High-power fan hoisting control system and method based on low-temperature environment narrow site

Abstract

The invention belongs to the technical field of fan hoisting, and particularly relates to a high-power fan hoisting control system and method based on a low-temperature environment narrow site, and the method comprises the steps: collecting environment data of a construction site, including real-time temperature and parameters such as humidity and wind speed related to the temperature, building a high-precision temperature prediction model, and accurate temperature prediction of the construction environment is realized. Based on the temperature prediction, dynamic adjustment of hoisting power parameters can be implemented, intelligent planning and optimization of a hoisting path can be realized by avoiding a low-temperature sensitive area in the hoisting path, and efficient operation under low-temperature and narrow site conditions is ensured. By means of the comprehensive management mode, the safety and efficiency of operation under the extreme environment condition are remarkably improved, and performance reduction and misoperation of a power system caused by low temperature are effectively avoided.

Description

Technical Field

[0001] The present invention belongs to the technical field of fan hoisting, and in particular relates to a control system and method for hoisting a high-power fan in a small space in a low-temperature environment. Background Art

[0002] As the global demand for renewable energy increases, wind power generation has become a key means to reduce carbon emissions and achieve sustainable energy development. As the core equipment for wind power generation, the demand for wind turbine installation is growing. With the continuous increase in the capacity of single units, high-power wind turbines have gradually become the mainstream. However, many high-quality wind energy resources are concentrated in remote and harsh areas, such as mountains, plateaus and polar regions. This leads to complex construction challenges such as high altitude and small sites when hoisting high-power wind turbines in these areas.

[0003] In the prior art, wind turbine hoisting systems mostly focus on hoisting safety control under normal circumstances. For example, the Chinese invention patent with publication number CN113321126B proposes a wind turbine hoisting platform and a hoisting method thereof, which uses pressure sensors and infrared detection devices installed at the connection position between the hoisting device and the hoisted equipment, as well as a control system that receives data from these sensors to judge the safety of the hoisting process, and adjust the hoisting position to improve safety when necessary.

[0004] However, although the above scheme has achieved basic safety monitoring and dynamic adjustment through real-time data collection, it has not fully considered the impact of low temperature and narrow site conditions on lifting operations. Specifically, it is reflected in the following aspects: 1. Performance degradation of the power system in low temperature environment: The power system (such as hydraulic drive and motor control) that the fan lifting relies on in low temperature environment has significantly reduced performance due to problems such as increased viscosity of hydraulic oil and changes in motor winding resistance. If the lifting power is not adaptively adjusted according to the temperature, it is easy to cause frequent failures of the lifting equipment or even mechanical damage, increasing the risk of lifting failure.

[0005] 2. Limited hoisting paths in narrow sites: The hoisting paths of high-power wind turbines in narrow sites are already limited. Ground frost heave and deformation under low temperature conditions will further compress the moving path of the hoisting equipment, making the originally limited safety redundant space even tighter. If the low temperature factor is not fully considered when adjusting the hoisting position, the risk of collision and the difficulty of adjusting the hoisting position will be increased invisibly, ultimately leading to poor hoisting adjustment effects. Summary of the invention

[0006] The present invention aims to solve the deficiencies in the prior art and proposes a high-power wind turbine hoisting control system and method based on a small site in a low-temperature environment, focusing on optimizing the hoisting control of high-power wind turbines under conditions of low temperature and limited space to improve the safety and efficiency of operations.

[0007] The purpose of the present invention can be achieved through the following technical solutions: The first aspect of the present invention provides a high-power wind turbine hoisting control system based on a small site in a low-temperature environment, including the following modules: Data acquisition module: using environmental sensors and three-dimensional laser scanners to collect real-time temperature data of the construction site in a low-temperature environment, temperature-related environmental factor data and three-dimensional space point cloud data of the small site during the hoisting process.

[0008] Temperature prediction module: Build a temperature prediction model based on historical temperature time series data and historical environmental factor time series data related to temperature, and output the temperature prediction curve within the future set time window.

[0009] Power adjustment module: adjusts the power parameters of the lifting equipment according to the temperature prediction curve under the mapping relationship between temperature range and power parameters.

[0010] Path optimization module: Identify the distribution of obstacles on the construction site based on 3D point cloud data, generate an initial lifting path based on the size of the wind turbine, and dynamically optimize the path nodes according to the temperature prediction curve when performing the lifting operation according to the initial lifting path.

[0011] Warning execution module: During the power adjustment and path optimization process, it is configured to trigger a graded warning based on the comparison results between the predicted temperature and the warning threshold.

