the generator set is predicted based on its historical power generation.
[0010] The health consumption value for the first prediction period is calculated based on the power generation capacity of the first prediction period, and the expected health value for the first prediction period is calculated based on the current health value of the generator set and the health consumption value for the first prediction period.
[0011] The health value of the first predicted time period is compared with the preset health threshold, and the generator set is determined based on the comparison result to determine whether maintenance is required during the first predicted time period.
[0012] If it is determined that maintenance of the generator set is required during the first predicted time period, a first pre-maintenance instruction is sent to the control terminal. The first pre-maintenance instruction is used to instruct the staff to perform maintenance on the generator set after the first predicted time period.
[0013] Using the above scheme, after each maintenance of the generator set, the health value of the generator set is reset to a preset value. Furthermore, since the generator set consumes different amounts of power under different loads, this scheme predicts the power generation for the first prediction period based on historical power generation data, and further calculates the required consumption for the first prediction period based on the predicted power generation. Based on the current health value and the health value required for the first prediction period, the health value for the first prediction period is estimated. If the generator set health value is low in the first prediction period, maintenance is required. This scheme can predict the consumption of the generator set and the status of the generator set by predicting the load, and determine whether maintenance is required. It can pre-assess the health status of the power generation equipment, helping staff to accurately grasp the timing of maintenance.
[0014] In some embodiments of the present invention, in the step of predicting the power generation of a first prediction time period based on the historical power generation of the generator set, the power generation of multiple time points in multiple historical time periods is constructed as a first input vector, and the first input vector is input into a pre-trained long short-term memory network model, wherein the long short-term memory network model outputs the predicted power generation of multiple time points in the first prediction time period.
[0015] In some embodiments of the present invention, in the step of calculating the health consumption value of the first predicted time period based on the power generation of the first predicted time period, the power generation of multiple time points in the first predicted time period is constructed as a second input vector, and the second input vector is input into a pre-trained convolutional neural network model, wherein the convolutional neural network model outputs the corresponding health consumption value.
[0016] The above scheme uses two models for joint processing. The first model predicts the power generation in the future time period, which corresponds to the load intensity of the generator set. The second model determines the consumption of the generator set based on the load intensity, and the corresponding decline in the health level is the health consumption value. This method efficiently and accurately determines the consumption of the generator set.
[0017] In some embodiments of the present invention, in the step of calculating the expected health value of the first predicted time period based on the current health value of the generator set and the health consumption value of the first predicted time period, the difference between the current health value of the generator set and the health consumption value of the first predicted time period is calculated to obtain the health value of the first predicted time period.
[0018] Using the above scheme, the health value of the generator set is reset after each maintenance, and the consumption value of the health value is calculated after each time period. Different loads correspond to different consumption values, and the consumption value can be 0. When the consumption value is lower than the threshold, a maintenance command is issued.
[0019] In some embodiments of the present invention, the method further includes: receiving a maintenance completion command in real time, and resetting the health value of the generator set based on the maintenance completion command.
[0020] In some embodiments of the present invention, if it is determined that maintenance of the generator set is not required during the first predicted time period, the step of determining whether maintenance of the generator set is required during the first predicted time period based on the comparison result further includes:
[0021] Calculate the power generation for the second forecast period based on historical power generation and the power generation predicted for the first forecast period.
[0022] The health consumption value for the second prediction period is calculated based on the power generation capacity of the second prediction period. The expected health value for the second prediction period is calculated based on the current health value of the generator set, the health consumption value for the first prediction period, and the health consumption value for the second prediction period.
[0023] The health value for the second predicted time period is compared with the preset health threshold, and the comparison result determines whether the generator set needs to be maintained during the second predicted time period.
[0024] If it is determined that maintenance of the generator set is required during the second predicted time period, a second pre-maintenance instruction is sent to the control terminal, which includes the expected maintenance time.
[0025] Using the above scheme, once it is determined that the generator set does not need to be maintained in the first predicted time period, the health value of the next predicted time period is continuously estimated in the same way until it is determined that a second pre-maintenance instruction needs to be issued. By continuously making subsequent determinations, the time point when maintenance is required can be fuzzily estimated, and the estimated maintenance time can be fed back to the staff through the second maintenance instruction. This allows the staff to make predictions for maintenance over a longer period of time, enabling them to better arrange their work or carry out maintenance in advance, avoiding an excessive number of maintenance tasks that are difficult to handle.
