Secondary channel compensation modeling method in equipment operation state

Abstract

The invention provides a ship low-frequency line spectrum vibration noise active control secondary channel compensation modeling method in an equipment operation state. Aiming at the problems that the existing offline identification needs to frequently start and stop equipment, the calculation burden of online identification is large, the indirect identification precision is low and the like, a least mean square (LMS) algorithm is adopted, and a mapping relation between a reference signal and a primary response signal is firstly constructed; driving an actuator by using the reference signal to obtain a mixed signal superposed with the primary response; predicting primary response based on the mapping relation, and separating secondary response; and secondary channel modeling is completed by using the LMS algorithm again. The method breaks through the limitation of traditional modeling, can achieve high-precision modeling without shutdown, effectively reduces the calculation burden, is not interfered by the change of the structural mode of a controlled object, greatly improves the debugging efficiency of an active control system of large marine equipment, and brings an innovative technical breakthrough to the field of ship vibration noise control.

Description

Technical Field

[0001] The present invention relates to the field of active control of low-frequency line spectrum vibration noise of ships, and in particular to a compensation modeling method capable of achieving high-precision secondary channel modeling when equipment is in operation. Background Art

[0002] In the marine industry, active control technology has become an important approach to addressing low-frequency line-spectrum vibration and noise. However, due to the typically large scale of controlled objects and the complex transmission paths of vibration and noise, conventional control strategies struggle to achieve optimal results. Currently, a universal active control design approach has yet to be established in engineering.

[0003] Currently, active control strategies based on offline identification to establish secondary channel models are widely used, but they have numerous drawbacks. Any hardware solution change requires re-identification. To achieve optimal identification results, the controlled object is often shut down to reduce interference from the primary response during the identification process. However, in actual ship operations, the startup and shutdown of onboard equipment is extremely complex, and frequent startup and shutdown operations are often not possible on site.

[0004] To continue debugging after the hardware solution change, engineers tried various methods. For example, they excited the secondary source with a higher energy level, hoping to make the error signal, which combines the primary and secondary responses, significantly higher than the pre-mixing value, thereby ensuring the identification accuracy of the main channel. However, because the secondary source's response in the coupled channel was sometimes too small, achieving the goal of a mixed response greater than the primary response was difficult, ultimately leading to significant deviations in the identification results.

[0005] Alternatively, there are methods that use adjacent frequencies instead of the target frequency for identification, or interpolate the frequency response information of the target frequency using two frequencies before and after the target frequency. However, these indirect methods are only feasible when the frequency response characteristics of the secondary channel change slowly. If the frequency response characteristics at the target frequency change dramatically, the obtained result will have a large error compared to the true value, thereby reducing the convergence margin of the control system and significantly reducing the control effect.

[0006] Although online identification can theoretically solve some of the above problems, in actual engineering applications, this method will introduce additional computational burdens and have an adverse effect on the control effect, which makes engineers have concerns about its application.

[0007] In summary, the existing secondary channel modeling method in active control of ship vibration and noise has many defects. A new method is urgently needed to solve these problems in order to improve modeling accuracy, reduce the number of equipment starts and stops during actual ship commissioning, improve commissioning efficiency, and ensure that good commissioning results are achieved within an effective time. Summary of the Invention

[0008] This invention aims to provide a secondary channel compensation modeling method for equipment operation, addressing the problems of existing secondary channel modeling methods for active control of low-frequency line spectrum vibration noise in ships. Specifically, it overcomes the drawbacks of offline identification, such as the frequent equipment startup and shutdown required, the heavy computational burden of online identification, and the low accuracy of indirect identification methods. It achieves high-precision secondary channel models while the equipment is operating, reduces the number of equipment startups and shutdowns during actual ship commissioning, improves commissioning efficiency, and ensures that the control system achieves good commissioning results within a reasonable time.

[0009] The technical solution of the secondary channel compensation modeling method in the device operation state includes the following steps:

[0010] Step 1:

[0011] Establishing a mapping relationship: Using the traditional least mean square (LMS) algorithm, a reference signal x is generated within the controller and used as the input. The primary response d is used as the output. The LMS algorithm then establishes a mapping relationship f(·) between the two. This mapping relationship forms the basis for subsequent steps and reflects the inherent connection between the reference signal x and the primary response d.

[0012] Step 2;

[0013] Generate and mix signals: Use the reference signal x as the source signal to drive the actuator to work, and the actuator generates a secondary response e s , the secondary response e s It is superimposed with the primary response d to form a mixed signal e.

[0014] Step 3:

[0015] Predicting the primary response: In the secondary channel modeling stage, the mapping relationship f(·) established in step 1 is used to predict the primary response d′ at the current moment.