[0012] The second aspect of the present invention proposes a high-power wind turbine hoisting control method based on a small site in a low-temperature environment, comprising the following steps: Step 1: Collect real-time temperature data and environmental factor data in a low-temperature environment through environmental sensors deployed at the boundary of the hoisting site and key parts of the equipment, and use a three-dimensional laser scanner to obtain three-dimensional spatial point cloud data of the small site.

[0013] Step 2: Build a temperature prediction model based on the historical temperature time series data and related environmental factor data collected in step 1, and output the temperature prediction curve within the future set time window.

[0014] Step 3: According to the temperature prediction curve generated in step 2, the preset temperature range and power parameter mapping relationship are matched to adjust the power parameters of the lifting equipment.

[0015] Step 4: Based on the three-dimensional point cloud data obtained in step 1, identify the distribution of obstacles on the construction site and mark the low-temperature sensitive areas, and generate the initial lifting path in combination with the fan size parameters; allocate a dynamic buffer zone for the low-temperature sensitive area according to the temperature prediction curve in step 2, thereby dynamically optimizing the initial lifting path.

[0016] Step 5: During the power adjustment and route optimization process, it is configured to trigger a graded warning based on the comparison result between the predicted temperature and the warning threshold.

[0017] Combining all the above technical solutions, the positive effects of the present invention are as follows: 1. The present invention constructs an accurate temperature prediction model by collecting environmental data of the construction site to achieve accurate prediction of changes in construction environment temperature, and dynamically adjusts the lifting power parameters based on the temperature prediction, and optimizes the lifting path to avoid key nodes in low-temperature sensitive areas. This not only fills the shortcomings of the existing technology in low-temperature and small-site lifting scenarios, but also significantly improves the safety and efficiency of operations under extreme environmental conditions.

[0018] 2. The present invention can timely remind the hoisting operators to take control measures in advance by triggering a graded warning mechanism based on the predicted temperature, thereby significantly reducing the adverse effects of low temperature on the performance of the hoisting equipment. This not only effectively ensures the safety and reliability of the hoisting operation, but also greatly extends the service life of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0020] Figure 1 This is a schematic diagram of the system module structure provided in Example 1 of the present invention.

[0021] Figure 2 This is a flow chart of the implementation of the power adjustment module in the present invention.

[0022] Figure 3 This is a diagram of the method steps provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] Example 1

[0025] The present invention provides a high-power fan hoisting control system based on a small site in a low-temperature environment, comprising a data acquisition module, a temperature prediction module, a power adjustment module, a path optimization module and an early warning execution module.

[0026] See also Figure 1 As shown, the module composition and relationship are as follows: the data acquisition module provides real-time environmental data and detailed three-dimensional spatial information for the temperature prediction module.

[0027] The temperature prediction module uses real-time environmental data to generate a temperature prediction curve, which is then passed to the power adjustment module and the path optimization module for corresponding parameter adjustment and path optimization.

[0028] The power adjustment module outputs the adjusted power parameters which are directly applied to the lifting equipment and are also fed back to the early warning execution module for evaluating whether early warning actions need to be triggered.

[0029] The hoisting path optimized by the path optimization module is applied to the hoisting equipment to adjust the hoisting position, and is also fed back to the early warning execution module to evaluate whether it is necessary to trigger an early warning action.

[0030] The early warning execution module outputs an early warning signal to the operator or control system to guide the adoption of appropriate preventive measures or emergency responses.

[0031] The above modules work together to achieve comprehensive management and optimization of the hoisting operation of high-power wind turbines in low temperature and small space conditions, ensuring the safety and efficiency of the operation. Each module not only completes its specific tasks independently, but also supports each other through data sharing and feedback mechanisms, forming a highly integrated intelligent control system.

[0032] The data acquisition module uses environmental sensors and a three-dimensional laser scanner to collect temperature data of the construction site in a low-temperature environment, temperature-related environmental factor data, and three-dimensional spatial point cloud data of a small site in real time during the lifting process.

[0033] In the specific implementation of the above scheme, environmental sensors are deployed at the boundaries of the hoisting site and key parts of the equipment, integrating multiple sensors such as temperature sensors, humidity sensors, and wind speed sensors to comprehensively monitor the construction environment. Among them, the temperature sensor is specifically used to collect temperature data in a low-temperature environment, while the humidity sensor, wind speed sensor, etc. are responsible for obtaining environmental factor data such as humidity and wind speed related to temperature. In addition, the 3D laser scanner accurately scans the spatial structure of the construction site and generates detailed 3D spatial point cloud data. Through these comprehensive data, the system can achieve a comprehensive perception of the construction environment.

[0034] The temperature prediction module is used to build a temperature prediction model based on historical temperature time series data and historical environmental factor time series data related to temperature, and output a temperature prediction curve within a future set time window.

