A control method for magnetically levitated ORC waste heat generator sets

Abstract

This invention discloses a control method for a magnetically levitated ORC waste heat generator set, relating to the technical field of magnetically levitated organic Rankine cycle waste heat generator sets. The method includes acquiring environmental information data of the magnetically levitated ORC waste heat generator set and determining the scenario in which the set is located. If the scenario is a land-based scenario, a first control current of the magnetic bearing in the set is adjusted using a first adjustment strategy. If the scenario is a marine scenario, a second control current of the magnetic bearing is adjusted using a second adjustment strategy different from the first. The method automatically adjusts the corresponding control scheme based on the actual on-site conditions perceived by the unit, dynamically adjusting the magnetic bearing current to maintain the optimal rotor levitation gap, preventing overheating of the magnetic bearing, reducing mechanical losses, and ensuring the unit always operates within a high-efficiency range. This ensures a high conversion efficiency of waste heat to electrical energy even under complex operating conditions.

Description

Technical Field

[0001] This invention relates to the field of magnetic levitation organic Rankine cycle (ORC) waste heat generator technology, and specifically to a control method for magnetic levitation ORC waste heat generators. Background Technology

[0002] As a key component of industrial waste heat recovery and utilization, magnetic levitation ORC waste heat generator sets are currently mainly divided into onshore and offshore units. Therefore, different operating environments place differentiated demands on unit control: onshore scenarios typically do not present significant swaying issues, but some scenarios exhibit severe high-frequency vibrations; in marine scenarios, significant swaying poses a significant threat to the stability of the magnetic levitation system. Current technologies cannot automatically adjust the equipment according to the application scenario, requiring separate control schemes for different scenarios, resulting in insufficient scenario adaptability and high design costs. Summary of the Invention

[0003] To overcome the problems existing in related technologies, the present invention provides a control method for a magnetically levitated ORC waste heat generator set.

[0004] According to an embodiment of the present invention, a control method for a magnetically levitated ORC waste heat generator set is provided, the method comprising:

[0005] Acquire environmental information data of the magnetic levitation ORC waste heat generator set;

[0006] Based on the environmental information data, the scene in which the magnetic levitation ORC waste heat generator set is located is determined;

[0007] If the scene is a land scene, the first control current of the magnetic bearing in the magnetic levitation ORC waste heat generator set is adjusted by the first adjustment strategy.

[0008] If the scenario is an ocean scenario, the second control current of the magnetic bearing in the magnetic levitation ORC waste heat generator set is adjusted by a second adjustment strategy different from the first adjustment strategy.

[0009] In some embodiments of the present invention, acquiring environmental information data of the magnetic levitation ORC waste heat generator set includes:

[0010] The sway amplitude data of the magnetic levitation ORC waste heat generator set is obtained through an inertial measurement unit;

[0011] The environmental information data is determined based on the shaking amplitude data.

[0012] In some embodiments of the present invention, determining the scene in which the magnetic levitation ORC waste heat generator unit is located based on the environmental information data includes:

[0013] If the sway amplitude data is greater than or equal to the sway amplitude threshold, then the magnetic levitation ORC waste heat generator set is in the marine scene.

[0014] If the sway amplitude data is less than the sway amplitude threshold, then the magnetic levitation ORC waste heat generator set is in the land scenario.

[0015] In some embodiments of the present invention, adjusting the first control current of the magnetic bearing in the magnetically levitated ORC waste heat generator set through a first adjustment strategy includes:

[0016] Obtain the first offset distance of the rotor in the magnetic bearing from its working position;

[0017] Obtain the second offset distance of the rotor in the magnetic bearing from its working position in the next adjacent period;

[0018] Determine the first deviation distance between the first offset distance and the second offset distance;

[0019] Obtain the first time period between the occurrence of the first offset distance and the occurrence of the second offset distance;

[0020] The first control current is determined based on the second offset distance, the first deviation distance, and the first time period.

[0021] In some embodiments of the present invention, the sway amplitude data includes sway amplitude values ​​and acceleration values;

[0022] The adjustment of the second control current of the magnetic bearing in the magnetically levitated ORC waste heat generator set using a second adjustment strategy different from the first adjustment strategy includes:

[0023] Obtain the vibration amplitude data of the magnetically levitated ORC waste heat generator set, wherein the vibration amplitude data includes vibration amplitude values;

[0024] Based on the sway amplitude value, the acceleration value, and the vibration amplitude value, the first current correction parameter is obtained;

[0025] Obtain the third offset distance of the rotor in the magnetic bearing from its working position;

[0026] Obtain the fourth offset distance of the rotor in the magnetic bearing deviating from the working position in the next adjacent time, and obtain the second deviation distance between the third offset distance and the fourth offset distance;

[0027] Obtain the second time period between the occurrence of the third offset distance and the occurrence of the fourth offset distance;

[0028] The second current correction parameter is calculated based on the fourth offset distance, the second deviation distance, and the second time period.