[0016] d'=f(x)

[0017] In this way, the primary response d′ at the current moment can be reasonably estimated when the equipment is in operation, which is used to separate the secondary response e s Provide data support.

[0018] Step 4:

[0019] Separate the secondary response: Restore the actuator's secondary response e based on the mixed signal e and the predicted primary response d' at the current moment s .

[0020] e s =e-d'=ef(x)

[0021] Specifically, the secondary response e is accurately separated from the mixed signal e by the above calculation method. s part, making the subsequent modeling of secondary channels more accurate.

[0022] Step 5:

[0023] Complete secondary channel modeling: reference signal x and restored secondary response e s The LMS algorithm is used again to obtain the secondary channel model Sz for the input and output signals, thus completing the secondary channel modeling of the controlled device under the working state. After steps 1-5, the secondary channel model can be accurately constructed during the continuous operation of the device.

[0024] Beneficial effects

[0025] Modeling under running status: Different from the traditional offline identification method, the present invention can realize the modeling of secondary channels under the running status of the equipment, without the need to frequently start and stop the equipment for modeling. It adapts to the actual complex start and stop conditions of actual ship equipment, greatly improving the timeliness and convenience of modeling.

[0026] Reducing computational burden and avoiding interference: The method of the present invention does not increase the computational burden of the controller during the control process, nor does it require the addition of additional noise, effectively avoiding adverse effects on the control system and ensuring the stability and reliability of the control system.

[0027] High-precision modeling: Compared to commonly used engineering methods such as replacing frequencies with similar ones or interpolating frequencies on both sides, this method is unaffected by modal variations in the controlled object's structure. Even when the secondary response is significantly smaller than the primary response, high modeling accuracy is achieved, improving the control system's convergence margin and effectiveness.

[0028] Improved commissioning efficiency: For large marine equipment, the start-up and shutdown operations are complex and costly. The method of the present invention avoids frequent equipment start-up and shutdown during the commissioning process, significantly shortening the commissioning time, improving on-site commissioning efficiency, and reducing commissioning costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 Schematic diagram of the present invention DETAILED DESCRIPTION

[0030] To make the objectives, technical solutions, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is provided with reference to the accompanying drawings. It should be noted that the following embodiments are only some specific embodiments of the present invention and are not intended to be limiting of the technical solutions of the present invention. In actual applications, adjustments and variations may be made according to specific circumstances.

[0031] In this embodiment, the vibration and noise control of a large ship propulsion system is used as an example. During operation, this system generates strong low-frequency line spectrum vibration and noise, which seriously affects the ship's performance and the crew's working and living environment. The secondary channel compensation modeling method of the present invention is applied to address this issue during the equipment's operating state.

[0032] The technical solution of the secondary channel compensation modeling method under the operating state of the device includes the following steps:

[0033] Step 1:

[0034] Establishing a mapping relationship: The traditional least mean square (LMS) algorithm is used within the ship's controller. The relevant parameters of the controller are set to generate a reference signal x. This reference signal x can be obtained by processing the speed signal of the ship's propulsion system or other characteristic signals related to vibration noise. The reference signal x is used as the input signal, and the primary response d is collected at the same time. The primary response d can be obtained by installing vibration sensors at specific locations on the ship. These locations are usually areas where vibration is more obvious and has a greater impact on the overall performance of the ship. The input reference signal x and the output primary response d are calculated and analyzed using the LMS algorithm to establish a mapping relationship f(·) between the two. In actual operation, the collected signal needs to be preprocessed by filtering, amplifying, etc. to improve the signal quality and ensure the accuracy of the mapping relationship f(·).

[0035] Step 2:

[0036] Generate and mix signals: Use the reference signal x obtained after establishing the mapping relationship as the source signal and input it into the actuator to drive the actuator to work. The actuator generates a secondary response e according to the reference signal x. s In ship propulsion systems, actuators can be active vibration reduction devices installed near the vibration source, such as piezoelectric ceramic drives. Secondary response s After generation, it will be superimposed in space with the primary response d to form a mixed signal e. During this process, it is necessary to ensure the stability of the actuator's working state to avoid abnormal secondary responses due to actuator failure.

[0037] Step 3:

[0038] Predicting the primary response: Entering the secondary channel modeling phase, using the mapping relationship f(·) established in step 1. Input the current reference signal x into this mapping relationship and predict the current primary response d′ through calculation.

[0039] d'=f(x)

[0040] Because the operating conditions of a ship's propulsion system change during operation, the reference signal x needs to be updated in real time to ensure that the predicted primary response d′ accurately reflects the current system state. During the prediction process, the mapping relationship f(·) can be appropriately adjusted and optimized based on actual conditions. For example, the LMS algorithm parameters can be fine-tuned based on factors such as the ship's operating time and load changes.