[0035] The specific implementation process of the above module is as follows: collect historical temperature time series data and historical environmental factor time series data of the area where the construction site is located.

[0036] For example, the historical environmental factor time series data includes, but is not limited to, historical humidity time series data, historical wind speed time series data, etc. The time series may be every hour or every ten minutes corresponding to every day.

[0037] It should be emphasized that when collecting historical temperature time series data and historical environmental factor time series data, it is necessary to cover data from the most recent multiple years as much as possible to provide a richer sample size. By analyzing data over a longer period of time, potential trends and patterns can be identified, thereby improving the accuracy of predictions about future temperature changes and reducing deviations caused by a single abnormal year.

[0038] The collected historical temperature time series data and historical environmental factor time series data are cleaned and standardized.

[0039] In the specific operation of the above scheme, the main purpose of data cleaning is to remove outliers and other data problems that may affect the accuracy of the model. Common outliers include extreme or unreasonable readings caused by sensor failure, transmission errors, or other unnatural phenomena. Standardization is to align the timestamps of the cleaned data and convert it into a standard format suitable for modeling to ensure data consistency and comparability.

[0040] Use time series analysis methods (such as Fourier transform, wavelet transform, etc.) to extract periodic change characteristics under low temperature environment from historical temperature time series data. For example, identify the temperature fluctuation pattern within a day (such as large temperature difference between morning and evening) and seasonal changes (such as long duration of low temperature in winter).

[0041] The correlation between temperature and environmental factors was analyzed by statistical methods, and the environmental factors that were correlated with temperature prediction in low temperature environments were screened out and recorded as correlated environmental factors.

[0042] As a specific implementation of the above scheme, statistical methods can use the Pearson correlation coefficient or machine learning models (such as linear regression) to quantify the strength of the relationship between temperature and various environmental factors (such as humidity, wind speed, etc.).

[0043] Furthermore, the screening of associated environmental factors can set a preset correlation threshold value to determine which environmental factors have a substantial impact on the temperature prediction in a low temperature environment. The correlation index between each environmental factor and temperature is compared with the correlation threshold value, and those environmental factors whose correlation index reaches or exceeds the preset threshold value are selected as associated environmental factors. These selected environmental factors represent the characteristic variables that have the greatest impact on the temperature prediction in a low temperature environment.

[0044] The above-mentioned correlation-based screening method not only improves the selection accuracy of model input features, but also reduces the impact of irrelevant or redundant features on model performance. By focusing on key associated environmental factors, a more concise and efficient temperature prediction model can be constructed, thereby improving its prediction accuracy and stability in complex low-temperature environments. This method ensures that the selected characteristic variables can reflect the main driving factors of temperature changes to the greatest extent, providing reliable temperature prediction support for high-power wind turbine hoisting operations.

[0045] Based on the periodic change characteristics in low temperature environment and related environmental factors, a prediction algorithm is selected to construct a temperature prediction model.

[0046] By incorporating the periodic change characteristics and related environmental factors in the low temperature environment into the data characteristic analysis when constructing the temperature prediction model, the most suitable prediction algorithm (such as autoregressive integrated moving average model, random forest, support vector machine, long short-term memory network) can be selected to improve the adaptability and prediction accuracy of the model.

[0047] The specific process of selecting a prediction algorithm is described in detail in existing literature and technical documents and will not be repeated here.

[0048] The historical data is divided into a training set and a test set. The training set is used to train the temperature prediction model. The input of the model includes the historical temperature sequence and its corresponding associated environmental factor sequence, and the output is the temperature prediction value within the target time period.

[0049] In the innovative implementation of the above solution, in order to adapt to the real-time changing environmental conditions, the rolling time window technology is used to dynamically update the model parameters. Specifically, the system will retrain the model at a fixed time interval (such as every 5 minutes), adding the latest temperature and related environmental factor data to the training set to improve the prediction accuracy.

[0050] After the model training is completed, the test set is used to validate it and the prediction performance is evaluated by calculating the prediction error.

[0051] The above-mentioned evaluation of prediction performance by calculating the prediction error can be evaluated using the mean square error and mean absolute error indicators. If the prediction error of the model is large, it indicates that its generalization ability and prediction accuracy are insufficient. At this time, it is necessary to further optimize the model structure or adjust the parameters. Specifically, the hyperparameters of the model (such as learning rate, number of hidden layer nodes, time step, etc.) can be tuned to improve the generalization ability and prediction accuracy of the model.

[0052] After the prediction performance is evaluated and met, the relevant environmental factors are screened out based on the temperature data and environmental factor data collected in real time during the lifting process, and are combined with the temperature data as input to the trained temperature prediction model.

[0053] The temperature prediction model generates a temperature prediction curve within a future set time window based on the input data.