[0029] The second control current is obtained based on the first current correction parameter and the second current correction parameter.

[0030] In some embodiments of the present invention, the environmental information data further includes swaying frequency and vibration frequency;

[0031] After acquiring the environmental information data of the magnetic levitation ORC waste heat generator set, the control method further includes:

[0032] Determine whether the shaking frequency and the vibration frequency are greater than preset values;

[0033] If so, increase the alarm threshold and the number of filters for the magnetic levitation controller.

[0034] In some embodiments of the present invention, the control method further includes:

[0035] The operating condition information data of the magnetic levitation ORC waste heat generator set is obtained. The operating condition information data includes: the temperature and pressure on the heat source side of the magnetic levitation ORC waste heat generator set, the temperature of the cooling tower in the magnetic levitation ORC waste heat generator set, and the ambient temperature near the magnetic levitation ORC waste heat generator set.

[0036] The working condition information data is integrated and classified using a convolutional neural network to obtain the first detection result;

[0037] Based on the first detection result, adjust the speed of the motor and / or the speed of the working fluid pump and / or the frequency of the circulating water pump.

[0038] In some embodiments of the present invention, the control method further includes:

[0039] Obtain safety information data of the magnetic levitation ORC waste heat generator set, the safety information data including monitoring videos of the magnetic levitation ORC waste heat generator set and its surrounding environment;

[0040] Based on the safety information data, determine whether there are any safety hazards in the magnetic levitation ORC waste heat generator set. If so, issue an alarm.

[0041] In some embodiments of the present invention, determining whether there is a safety hazard in the magnetic levitation ORC waste heat generator set based on the safety information data, and issuing an alarm if so, includes:

[0042] The SlowFast network model was used to detect structural changes, changes in the surrounding environment, and changes in human behavior of the magnetically levitated ORC waste heat generator unit in the surveillance video, and a second detection result was obtained.

[0043] Based on the second detection result, it is determined whether the magnetic levitation ORC waste heat generator set has any safety hazards;

[0044] If so, an alarm will be triggered, wherein the alarm will include the spatial location information and the time of occurrence information of the safety hazard.

[0045] In some embodiments of the present invention, before acquiring the environmental information data of the magnetic levitation ORC waste heat generator set, the method further includes:

[0046] Perform a self-test on the magnetic levitation ORC waste heat generator set to determine if there is a fault in the magnetic levitation ORC waste heat generator set;

[0047] If so, the magnetically levitated ORC waste heat generator set will be automatically reset;

[0048] If not, the magnetic levitation ORC waste heat generator set is turned on. When the motor of the magnetic levitation ORC waste heat generator set reaches the rated speed, the electricity generated by the magnetic levitation ORC waste heat generator set is connected to the user.

[0049] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0050] The control method for a magnetically levitated ORC waste heat generator set disclosed in this invention automatically adjusts the corresponding control scheme according to the actual on-site conditions of the magnetically levitated ORC waste heat generator set. The control system can make appropriate adjustments according to changes in external factors, dynamically adjust the magnetic bearing current to maintain the optimal levitation gap of the rotor, prevent the magnetic bearing from overheating, reduce mechanical losses, and ensure that the magnetically levitated ORC waste heat generator set is always in the high-efficiency operating range, ensuring that it maintains a high conversion efficiency of waste heat to electrical energy even under complex working conditions.

[0051] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0052] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0053] Figure 1 This is a flowchart illustrating a control method for a magnetically levitated ORC waste heat generator set according to one embodiment. Detailed Implementation

[0054] 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 numbers 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.

[0055] As a key piece of equipment for industrial waste heat recovery and utilization, magnetic levitation ORC waste heat generator sets are currently mainly divided into onshore and offshore units. Onshore units are applicable to numerous fields, including steel mills, glass factories, and chemical plants. Offshore units are primarily used in the marine industry. Therefore, different working environments place differentiated demands on unit control. For example, while large-scale swaying is generally not an issue on land, some scenarios experience severe high-frequency vibrations; in marine scenarios, large-scale swaying poses a significant threat to the stability of the magnetic levitation system. Current technologies cannot automatically adjust the equipment according to the application scenario, requiring separate control schemes for different scenarios, resulting in insufficient adaptability and high design costs.

[0056] During unit operation, changes in ambient temperature and humidity, as well as temperature and pressure of the heat source, directly impact power generation efficiency. Current technologies struggle to adjust control strategies in advance based on these data changes, leading to unstable power generation efficiency. Furthermore, prolonged operation in a vibrating environment can cause sensors to loosen or detach, and working fluid leaks. Environmental or human factors can also damage the unit's structure, threatening equipment safety and the safety of surrounding personnel. Current safety monitoring methods are lagging and unable to promptly identify structural anomalies and dangerous human behavior; therefore, a comprehensive safety assurance system is urgently needed.