[0041] Step 4:

[0042] Separate the secondary response: Based on the mixed signal e and the predicted primary response d′ at the current moment, restore the actuator's secondary response e through a specific calculation method s The specific calculation process can be e s =e-d'=ef(x). Through this difference calculation, the secondary response e can be accurately separated from the mixed signal e. s In actual calculations, issues such as signal accuracy and noise interference must be considered. Digital signal processing technology can be used to perform noise reduction and filtering on the signal to improve the accuracy of the separation results.

[0043] Step 5:

[0044] Complete the secondary channel modeling: the reference signal x and the separated secondary response e s The LMS algorithm is used as the input and output signals for calculation again. By continuously iterating the LMS algorithm and adjusting the parameters in the algorithm, the secondary channel model Sz is finally obtained, thus completing the secondary channel modeling of the controlled equipment (i.e., the ship propulsion system) under the working state.

[0045] After modeling is complete, the secondary channel model Sz can be verified and evaluated. For example, by comparing actual measured vibration and noise data with the model's predicted data, the accuracy and reliability of the model can be determined. If the model accuracy does not meet the requirements, the previous steps can be repeated to adjust and optimize the relevant parameters and re-model the model.

[0046] The above examples demonstrate that the method of the present invention can effectively model secondary channels while equipment is operating. Compared with traditional methods, it avoids the numerous problems associated with frequent equipment startups and shutdowns, reduces computational burdens, improves modeling accuracy, and enhances the debugging efficiency of active control systems for large marine equipment, providing reliable technical support for active control of ship vibration and noise.

Claims

1. A secondary channel compensation modeling method under the operation state of the device, characterized in that: The steps include: Step 1 Establishing a mapping relationship: Using the traditional least mean square algorithm (LMS), the controller generates a reference signal x as the input signal and the primary response d as the output signal. The LMS algorithm is used to establish a mapping relationship f(·) between the two. Step 2 Generate and mix signals: Use the reference signal x as the source signal to drive the actuator to work, and the actuator generates a secondary response e s , the secondary response e s It is superimposed with the primary response d to form a mixed signal e; Step 3 Predicting the primary response: In the secondary channel modeling phase, the mapping relationship f(·) established in step 1 is used to predict the primary response d′ at the current moment; Step 4 Separate the secondary response: Restore the actuator's secondary response e based on the mixed signal e and the predicted primary response d' at the current moment s ; Step 5 Complete secondary channel modeling: reference signal x and restored secondary response e s For input and output signals, the LMS algorithm is used again to obtain the secondary channel model Sz to complete the secondary channel modeling under the working state of the controlled device.

2. The secondary channel compensation modeling method in the device operation state according to claim 1, characterized in that: When establishing the mapping relationship, the reference signal x is obtained after processing the speed signal of the ship's propulsion system or other characteristic signals related to vibration noise. The primary response d is obtained through a vibration sensor installed at a specific position on the ship. In actual operation, the collected signal needs to be pre-processed by filtering, amplification, etc.

3. The secondary channel compensation modeling method in the device operation state according to claim 1, characterized in that: During the signal generation and mixing process, the actuator is an active vibration reduction device installed near the vibration source, such as a piezoelectric ceramic driver, and the actuator must be in a stable working state.

4. The secondary channel compensation modeling method in the device operation state according to claim 1, characterized in that: When predicting the primary response, the reference signal x needs to be updated in real time due to the changes in the operating conditions of the ship propulsion system. At the same time, the parameters of the LMS algorithm can be fine-tuned according to factors such as the ship's operating time and load changes to optimize the mapping relationship f(·).

5. The secondary channel compensation modeling method in the device operation state according to claim 1, characterized in that: When separating the secondary responses, the difference is used to calculate e s =e-d'=ef(x), separating the secondary response e from the mixed signal e s , and in actual calculations, digital signal processing technology is used to perform noise reduction, filtering and other processing on the signal to improve the accuracy of the separation results.

6. The secondary channel compensation modeling method in the device operation state according to claim 1, characterized in that: After completing the secondary channel modeling, the secondary channel model Sz is verified and evaluated by comparing the actual measured vibration and noise data with the model prediction data. If the secondary channel model accuracy does not meet the requirements, return to the previous steps to adjust the relevant parameters and re-model.

7. The secondary channel compensation modeling method in the device operation state according to claim 1, characterized in that: The method can realize modeling of the secondary channel while the device is in operation, without the need to frequently start and stop the device for modeling.

8. The secondary channel compensation modeling method in the device operation state according to claim 1, characterized in that: The method does not increase the calculation burden of the controller during the control process, and does not require the addition of additional noise.

Images