[0054] It is important to note that since the installation of high-power wind turbines usually takes one to two weeks, significant temperature changes may occur during this period. Therefore, the time window set in the future should not be too long to avoid increased uncertainty and decreased accuracy caused by long-term predictions. For example, setting the time window to 2 hours can ensure that the temperature forecast has higher accuracy and reliability and reflects short-term temperature fluctuations in a timely manner.

[0055] It should be noted that relying solely on real-time temperature data for adjustment during wind turbine hoisting in a low-temperature environment may lead to response lag and operation delay, which in turn may cause problems such as equipment performance degradation, power system failure or hoisting path deviation, and will also expose the hoisting equipment to harsh environments for a long time. However, temperature prediction can identify and avoid adverse conditions in advance, optimize the hoisting path and power parameters, thereby effectively avoiding equipment in extreme environments, reducing operational risks and equipment damage, and improving the safety and efficiency of operations as well as the controllability and reliability of the overall construction process.

[0056] It should be pointed out that when predicting the temperature of the construction site, a dedicated temperature prediction model is chosen instead of directly using the data provided by the meteorological station. The main reason is that although the latter can provide a certain reference for the wind turbine hoisting operation, it may be insufficient in terms of refined management and safe operation. Specifically: the data from the meteorological station is usually based on the average value or grid forecast results of a larger area, which cannot accurately reflect the specific local microclimate conditions of the construction site. The construction site often requires higher resolution (such as hourly or shorter time intervals) and higher accuracy temperature forecasts to ensure the safety and efficiency of operations. In contrast, a dedicated temperature prediction model can be combined with the temperature data collected in real time on site for rolling updates and corrections to provide more accurate short-term forecasts. This method can not only capture sudden temperature changes, but also adjust the prediction frequency and accuracy according to actual needs, so as to better support the dynamic management and decision-making of the construction site. In this way, the reliability and applicability of temperature prediction can be significantly improved.

[0057] As an implementation example of the above temperature prediction, suppose a construction site is located in a cold northern region, the current time is 3 a.m. in winter, the ambient temperature is -15°C, the wind speed is 8m / s, and the humidity is 70%. The system predicts the temperature based on the following information: the temperature change pattern in the early morning hours is extracted from historical data, and it is found that the temperature drops by an average of 1°C per hour, but when the wind speed exceeds 5m / s, the cooling rate may accelerate to 1.5°C per hour.

[0058] Statistical results show that wind speed is one of the main environmental factors affecting temperature changes, while humidity has a smaller impact.

[0059] Based on historical data, the system trained an LSTM model and verified its prediction error (MSE=0.5℃) in the test set. The model performed well and met the prediction requirements.

[0060] The system takes the current temperature (-15℃) and wind speed (8m / s) as input, and combines the historical temperature drop pattern to generate a temperature forecast curve for the next 2 hours.

[0061] See also Figure 2 As shown, the power parameter adjustment module is used to adjust the power parameters of the lifting equipment according to the temperature prediction curve under the mapping relationship between the temperature range and the power parameter.

[0062] The specific implementation process of the above module is as follows: the operating temperature range is divided into multiple temperature intervals according to the working characteristics of the fan hoisting equipment.

[0063] For each temperature range, an appropriate mapping relationship between it and the dynamic parameters is established.

[0064] The above mentioned power parameters include but are not limited to starting current, operating frequency and hydraulic system pressure.

[0065] The appropriate mapping relationship between power parameters and temperature ranges is based on experimental data, manufacturer recommended values ​​or historical operating experience to ensure that the equipment can operate efficiently and stably under various temperature conditions.

[0066] In the example of the above implementation scheme, it is assumed that we divide the operating temperature range into the following three temperature intervals in advance and map appropriate power parameters for each interval, as shown in Table 1.

[0067] Table 1: Mapping relationship between some temperature ranges and suitable dynamic parameters

[0068]

[0069] The temperature data collected in real time during the lifting process and the temperature prediction curve within the future set time window provided by the temperature prediction module are integrated into the current expected temperature range.

[0070] As an example of the above solution, it is assumed that the real-time temperature collected at a certain time (T0) is -12°C.

[0071] The temperature prediction module generates a temperature prediction curve for the next 2 hours (set time window) based on historical data and current environmental conditions. The curve shows that at T0+30 minutes, the predicted temperature is -14°C.

[0072] At T0+60 minutes, the predicted temperature is -16°C.

[0073] At T0+90 minutes, the predicted temperature is -15℃.

[0074] At T0+120 minutes, the predicted temperature is -13°C.

[0075] Integrate the real-time temperature data with the temperature prediction curve within the future set time window to calculate the comprehensive temperature range covering the current and future period.