[0057] An exemplary embodiment of the present invention provides a control method for a magnetically levitated ORC waste heat generator set, such as... Figure 1 As shown, the method includes:

[0058] S100: Obtain environmental information data of the magnetic levitation ORC waste heat generator set.

[0059] In this step, the environmental information data includes sway amplitude data and vibration amplitude data. The sway amplitude data can be measured using an inertial measurement unit (IMU) to detect whether the current attitude of the magnetic levitation ORC waste heat generator set is subject to significant tilting or swaying. The vibration amplitude data can be measured using a vibration sensor. The measurement results are then transmitted to the PLC module (Programmable Logic Controller). The PLC module automatically determines the application scenario of the magnetic levitation ORC waste heat generator set based on the environmental information data and then automatically adjusts the magnetic levitation ORC waste heat generator set according to an appropriate adjustment strategy.

[0060] S200: Based on environmental information data, determine the scene in which the magnetic levitation ORC waste heat generator set is located.

[0061] In this step, while there is generally no significant swaying issue in land-based scenarios, severe high-frequency vibrations are present; in marine scenarios, significant swaying occurs, and this large-scale swaying poses a serious threat to the stability of the magnetic levitation system. Therefore, the system can automatically determine whether the magnetic levitation ORC waste heat generator unit is in a land-based or marine scenario based on the swaying amplitude data.

[0062] S210. If the scene is a land scene, the first control current of the magnetic bearing in the magnetic levitation ORC waste heat generator set is adjusted by the first adjustment strategy.

[0063] In this step, when environmental information data shows that the magnetic levitation ORC waste heat generator set has no obvious shaking but exhibits vibration, the system determines that the magnetic levitation ORC waste heat generator set is in a land-based scenario. When the rotor rotates at high speed, it may cause mass eccentricity. This eccentricity generates periodic centrifugal force, causing the rotor to vibrate at high frequency around its equilibrium position, deviating from its normal operating position. Therefore, by acquiring the amount of rotor offset from its operating position, the required current adjustment can be calculated based on this offset, and then the control current can be changed to use electromagnetic force to pull the rotor back to its operating position. By precisely adjusting the magnetic bearing control current, the rotor is kept at the preset operating position, reducing vibration, noise, and equipment wear caused by positional deviation, improving the stability and efficiency of the magnetic levitation ORC waste heat generator set, and ensuring its long-term reliable operation in land-based scenarios.

[0064] S220. If the scene is an ocean scene, the second control current of the magnetic bearing in the magnetic levitation ORC waste heat generator set is adjusted by a second adjustment strategy different from the first adjustment strategy.

[0065] In this step, when environmental information data shows that the magnetic levitation ORC waste heat generator set is experiencing significant shaking, the system determines that the set is in a marine environment. In a marine environment, the set will not only experience shaking but also vibration. When the set detects significant shaking and vibration, it calculates the correction current for the magnetic bearing based on the shaking amplitude, acceleration, and vibration amplitude. Then, by acquiring the rotor's offset from its working position, it calculates the required current adjustment. Finally, it determines the final control current based on the correction current and the current adjustment, and then changes the control current to use electromagnetic force to pull the rotor back to its working position. By comprehensively considering factors such as shaking amplitude, acceleration, and vibration amplitude, the system accurately calculates and adjusts the control current for the magnetic bearing, ensuring the magnetic levitation motor maintains a stable levitation state even in a marine environment with significant shaking and vibration. This reduces equipment failures caused by environmental interference, ensures the normal operation of the magnetic levitation ORC waste heat generator set, and improves its adaptability and reliability in marine environments.

[0066] In some embodiments of the present invention, obtaining environmental information data of the magnetically levitated ORC waste heat generator set includes: acquiring sway amplitude data of the magnetically levitated ORC waste heat generator set through an inertial measurement unit (IMU), and determining environmental information data based on the sway amplitude data. Exemplarily, the IMU can be formed by a combination of a three-axis gyroscope and a three-axis accelerometer. The IMU is installed on a rigid structural part of the magnetically levitated ORC waste heat generator set, and as close as possible to the center of gravity of the magnetically levitated ORC waste heat generator set, to ensure accurate capture of the overall sway of the magnetically levitated ORC waste heat generator set and reduce measurement errors caused by installation position deviations. The amplitude value of the sway of the magnetically levitated ORC waste heat generator set is acquired through the three-axis gyroscope, and the magnitude and direction of the acceleration of the sway of the magnetically levitated ORC waste heat generator set are measured through the three-axis accelerometer. The data acquired by both are transmitted to a PLC module for appropriate filtering processing to obtain sway amplitude data used to determine the environment in which the magnetically levitated ORC waste heat generator set is located, ensuring the real-time nature and accuracy of the parameter information.