[0026] In some embodiments of the present invention, if it is determined that maintenance of the generator set is not required in the second predicted time period, in the step of determining whether maintenance of the generator set is required in the first predicted time period based on the comparison result, the calculation method of the health value corresponding to the second predicted time period is adopted to continuously calculate the health value of the next predicted time period until it is determined that maintenance of the generator set is required in that time period, and a second pre-maintenance command is sent to the control terminal.
[0027] In some embodiments of the present invention, in the step of calculating the expected health value of the second predicted time period based on the current health value of the generator set, the health consumption value of the first predicted time period, and the health consumption value of the second predicted time period, the difference between the current health value of the generator set and the sum of the health consumption values of the first and second predicted time periods is calculated to obtain the health value of the second predicted time period.
[0028] Another aspect of the present invention relates to an electronic device having a computer program stored thereon, which, when executed by a processor, implements the steps of the aforementioned device health status prediction method.
[0029] Another aspect of the present invention relates to a device health status prediction system, the system comprising a computer device including a processor and a memory, the memory storing computer instructions, the processor executing the computer instructions stored in the memory, and the system implementing the steps of the method when the computer instructions are executed by the processor.
[0030] In summary, the present invention has the following beneficial effects:
[0031] 1. This solution can predict the power consumption of the generator set and its status by forecasting the load, thus determining whether maintenance is needed. It allows for a preliminary assessment of the health status of the power generation equipment, helping staff accurately determine when maintenance is necessary.
[0032] 2. This solution uses two models for joint processing. The first model predicts the power generation in the future time period, which corresponds to the load intensity of the generator set. The second model determines the consumption of the generator set based on the load intensity, which corresponds to the decline in the health level, i.e. the health consumption value. This method efficiently and accurately determines the consumption of the generator set.
[0033] 3. This solution resets the health value of the generator set after each maintenance, and then calculates the consumption value of the health value after each time period. Different loads correspond to different consumption values, and the consumption value can be 0. When the consumption value is lower than the threshold, a maintenance command is issued.
[0034] 4. When this scheme determines that the generator set does not need to be maintained in the first predicted time period, it continuously uses the same method to estimate the health value of the next predicted time period until it determines that a second pre-maintenance instruction needs to be issued. By continuously determining the time point that needs maintenance, it can make a fuzzy estimate of the time point that needs maintenance, and feed back the estimated maintenance time to the staff through the second maintenance instruction. This allows the staff to make predictions for maintenance over a longer period of time, enabling them to better arrange their work or carry out maintenance in advance, avoiding too many maintenance tasks that are difficult to handle. Attached Figure Description
[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0036] Figure 1 This is a schematic diagram of one embodiment of the device health status prediction method of the present invention;
[0037] Figure 2 This is a schematic diagram of another embodiment of the device health status prediction method of the present invention. Detailed Implementation
[0038] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention as detailed in the appended claims.
[0039] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular forms “a,” “the,” and “the” used in this invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.
[0040] like Figure 1 As shown, the present invention provides a method for predicting the health status of equipment, the method comprising the following steps:
[0041] Step S100: Predict the power generation for the first prediction period based on the historical power generation of the generator set.
[0042] In practice, the power generation of historical power generation includes the power generation at multiple points in history.
[0043] Step S200: Calculate the health consumption value of the first prediction time period based on the power generation of the first prediction time period, and calculate the expected health value of the first prediction time period based on the current health value of the generator set and the health consumption value of the first prediction time period.
[0044] Step S300: Compare the health value of the first predicted time period with the preset health threshold, and determine whether the generator set needs to be maintained during the first predicted time period based on the comparison result.
[0045] In the specific implementation process, in the step of determining whether the generator set needs to be maintained in the first predicted time period based on the comparison results, if the health value corresponding to the first predicted time period is less than the preset health threshold, it is determined that maintenance is required.
[0046] Step S410: If it is determined that the generator set needs to be maintained during the first predicted time period, a first pre-maintenance instruction is sent to the control terminal. The first pre-maintenance instruction is used to instruct the staff to maintain the generator set after the first predicted time period.
[0047] In practice, the control terminal can be a staff member's mobile device or a server in the central control center.
[0048] Using the above scheme, after each maintenance of the generator set, the health value of the generator set is reset to a preset value. Furthermore, since the generator set consumes different amounts of power under different loads, this scheme predicts the power generation for the first prediction period based on historical power generation data, and further calculates the required consumption for the first prediction period based on the predicted power generation. Based on the current health value and the health value required for the first prediction period, the health value for the first prediction period is estimated. If the generator set health value is low in the first prediction period, maintenance is required. This scheme can predict the consumption of the generator set and the status of the generator set by predicting the load, and determine whether maintenance is required. It can pre-assess the health status of the power generation equipment, helping staff to accurately grasp the timing of maintenance.