[0076] The current real-time temperature is -12℃.

[0077] The lowest predicted temperature in the next 2 hours is -16℃ and the highest predicted temperature is -13℃.

[0078] Therefore, the current expected temperature range is -16°C to -12°C.

[0079] The current expected temperature range is compared and analyzed with the divided temperature intervals. If the current predicted temperature range is completely within a single temperature interval, the current power parameters of the lifting equipment are adjusted according to the appropriate power parameters mapped to the temperature interval.

[0080] If the current temperature range spans multiple temperature intervals, these involved temperature intervals are recorded as covered temperature intervals.

[0081] The covering temperature interval corresponding to the lowest temperature among all the covering temperature intervals is selected as the severe temperature interval.

[0082] The temperature range and suitable power parameters applied to the above are mapping examples. Assuming that the current expected temperature range is -18°C to -28°C, comparing the current expected temperature range with the divided temperature range, it can be seen that it spans the two temperature ranges of -20°C to -10°C and -30°C to -20°C. In this case, the covering temperature range corresponding to the lowest temperature, i.e. -30°C to -20°C, is selected as the severe temperature range.

[0083] Based on the appropriate mapping relationship between temperature ranges and power parameters, the appropriate power parameters corresponding to the severe temperature range are extracted.

[0084] The current power parameters of the lifting equipment are adjusted according to the appropriate power parameters mapped in the severe temperature range.

[0085] It should be added that the above-mentioned adjustment of the current power parameters by using appropriate power parameters is intended to achieve a smooth transition from the current power parameters to the appropriate power parameters.

[0086] In the above solution, selecting appropriate power parameters in the harsh temperature range for adjustment can ensure that the equipment can still operate normally in the most severe operating environment. This method directly optimizes the most unfavorable situation and avoids the risks caused by underestimating the impact of the environment.

[0087] The present invention utilizes real-time temperature data and a temperature prediction curve within a future set time window to comprehensively consider current and expected temperature changes when performing hoisting operation adjustments, thereby making the adjustments more comprehensive and forward-looking. This helps to prepare for adjustments in advance, reducing downtime due to equipment failure or environmental factors, and at the same time maintaining optimal performance of the equipment under adverse conditions such as low temperature, thereby reducing failure rates and accident risks.

[0088] The path optimization module identifies the distribution of obstacles on the construction site based on the three-dimensional point cloud data, generates an initial lifting path based on the size of the wind turbine, and dynamically optimizes the path nodes according to the temperature prediction curve when performing the lifting operation according to the initial lifting path.

[0089] Preferably, the initial lifting path refers to the following generation process: a point cloud model of the construction site is constructed according to the three-dimensional space point cloud data of the narrow site.

[0090] In the specific implementation of the preferred scheme, the process of constructing the point cloud model of the construction site is as follows: the three-dimensional spatial point cloud data of the small site is pre-processed through point cloud processing technology (such as filtering, downsampling and smoothing, etc.); then, a surface reconstruction algorithm (such as Poisson reconstruction or triangulated mesh generation) is used to convert the point cloud data into a three-dimensional site point cloud model that accurately describes the terrain and structural characteristics of the construction site.

[0091] Based on the size parameters of the wind turbine (including height, diameter, blade length, etc.), the minimum safe envelope space for wind turbine hoisting is simulated in the point cloud model.

[0092] It is important to know that the minimum safe envelope space refers to the minimum safe envelope space required during the wind turbine hoisting process. This envelope space not only takes into account the physical size of the wind turbine itself, but also includes the additional clearance required to ensure safe operation to avoid collision with surrounding obstacles.

[0093] The geometric topological analysis method is used to identify the obstacle distribution in the point cloud model of the construction site containing the minimum safe envelope space of the wind turbine, thereby marking the passable area and obstacle area in the construction site.

[0094] It is important to understand that the geometric topological analysis method combines the principles of geometry and topology to analyze spatial structure and shape characteristics. Geometry focuses on the size, shape, and relative position of objects, while topology studies the continuity and connectivity of space. Through this combination, the geometric topological analysis method can effectively process point cloud models and extract useful spatial information from them.

[0095] In wind turbine hoisting operations, the point cloud model of the construction site contains detailed terrain and obstacle information. The geometric topological analysis method can be used to identify these obstacles. The specific identification process is described in detail in the existing technical literature and will not be repeated here.

[0096] In the specific implementation of the above solution, the process of extracting the passable area is to determine the passable area for free movement of the lifting equipment in the site point cloud model based on the location information of the obstacles.

[0097] The initial lifting path is generated by using the marked traversable area and obstacle area combined with the kinematic constraints of the lifting equipment (such as maximum steering angle, minimum turning radius and maximum climbing ability).