[0067] In some embodiments of the present invention, determining the scene of the magnetically levitated ORC waste heat generator set based on environmental information data includes: if the sway amplitude data is greater than or equal to a sway amplitude threshold, the magnetically levitated ORC waste heat generator set is in a marine scene; if the sway amplitude data is less than the sway amplitude threshold, the magnetically levitated ORC waste heat generator set is in a land scene. Exemplarily, the sway amplitude data includes the sway amplitude value measured by a gyroscope and the acceleration value measured by an accelerometer. In a land scene, the sway amplitude of the magnetically levitated ORC waste heat generator set is usually small; therefore, the sway amplitude value generally does not exceed ±3°. In a marine scene, affected by ocean waves, the sway amplitude of the magnetically levitated ORC waste heat generator set is often larger, with the minimum sway amplitude value often being ±5° or higher. Therefore, the sway amplitude threshold can be set, for example, to ±3°, ±5°, etc., and those skilled in the art can set it according to actual conditions; no limitation is made here. In addition, to avoid misjudgment caused by momentary interference, a duration can be set as a judgment condition. For example, the duration can be 2 seconds, 3 seconds, 4 seconds or other times. That is, for example, when the sway amplitude threshold is greater than or equal to ±5° and the duration exceeds 3 seconds, it can be determined that the magnetic levitation ORC waste heat generator is in an ocean scene; if it only momentarily exceeds the sway amplitude threshold and then quickly returns to the range of the sway amplitude threshold, then there is no need to switch scenes.

[0068] In this embodiment, by comparing the shaking amplitude data with the shaking amplitude threshold, the control unit can quickly and accurately make scenario judgments, providing a basis for the selection of subsequent control strategies. The appropriate adjustment strategy can be automatically adjusted according to different operating scenarios, enabling the same control scheme to handle more operating conditions, improving the product's scenario adaptability, saving design costs, and improving product stability.

[0069] In some embodiments of the present invention, adjusting the first control current of the magnetic bearing in the magnetically levitated ORC waste heat generator set through a first adjustment strategy includes: when it is determined that the magnetically levitated ORC waste heat generator set is in a land scenario, firstly, obtaining the first offset distance of the rotor in the magnetic bearing from its working position. Position sensors, such as eddy current displacement sensors or Hall sensors, can be installed radially and axially within the magnetic bearing to obtain the real-time offset position of the rotor, ensuring comprehensive and accurate capture of the rotor's positional changes in the radial and axial directions. Then, obtaining the second offset distance of the rotor from its working position in the magnetic bearing for the next adjacent time. Determining the first deviation distance between the first and second offset distances, that is, obtaining the first deviation distance by the difference between two consecutively measured rotor offset distances, and simultaneously obtaining the first time period between the occurrence of the first and second offset distances. Based on the second offset distance, the first offset distance, and the first time period, determining the first control current. The formula used is:

[0070]

[0071] in, The first control current (unit: ampere (A)) is used to adjust the magnetic levitation bearing. It is the deviation coefficient (its value range is determined by the system stiffness, current limit and offset range). This is the second offset distance (unit: mm). The first time period (in seconds) between the occurrence of the first offset distance and the occurrence of the second offset distance. The first deviation distance (unit: mm). This is the current drift parameter, used to adjust the initial current to 0, ensuring that the initial current is 0 when there is no offset (its range is determined by the system's initial zero drift error).

[0072] In this way, the system collects the real-time offset data of the rotor and automatically adjusts the current value required by the magnetic levitation bearing according to the actual offset position, so that the rotor is always in a stable working position, keeps the air gap of the magnetic bearing uniform, avoids bearing damage caused by rotor wear, and enables the magnetic bearing to maintain the rotor in the ideal working position more accurately, reducing mechanical friction and energy loss, extending the service life of the equipment, and ensuring the continuous, efficient and stable operation of the magnetic levitation ORC waste heat generator set under complex working conditions, thus ensuring the waste heat recovery efficiency.

[0073] In some embodiments of the present invention, the swaying amplitude data includes swaying amplitude values ​​and acceleration values. Specifically, a gyroscope can be used to obtain the swaying amplitude value of the magnetically levitated ORC waste heat generator set, and an accelerometer can be used to measure the magnitude of the swaying acceleration value of the magnetically levitated ORC waste heat generator set.