[0049] In some embodiments of the present invention, in the step of predicting the power generation of a first prediction time period based on the historical power generation of the generator set, the power generation of multiple time points in multiple historical time periods is constructed as a first input vector, and the first input vector is input into a pre-trained Long Short-Term Memory (LSTM) network model, which outputs the predicted power generation of multiple time points in the first prediction time period.
[0050] LSTM controls the retention and forgetting of information at different points in time by introducing three important gating mechanisms—the forget gate, the input gate, and the output gate. These gating mechanisms allow the network to dynamically decide which information should be retained in memory and which should be forgotten, thus effectively solving the problem of long-term dependencies.
[0051] In some embodiments of the present invention, in the step of calculating the health consumption value of the first predicted time period based on the power generation of the first predicted time period, the power generation of multiple time points in the first predicted time period is constructed as a second input vector, and the second input vector is input into a pre-trained convolutional neural network model, wherein the convolutional neural network model outputs the corresponding health consumption value.
[0052] In the specific implementation process, the Convolutional Neural Networks (CNN) model is a deep learning model that includes multiple convolutional layers, pooling layers, fully connected layers, and classification layers. The classification layer outputs the health consumption value.
[0053] The above scheme uses two models for joint processing. The first model predicts the power generation in the future time period, which corresponds to the load intensity of the generator set. The second model determines the consumption of the generator set based on the load intensity, and the corresponding decline in the health level is the health consumption value. This method efficiently and accurately determines the consumption of the generator set.
[0054] In some embodiments of the present invention, in the step of calculating the expected health value of the first predicted time period based on the current health value of the generator set and the health consumption value of the first predicted time period, the difference between the current health value of the generator set and the health consumption value of the first predicted time period is calculated to obtain the health value of the first predicted time period.
[0055] Using the above scheme, the health value of the generator set is reset after each maintenance, and the consumption value of the health value is calculated after each time period. Different loads correspond to different consumption values, and the consumption value can be 0. When the consumption value is lower than the threshold, a maintenance command is issued.
[0056] In some embodiments of the present invention, the method further includes: receiving a maintenance completion command in real time, and resetting the health value of the generator set based on the maintenance completion command.
[0057] like Figure 2 As shown, in some embodiments of the present invention, if it is determined that maintenance of the generator set is not required during the first predicted time period, the step of determining whether maintenance of the generator set is required during the first predicted time period based on the comparison result further includes:
[0058] Step S421: Calculate the power generation for the second prediction period based on the historical power generation and the predicted power generation for the first prediction period.
[0059] Step S422: Calculate the health consumption value of the second prediction time period based on the power generation of the second prediction time period, and calculate the expected health value of the second prediction time period based on the current health value of the generator set, the health consumption value of the first prediction time period, and the health consumption value of the second prediction time period.
[0060] In practice, the method for calculating the power generation in the second prediction period is the same as the method for calculating the power generation in the first prediction period.
[0061] In practice, the method for calculating the health consumption value for the second prediction period is the same as the method for calculating the health consumption value for the first prediction period.
[0062] Step S423: Compare the health value of the second predicted time period with the preset health threshold, and determine whether the generator set needs to be maintained during the second predicted time period based on the comparison result.
[0063] Step S424: If it is determined that the generator set needs to be maintained during the second predicted time period, a second pre-maintenance instruction is sent to the control terminal. The second pre-maintenance instruction includes the expected maintenance time point.
[0064] In practice, the maintenance time point is the last time point of the corresponding time period.
[0065] Using the above scheme, once it is determined that the generator set does not need to be maintained in the first predicted time period, the health value of the next predicted time period is continuously estimated in the same way until it is determined that a second pre-maintenance instruction needs to be issued. By continuously making subsequent determinations, the time point when maintenance is required can be fuzzily estimated, and the estimated maintenance time can be fed back to the staff through the second maintenance instruction. This allows the staff to make predictions for maintenance over a longer period of time, enabling them to better arrange their work or carry out maintenance in advance, avoiding an excessive number of maintenance tasks that are difficult to handle.
[0066] In some embodiments of the present invention, if it is determined that maintenance of the generator set is not required in the second predicted time period, in the step of determining whether maintenance of the generator set is required in the first predicted time period based on the comparison result, the calculation method of the health value corresponding to the second predicted time period is adopted to continuously calculate the health value of the next predicted time period until it is determined that maintenance of the generator set is required in that time period, and a second pre-maintenance command is sent to the control terminal.