[0098] In the innovative implementation of the above solution, in order to cope with real-time changing environmental conditions (such as new obstacles), a dynamic adjustment mechanism is introduced, allowing the system to instantly update the path planning plan based on the latest sensor feedback to ensure the safety and efficiency of the lifting operation.

[0099] The above process constructs an accurate site point cloud model, simulates the minimum safe envelope space for wind turbine hoisting, and uses advanced geometric topology analysis methods to identify the traversable area and obstacle distribution. Finally, the kinematic constraints of the hoisting equipment are combined to generate the optimal initial hoisting path. This not only improves the safety and success rate of the hoisting operation, but also enhances the flexibility and adaptability of the system, and can effectively cope with the complex and changeable construction site environment.

[0100] Further preferably, dynamically optimizing the path nodes according to the temperature prediction curve includes the following: marking low temperature sensitive areas in the point cloud model of the construction site.

[0101] The above-mentioned medium and low temperature sensitive area labeling process is as follows: collect historical performance failure records of the hoisting equipment operating in a low temperature environment, and extract the location where the hoisting equipment failure occurs from the historical performance failure records.

[0102] It should be pointed out that the historical performance failure records mentioned above refer to the failure event records caused by performance degradation of the lifting equipment. The low-temperature sensitive areas are marked with historical performance failure records because the historical performance failure records provide the actual operating performance of the equipment under specific environmental conditions (such as low temperature). These data reflect the performance and potential problems of the equipment in the real operating environment. Compared with theoretical analysis or simulation, actual failure records can more accurately reflect which areas are prone to performance degradation or failure, thereby improving the accuracy of marking.

[0103] The location of the lifting equipment failure in each historical performance failure record is classified into a category using a spatial clustering algorithm to form low-temperature sensitive areas. These areas may have a significant impact on the operation of the lifting equipment in a low-temperature environment due to terrain characteristics (such as water accumulation and freezing in low-lying areas).

[0104] Use visualization tools to mark low-temperature sensitive areas in the construction site point cloud model.

[0105] In particular, when labeling low-temperature sensitive areas, the labeling accuracy of low-temperature sensitive areas can be optimized through iterative historical lifting data.

[0106] For each low-temperature sensitive area, its proportion in all historical performance failure records is calculated, and a priority label is assigned to each low-temperature sensitive area based on the proportion.

[0107] The specific example of assigning the above-mentioned medium priority label is as follows: different priority labels are divided according to the threshold value set according to the calculated proportion. For example, the area with a proportion greater than a higher threshold (such as 50%) is marked as high priority.

[0108] Regions with percentages between medium thresholds (e.g., 10%-50%) are marked as medium priority.

[0109] Regions with a percentage below a lower threshold (such as less than 10%) are marked as low priority.

[0110] It should be explained that the priority labels assigned to the low-temperature sensitive areas above reflect the risk levels of the low-temperature sensitive areas. Low-temperature sensitive areas with higher priorities usually mean that a larger proportion of equipment failures occur in these areas, indicating that these areas are at higher risk.

[0111] Allocate buffer zones for low-temperature sensitive areas in path planning based on temperature prediction curves.

[0112] It should be pointed out that the buffer zone actually refers to an additional space area set around the low-temperature sensitive area. Its main purpose is to provide additional safety margin for the lifting path. Even if there are small operational errors or environmental changes, it can prevent the equipment from directly entering the low-temperature sensitive area when performing lifting tasks, thereby reducing potential risks.

[0113] The specific allocation process is as follows: for each low-temperature sensitive area, an initial buffer is allocated according to the priority label.

[0114] The initial buffer allocation operation is as follows: 1) Set a basic buffer width as the starting value for all low temperature sensitive areas. This width can be determined based on actual requirements such as device size and operating error range, and then define an expansion factor to adjust the buffer width based on priority.

[0115] 2) For low temperature sensitive areas marked as high priority, a larger expansion factor (such as 2 times or greater) is applied, and the base buffer width is multiplied by the expansion factor to obtain the final buffer width.

[0116] For example, if the base buffer width is 5 meters and the expansion factor is 2, the buffer width for the high priority area will be 10 meters.

[0117] 3) For low temperature sensitive areas marked as medium priority, a medium-sized expansion factor (such as 1.5 times) is applied, and the base buffer width is multiplied by the expansion factor to obtain the final initial buffer width.

[0118] For example, if the base buffer width is 5 meters and the expansion factor is 1.5, the initial buffer width for medium priority areas will be 7.5 meters.

[0119] 4) For low-temperature sensitive areas marked as low priority, a smaller expansion factor (such as 1 or slightly greater than 1) is applied, and the base buffer width is multiplied by the expansion factor to obtain the final initial buffer width.