[0074] Adjusting the second control current of the magnetic bearing in the magnetically levitated ORC waste heat generator set using a second adjustment strategy, different from the first adjustment strategy, includes: firstly, acquiring vibration amplitude data of the magnetically levitated ORC waste heat generator set, including vibration amplitude values; that is, environmental information data also includes vibration amplitude data, which can be measured using vibration sensors. Based on the swaying amplitude value, acceleration value, and vibration amplitude value, a first current correction parameter is obtained, which can be calculated using the following formula:

[0075]

[0076] in, The first current correction parameter (unit: ampere (A)). It is a proportionality coefficient (related to a linear correction of the sway amplitude). The value represents the amplitude of the sway (unit: mm). It is a coefficient for the acceleration value (linearly related to the acceleration value). This represents the acceleration value (unit: m / s²). It is the vibration coefficient (related to a linear correction of the vibration amplitude). The vibration amplitude value (unit: mm).

[0077] Then, the third offset distance of the rotor in the magnetic bearing from its working position is obtained, and the fourth offset distance of the rotor in the magnetic bearing from its working position in the next adjacent period is obtained, thus obtaining the second deviation distance between the third offset distance and the fourth offset distance; simultaneously, the second time period in which the third offset distance and the fourth offset distance occur is obtained; based on the fourth offset distance, the second deviation distance, and the second time period, the second current correction parameter is calculated; wherein, the second current correction parameter can be calculated using the following formula:

[0078]

[0079] in, This is the second current correction parameter (unit: ampere (A)). It is the deviation coefficient (its value range is determined by the system stiffness, current limit and offset range). This is the fourth offset distance (unit: mm). The second time period (in seconds) between the occurrence of the third and fourth offset distances. The second deviation distance (unit: mm). This is the current drift parameter, used to adjust the initial current to 0, ensuring that the initial current is 0 when there is no offset (its range is determined by the system's initial zero drift error).

[0080] Finally, the second control current is obtained based on the first current correction parameter and the second current correction parameter. The second control current can be calculated using the following formula:

[0081]

[0082] in, The second control current (unit: amperes (A)) The first current correction parameter (unit: ampere (A)). This is the second current correction parameter (unit: ampere (A)).

[0083] Thus, by comprehensively analyzing multi-dimensional dynamic data to precisely adjust the second control current of the magnetic bearing in the magnetic levitation ORC waste heat generator set, the stability and control accuracy of the magnetic levitation ORC waste heat generator set are significantly improved. By acquiring vibration amplitude data and calculating the first current correction parameter based on the sway amplitude value, acceleration value, and vibration amplitude value, the changes in sway characteristics during the operation of the magnetic levitation ORC waste heat generator set can be captured in real time, effectively suppressing rotor instability caused by sway and avoiding equipment damage or failure caused by sway. Furthermore, by acquiring the third offset distance of the magnetic bearing rotor from its current working position, the fourth offset distance of the next adjacent offset, and its second deviation distance, and combining this with the second time period of the offset, the second current correction parameter is calculated. Based on the actual offset position, the required current value of the magnetic levitation bearing is automatically adjusted, ensuring that the rotor is always in a stable working position, keeping the air gap of the magnetic bearing uniform, avoiding bearing damage caused by rotor wear, and ensuring waste heat recovery efficiency. Ultimately, by integrating the first and second current correction parameters, a dual correction mechanism for the magnetic bearing control current is formed, which solves the limitations of single parameter adjustment. This enables the magnetic bearing to maintain the rotor in the ideal working position more accurately, reducing mechanical friction and energy loss, extending the service life of the equipment, and ensuring the continuous, efficient and stable operation of the magnetic levitation ORC waste heat generator set under complex working conditions.

[0084] In some embodiments of the present invention, the environmental information data further includes sway frequency and vibration frequency. For example, continuous sway amplitude data and vibration amplitude data are sampled at fixed time intervals to obtain discrete time series data. Then, a Fourier transform algorithm is used to analyze these time series data to obtain the corresponding sway frequency and vibration frequency.

[0085] After acquiring environmental information data of the magnetic levitation ORC waste heat generator set, the control method further includes: determining whether the swaying frequency and vibration frequency are greater than preset values; if so, increasing the alarm threshold and filtering times of the magnetic levitation controller. When the swaying frequency or vibration frequency exceeds the preset value, for example, when the preset value for the swaying frequency can be 160Hz and the preset value for the vibration frequency can be 260Hz (of course, in other embodiments, those skilled in the art can set it according to the actual situation, which is not limited here), if the original alarm threshold is maintained, it is easy to trigger frequent alarms due to signal fluctuations. For example, high-frequency vibration may cause the vibration amplitude to momentarily exceed the conventional threshold, but it does not actually pose a threat to the magnetic levitation ORC waste heat generator set. By increasing the alarm threshold, these non-substantial exceedance signals caused by high-frequency interference can be filtered out, reducing the number of false alarms. This not only avoids operators being interfered with by invalid alarms, but also reduces the frequency of unnecessary shutdowns for inspection, saving maintenance time and labor costs. High-frequency swaying or vibration will cause the collected signal to contain more noise, and it is difficult to completely eliminate the interference by relying solely on the original filtering times. Therefore, it is necessary to increase the number of filtering iterations to more effectively smooth signal fluctuations, filter out high-frequency noise, and make the vibration amplitude and sway amplitude data input to the PLC controller closer to the true values. A stable signal provides a reliable basis for the current adjustment of the magnetic levitation controller, reduces control errors caused by signal distortion, thereby reducing rotor vibration amplitude and extending the service life of the magnetic bearing.