[0067] In some embodiments of the present invention, in the step of calculating the expected health value of the second predicted time period based on the current health value of the generator set, the health consumption value of the first predicted time period, and the health consumption value of the second predicted time period, the difference between the current health value of the generator set and the sum of the health consumption values of the first and second predicted time periods is calculated to obtain the health value of the second predicted time period.
[0068] Another aspect of the present invention relates to an electronic device having a computer program stored thereon, which, when executed by a processor, implements the steps of the aforementioned device health prediction method. The electronic device may be a tangible storage medium, such as random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, register, floppy disk, hard disk, removable storage disk, CD-ROM, or any other form of storage medium known in the art.
[0069] This invention also provides a device health status prediction system. The system includes a computer device, which includes a processor and a memory. The memory stores computer instructions, and the processor executes the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps of the method.
[0070] Those skilled in the art will understand that the exemplary components, systems, and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Whether implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention. When implemented in hardware, it can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this invention are programs or code segments used to perform the desired tasks. The programs or code segments can be stored in a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried in a carrier wave.
[0071] It should be clarified that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the present invention.
[0072] In this invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or in place of features of other embodiments.
[0073] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, various modifications and variations of the embodiments of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for predicting the health status of equipment, characterized in that, The steps of the method include: The power generation of the generator set is predicted based on its historical power generation. The health consumption value of the first prediction time period is calculated based on the power generation of the first prediction time period. The power generation of multiple time points in the first prediction time period is constructed as a second input vector. The second input vector is input into a pre-trained convolutional neural network model. The convolutional neural network model outputs the corresponding health consumption value. The expected health value of the first prediction time period is calculated based on the current health value of the generator set and the health consumption value of the first prediction time period. The health value of the first predicted time period is compared with the preset health threshold, and the generator set is determined based on the comparison result to determine whether maintenance is required during the first predicted time period. If it is determined that the generator set needs to be maintained during the first predicted time period, a first pre-maintenance instruction is sent to the control terminal. The first pre-maintenance instruction is used to instruct the staff to maintain the generator set after the first predicted time period. Calculate the power generation for the second forecast period based on historical power generation and the power generation predicted for the first forecast period. The health consumption value for the second prediction period is calculated based on the power generation capacity of the second prediction period. The expected health value for the second prediction period is calculated based on the current health value of the generator set, the health consumption value for the first prediction period, and the health consumption value for the second prediction period. The health value for the second predicted time period is compared with the preset health threshold, and the comparison result determines whether the generator set needs to be maintained during the second predicted time period. If it is determined that maintenance of the generator set is required during the second predicted time period, a second pre-maintenance instruction is sent to the control terminal, which includes the expected maintenance time.
2. The equipment health status prediction method according to claim 1, characterized in that: In the step of predicting the power generation of the first prediction period based on the historical power generation of the generator set, the power generation of multiple time points in multiple historical time periods is constructed as a first input vector, and the first input vector is input into a pre-trained long short-term memory network model. The long short-term memory network model outputs the predicted power generation of multiple time points in the first prediction period.
3. The equipment health status prediction method according to claim 1, characterized in that: In the step of calculating the expected health value for the first predicted time period based on the current health value of the generator set and the health consumption value for the first predicted time period, the difference between the current health value of the generator set and the health consumption value for the first predicted time period is calculated to obtain the health value for the first predicted time period.
4. The equipment health status prediction method according to claim 1, characterized in that: The method further includes the following steps: receiving a maintenance completion command in real time, and resetting the health value of the generator set based on the maintenance completion command.
5. The equipment health status prediction method according to claim 1, characterized in that: If it is determined that maintenance of the generator set is not required in the second predicted time period, then in the step of determining whether maintenance of the generator set is required in the first predicted time period based on the comparison result, the calculation method of the health value corresponding to the second predicted time period is adopted to continuously calculate the health value of the next predicted time period until it is determined that maintenance of the generator set is required in that time period, and then a second pre-maintenance command is sent to the control terminal.
6. The equipment health status prediction method according to claim 5, characterized in that: In the step of calculating the expected health value for the second predicted time period based on the current health value of the generator set, the health consumption value for the first predicted time period, and the health consumption value for the second predicted time period, the difference between the current health value of the generator set and the sum of the health consumption values for the first and second predicted time periods is calculated to obtain the health value for the second predicted time period.
7. An electronic device, characterized in that: It stores a computer program that, when executed by a processor, implements the steps of the device health prediction method as described in any one of claims 1-6.
8. A device health status prediction system, characterized in that: The system includes a computer device, which includes a processor and a memory. The memory stores computer instructions, and the processor executes the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps of the device health status prediction method according to any one of claims 1-6.