[0120] For example, if the base buffer width is 5 meters and the expansion factor is 1.2, the initial buffer width for low priority areas will be 6 meters.

[0121] The dynamic adjustment trigger conditions are set at the buffer zone boundary as follows: a) when the predicted temperature obtained according to the temperature prediction curve is higher than the primary warning threshold, the initial buffer zone is reduced according to the set ratio.

[0122] It is understandable that when the predicted temperature is lower than the primary warning threshold, it means that the current ambient temperature is relatively high and has not reached the low temperature range that requires an early warning. In this case, the equipment performance is less likely to be affected by low temperatures, the risk is low, and there is no need to expand the initial buffer zone. The initial buffer zone can be reduced, for example, the basic buffer zone can be maintained. Reducing the buffer zone helps improve the space utilization of the hoisting operation and reduce the path restriction problem caused by an excessively large buffer zone.

[0123] b) Maintaining the initial buffer zone when the forecast temperature is below the primary alert threshold and above the intermediate alert threshold.

[0124] It is understandable that when the predicted temperature is lower than the primary warning threshold but higher than the intermediate warning threshold, it means that the ambient temperature has entered the low temperature range but has not yet reached a higher risk level. At this time, the equipment performance may begin to be affected to a certain extent, but it is still within the controllable range. Within this temperature range, further reducing or expanding the buffer zone may lead to unnecessary operational complexity or waste of resources.

[0125] c) When the predicted temperature is lower than the intermediate warning threshold and higher than the advanced warning threshold, the initial buffer zone is expanded according to the set ratio.

[0126] It is understandable that when the predicted temperature is lower than the medium warning threshold but higher than the high warning threshold, it means that the ambient temperature has entered the low temperature high risk range. At this time, the performance of the equipment may be significantly affected and the risk of failure is significantly increased. The expansion of the buffer zone is to provide a larger safety margin around the low temperature sensitive area to reduce the possibility of equipment being exposed to a high risk environment.

[0127] The primary warning threshold, intermediate warning threshold and advanced warning threshold mentioned above are mentioned in the early warning execution module, and the primary warning threshold>intermediate warning threshold>advanced warning threshold.

[0128] It should be added that the situation where the predicted temperature is lower than the advanced warning threshold is not considered in the buffer allocation process. The reason is that when the predicted temperature drops below the threshold, the warning execution module will trigger the advanced warning and perform emergency braking measures. At this time, the lifting operation is already suspended, so there is no need to design an additional buffer for this situation. This processing method ensures that the system's safety mechanism can prioritize the safety of equipment and personnel under extreme low temperature conditions and avoid risks caused by continued operation.

[0129] According to the above operations, it can be seen that as the predicted temperature value decreases, the buffer range is appropriately expanded to increase the safety margin. Through real-time updated temperature forecast information, the size of the buffer zone around the low-temperature sensitive area is dynamically adjusted to ensure that it always adapts to the current and expected temperature conditions.

[0130] Identify path nodes on the initial lifting path that are located in or near low temperature sensitive buffer zones.

[0131] When the temperature forecast shows that the temperature within a future set time window is lower than the set warning threshold, the turning radius is shortened for the route nodes located in or close to the low temperature sensitive buffer zone.

[0132] The reason for shortening the turning radius for the path nodes located at or near the low-temperature sensitive buffer zone mentioned above is mainly because the low-temperature sensitive buffer zone often contains higher risk factors. Shortening the turning radius can help the equipment better avoid these risk factors, which helps to reduce the time that the equipment is exposed to high-risk areas for a long time under low temperature conditions, thereby reducing the risk of equipment performance degradation. In addition, a smaller turning radius usually means that the torque and power required by the equipment during the turning process are lower, thereby reducing the burden on the power system.

[0133] It should be noted that the warning thresholds mentioned above are primary warning thresholds. When the temperature forecast shows that the temperature will be lower than the primary warning threshold in the future, it indicates that the equipment performance may begin to be affected, although it has not yet reached a high-risk state. Taking preventive measures at this time (such as shortening the turning radius) can avoid potential risks in advance and prevent the situation from deteriorating to the point where a higher-level response is required. In addition, taking overly conservative measures too early (such as drastically adjusting the route when the temperature has not dropped significantly) may lead to unnecessary waste of resources and increased operational complexity.

[0134] Preferably, the smoothness and continuity of the entire path should be ensured while shortening the turning radius to avoid operational difficulties or other safety hazards caused by sudden changes in the path.

[0135] The warning execution module is configured to trigger a graded warning according to a comparison result between the predicted temperature and the warning threshold during the power adjustment and path optimization process.