[0086] In some embodiments of the present invention, the control method further includes:

[0087] The system acquires operating condition data for the magnetic levitation ORC waste heat generator set, including: temperature and pressure on the heat source side, temperature of the cooling tower within the set, and ambient temperature near the set. Temperature and pressure sensors can be installed on the heat source side, and additional temperature sensors can be added near the cooling tower and the set to detect changes in heat source conditions and the surrounding environment. The generator speed, working fluid pump speed, and circulating water pump frequency are adjusted based on the detected data. Timely adjustment of the internal control strategy of the magnetic levitation ORC waste heat generator set according to changes in ambient temperature and heat source conditions improves its operational stability.

[0088] The first detection result is obtained by integrating and classifying the working condition information data using a convolutional neural network.

[0089] The convolutional neural network (CNN) consists of convolutional layers, pooling layers, fully connected layers, and softmax layers. The collected operating condition information data undergoes multiple information extraction processes through the convolutional and pooling layers, followed by data integration through the fully connected layers. Finally, the softmax layer classifies the data and outputs the first detection result. The CNN used can employ existing network models. Historical data of the magnetic levitation ORC waste heat generator set under different operating conditions is collected, including operating condition information data under various states such as normal operation, high load, low load, and fault, as well as the optimal adjustment results for corresponding motor speed, working fluid pump speed, and circulating water pump frequency. The dataset is divided into training, validation, and test sets according to the required proportions. The preprocessed training data is input into the prepared CNN, and the network parameters are continuously adjusted through backpropagation until the model's accuracy on the validation set reaches the preset accuracy. Training is then stopped, and the model is ready for practical application. The CNN model is periodically retrained and optimized using new operating data to adapt to changes in the operating status of the magnetic levitation ORC waste heat generator set and the needs of different operating conditions.

[0090] Based on the initial detection results, the motor speed and / or the working fluid pump speed and / or the circulating water pump frequency are adjusted. The initial detection results are transmitted to the PLC controller, which adjusts the motor speed, working fluid pump speed, and circulating water pump frequency. By acquiring the temperature and pressure on the heat source side, the cooling tower temperature, and the ambient temperature, the operating environment and energy conversion conditions of the magnetic levitation ORC waste heat generator set can be comprehensively reflected. For example, the temperature and pressure on the heat source side directly reflect the intensity of waste heat energy supply, the cooling tower temperature reflects the heat dissipation capacity, and the ambient temperature affects the overall heat exchange efficiency of the magnetic levitation ORC waste heat generator set. Avoiding the limitations of single-parameter monitoring, convolutional neural networks extract deep correlation features from the data. Compared to traditional threshold judgments or simple algorithms, this allows for more accurate classification of complex operating conditions, ensuring that the initial detection results accurately reflect the operating requirements and reducing control deviations caused by classification errors. Adjustments based on the initial detection results ensure that the operating parameters of the motor, working fluid pump, and circulating water pump are highly matched to real-time operating conditions. Adjusting the speed or frequency of each device under different conditions avoids equipment damage, significantly improving the stability and reliability of the magnetic levitation ORC waste heat generator set, reducing the number of downtime maintenance, and ensuring that the magnetic levitation ORC waste heat generator set can be in the optimal operating state under any working condition, thus ensuring power generation efficiency and extending equipment life.

[0091] In some embodiments of the present invention, the control method further includes: acquiring safety information data of the magnetic levitation ORC waste heat generator set, the safety information data including monitoring video of the magnetic levitation ORC waste heat generator set and its surrounding environment; based on the safety information data, determining whether there are any safety hazards in the magnetic levitation ORC waste heat generator set, and if so, issuing an alarm. For example, an industrial-grade high-definition camera supporting infrared night vision and wide dynamic range can be used. In this way, the monitoring video can cover the entire magnetic levitation ORC waste heat generator set and its surrounding environment, monitoring in real time for abnormalities in the appearance of the magnetic levitation ORC waste heat generator set equipment, such as pipe leaks, loose or detached parts, and deformed casing; it can also monitor environmental risks in real time, such as the presence of flammable materials, excessive water accumulation, and operator violations; when a safety hazard is detected, the system will automatically issue an alarm, eliminating the need for continuous monitoring by dedicated personnel. The system automatically filters key information, reducing manpower input and avoiding missed reports due to fatigue or negligence. This improves the structural safety of the magnetic levitation ORC waste heat generator set, ensuring the safety of personnel around the set and the operational safety of the set itself.

[0092] In some embodiments of the present invention, based on safety information data, determining whether there are any safety hazards in the magnetic levitation ORC waste heat generator set, and if so, issuing an alarm notification includes:

[0093] The SlowFast network model was used to detect structural changes, changes in the surrounding environment, and changes in personnel behavior of a magnetically levitated ORC waste heat generator unit in surveillance video, yielding a second detection result. The SlowFast network consists of two branches: a low frame rate branch, suitable for spatial information recognition of a small number of consecutive images with long time intervals; and a high frame rate branch, suitable for temporal information recognition of a large number of consecutive images with short time intervals. The convolutional layers used in both branches maintain the same width and height dimensions but differ in the number of channels and the temporal dimension. This ensures that data is continuously fused during the recognition process while accurately identifying spatial and temporal information, thereby improving the accuracy of target recognition. Keyframes were extracted from the surveillance video at fixed intervals. The slow-frame rate branch extracted low frame rate frames to capture static changes in the structure and environment of the magnetically levitated ORC waste heat generator unit, while the high frame rate branch extracted high frame rate frames to capture dynamic changes in operator behavior. At the top layer of the network, a cross-attention mechanism was used to fuse the spatial features of the slow-frame rate branch and the temporal features of the high-frame rate branch, outputting a fused feature vector for classification.

[0094] Based on the second detection result, it is determined whether there are any safety hazards in the magnetic levitation ORC waste heat generator set. If so, an alarm is issued, including the spatial location information and the time of occurrence of the safety hazard. This can be achieved through audible and visual alarms to alert operators to dangerous behavior, or by sending text messages or notifications to the operator's mobile phone. This improves the structural safety of the magnetic levitation ORC waste heat generator set, ensuring the safety of personnel around the unit and the operational safety of the unit itself.

[0095] In some embodiments of the present invention, before acquiring environmental information data of the magnetic levitation ORC waste heat generator set, the method further includes: performing a self-check on the magnetic levitation ORC waste heat generator set to determine whether there is a fault. If so, the magnetic levitation ORC waste heat generator set is automatically reset. For example, the system can reset three times consecutively. If the fault cannot be eliminated after three resets, the system automatically issues an alarm. Of course, in other embodiments, the number of resets can be set to two, four, or six times, or other values. If not, it enters the standby startup stage, waiting for the operator to start the machine. When the operator starts the magnetic levitation ORC waste heat generator set, the motor levitation is automatically activated. The working fluid is circulated by the working fluid pump, blowing the motor impeller and putting the motor in a power generation state. When the motor of the magnetic levitation ORC waste heat generator set reaches the rated speed, the four-quadrant frequency converter is activated to control the motor to rotate at the rated speed, and the electricity generated by the magnetic levitation ORC waste heat generator set is connected to the user. Thus, the self-checking mechanism allows for a comprehensive inspection of the operating status of key components (such as magnetic bearings, motors, working fluid pumps, and four-quadrant frequency converters) before the start-up of the magnetic levitation ORC waste heat generator set, enabling timely detection of potential faults. Automatic reset can resolve some temporary and minor faults, preventing further damage to the equipment from starting with a fault, and improving the autonomy and efficiency of the startup process. When a set number of resets fails to eliminate the fault, the system automatically issues an alarm, avoiding wasted time due to ineffective resets and promptly reminding operators to conduct manual troubleshooting, ensuring professional handling of the fault and reducing the risk of startup failure due to delays in manual judgment or operational errors.

[0096] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein.

Claims

1. A control method for a magnetically levitated ORC waste heat generator set, characterized in that, include: Acquire environmental information data of the magnetic levitation ORC waste heat generator set; Based on the environmental information data, the scene in which the magnetic levitation ORC waste heat generator set is located is determined; If the scene is a land scene, the first control current of the magnetic bearing in the magnetic levitation ORC waste heat generator set is adjusted by the first adjustment strategy. If the scene is an ocean scene, the second control current of the magnetic bearing in the magnetic levitation ORC waste heat generator set is adjusted by a second adjustment strategy different from the first adjustment strategy. The acquisition of environmental information data for the magnetic levitation ORC waste heat generator set includes: The sway amplitude data of the magnetic levitation ORC waste heat generator set is obtained through an inertial measurement unit; The sway amplitude data includes sway amplitude values ​​and acceleration values; The adjustment of the first control current of the magnetic bearing in the magnetically levitated ORC waste heat generator set through the first adjustment strategy includes: Obtain the first offset distance of the rotor in the magnetic bearing from its working position; Obtain the second offset distance of the rotor in the magnetic bearing from its working position in the next adjacent period; Determine the first deviation distance between the first offset distance and the second offset distance; Obtain the first time period between the occurrence of the first offset distance and the occurrence of the second offset distance; The first control current is determined based on the second offset distance, the first deviation distance, and the first time period; The adjustment of the second control current of the magnetic bearing in the magnetically levitated ORC waste heat generator set using a second adjustment strategy different from the first adjustment strategy includes: Obtain the vibration amplitude data of the magnetically levitated ORC waste heat generator set, wherein the vibration amplitude data includes vibration amplitude values; Based on the sway amplitude value, the acceleration value, and the vibration amplitude value, the first current correction parameter is obtained; Obtain the third offset distance of the rotor in the magnetic bearing from its working position; Obtain the fourth offset distance of the rotor in the magnetic bearing deviating from the working position in the next adjacent time, and obtain the second deviation distance between the third offset distance and the fourth offset distance; Obtain the second time period between the occurrence of the third offset distance and the occurrence of the fourth offset distance; The second current correction parameter is calculated based on the fourth offset distance, the second deviation distance, and the second time period. The second control current is obtained based on the first current correction parameter and the second current correction parameter.

2. The control method for a magnetically levitated ORC waste heat generator set according to claim 1, characterized in that, The acquisition of environmental information data of the magnetic levitation ORC waste heat generator set also includes: The sway amplitude data of the magnetic levitation ORC waste heat generator set is obtained through an inertial measurement unit; The environmental information data is determined based on the shaking amplitude data.

3. The control method for a magnetically levitated ORC waste heat generator set according to claim 2, characterized in that, The determination of the scene in which the magnetic levitation ORC waste heat generator unit is located based on the environmental information data includes: If the sway amplitude data is greater than or equal to the sway amplitude threshold, then the magnetic levitation ORC waste heat generator set is in the marine scene. If the sway amplitude data is less than the sway amplitude threshold, then the magnetic levitation ORC waste heat generator set is in the land scenario.

4. The control method for a magnetically levitated ORC waste heat generator set according to claim 2, characterized in that, The environmental information data also includes swaying frequency and vibration frequency; After acquiring the environmental information data of the magnetic levitation ORC waste heat generator set, the control method further includes: Determine whether the shaking frequency and the vibration frequency are greater than preset values; If so, increase the alarm threshold and the number of filters for the magnetic levitation controller.

5. The control method for a magnetically levitated ORC waste heat generator set according to any one of claims 1-4, characterized in that, The control method further includes: The operating condition information data of the magnetic levitation ORC waste heat generator set is obtained. The operating condition information data includes: the temperature and pressure on the heat source side of the magnetic levitation ORC waste heat generator set, the temperature of the cooling tower in the magnetic levitation ORC waste heat generator set, and the ambient temperature near the magnetic levitation ORC waste heat generator set. The working condition information data is integrated and classified using a convolutional neural network to obtain the first detection result; Based on the first detection result, adjust the speed of the motor and / or the speed of the working fluid pump and / or the frequency of the circulating water pump.

6. The control method for a magnetically levitated ORC waste heat generator set according to any one of claims 1-4, characterized in that, The control method further includes: Obtain safety information data of the magnetic levitation ORC waste heat generator set, the safety information data including monitoring videos of the magnetic levitation ORC waste heat generator set and its surrounding environment; Based on the safety information data, determine whether there are any safety hazards in the magnetic levitation ORC waste heat generator set. If so, issue an alarm.

7. The control method for a magnetically levitated ORC waste heat generator set according to claim 6, characterized in that, Based on the safety information data, the system determines whether the magnetic levitation ORC waste heat generator set has any safety hazards. If so, an alarm is triggered, including: The SlowFast network model was used to detect structural changes, changes in the surrounding environment, and changes in human behavior of the magnetically levitated ORC waste heat generator unit in the surveillance video, and a second detection result was obtained. Based on the second detection result, it is determined whether the magnetic levitation ORC waste heat generator set has any safety hazards; If so, an alarm will be triggered, wherein the alarm will include the spatial location information and the time of occurrence information of the safety hazard.

8. The control method for a magnetically levitated ORC waste heat generator set according to claim 1, characterized in that, Before acquiring the environmental information data of the magnetic levitation ORC waste heat generator set, the following steps are also included: Perform a self-test on the magnetic levitation ORC waste heat generator set to determine if there is a fault in the magnetic levitation ORC waste heat generator set; If so, the magnetically levitated ORC waste heat generator set will be automatically reset; If not, the magnetic levitation ORC waste heat generator set is turned on. When the motor of the magnetic levitation ORC waste heat generator set reaches the rated speed, the electricity generated by the magnetic levitation ORC waste heat generator set is connected to the user.

Images