[0136] The specific operations for triggering the above-mentioned graded warning are as follows: when the predicted temperature is lower than the primary warning threshold, the primary warning is triggered and the equipment status check instruction is pushed to the visualization interface.

[0137] When the predicted temperature is lower than the intermediate warning threshold, an intermediate warning is triggered and the lifting speed is automatically reduced.

[0138] When the predicted temperature exceeds the advanced warning threshold, an advanced warning is triggered and emergency braking is performed.

[0139] It should be pointed out that the reason for setting different temperature warning thresholds is that the power system of the lifting equipment exhibits different performance characteristics under low temperature conditions. In particular, the hydraulic oil in the hydraulic system is affected by the low temperature, and the viscosity of the hydraulic oil increases significantly, resulting in an increase in the flow resistance of the hydraulic system and causing it to exhibit different physical properties. The hydraulic oil in the hydraulic system of the lifting equipment can be tested under a low temperature temperature gradient to obtain the inflection point temperature where the equipment performance changes significantly. These inflection point temperatures can be used as the basis for dividing different warning thresholds; in addition, based on the performance parameters of the hydraulic oil at different temperatures in the technical manual provided when the lifting equipment leaves the factory, these key parameters can be extracted as the basis for dividing different warning thresholds to ensure that the early warning mechanism can respond in time when the performance of the hydraulic system begins to be affected.

[0140] The present invention can timely remind the hoisting operators to take control measures in advance by triggering a graded warning mechanism based on the predicted temperature, thereby significantly reducing the adverse effects of low temperature on the performance of the hoisting equipment. This not only effectively ensures the safety and reliability of the hoisting operation, but also greatly extends the service life of the equipment.

[0141] It should be understood that the main reason for choosing to trigger the graded warning based on the predicted temperature rather than the real-time temperature is that the real-time temperature reflects the environmental conditions at the current moment, and the adjustment of the speed and position of the lifting usually takes a certain amount of time to complete. If the warning operation is performed only relying on the real-time temperature, it may cause the system to be unable to take effective measures in time when the problem is identified. Performing warning operations based on the predicted temperature can provide sufficient preparation time for these operations, ensuring that the system can complete the necessary adjustments before the actual temperature reaches the dangerous range, thereby improving the system's foresight and response efficiency. In addition, the real-time temperature may cause false alarms or missed alarms due to local fluctuations or measurement errors, and the predicted temperature can effectively reduce the occurrence of this situation through trend modeling and analysis, thereby improving the stability of the early warning system.

[0142] Example 2

[0143] See also Figure 3 As shown, the present invention proposes a high-power wind turbine hoisting control method based on a small site in a low-temperature environment, comprising the following steps: Step 1: Collect real-time temperature data and environmental factor data in a low-temperature environment through environmental sensors deployed at the boundary of the hoisting site and key parts of the equipment, and use a three-dimensional laser scanner to obtain three-dimensional spatial point cloud data of the small site.

[0144] Step 2: Build a temperature prediction model based on the historical temperature time series data and related environmental factor data collected in step 1, and output the temperature prediction curve within the future set time window.

[0145] Step 3: According to the temperature prediction curve generated in step 2, the preset temperature range and power parameter mapping relationship are matched to adjust the power parameters of the lifting equipment.

[0146] Step 4: Based on the three-dimensional point cloud data obtained in step 1, identify the distribution of obstacles on the construction site and mark the low-temperature sensitive areas, and generate the initial lifting path in combination with the fan size parameters; allocate a dynamic buffer zone for the low-temperature sensitive area according to the temperature prediction curve in step 2, thereby dynamically optimizing the initial lifting path.

[0147] Step 5: During the power adjustment and path adjustment process, it is configured to trigger a graded warning based on the comparison result between the predicted temperature and the warning threshold.

[0148] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A high-power fan hoisting control system based on a small site in a low-temperature environment, characterized by: Includes the following modules: Data acquisition module: Use environmental sensors and 3D laser scanners to collect temperature data of the construction site in a low-temperature environment, temperature-related environmental factor data, and 3D spatial point cloud data of a small site in real time during the hoisting process; Temperature prediction module: builds a temperature prediction model based on historical temperature time series data and historical environmental factor time series data related to temperature, and outputs the temperature prediction curve within the future set time window; Power adjustment module: adjusts the power parameters of the lifting equipment according to the temperature prediction curve under the mapping relationship between temperature range and power parameters; Path optimization module: Identify the distribution of obstacles on the construction site based on 3D point cloud data, generate the initial hoisting path based on the size of the wind turbine, and dynamically optimize the path nodes according to the temperature prediction curve when performing the hoisting operation according to the initial hoisting path; Warning execution module: During the power adjustment and path optimization process, it is configured to trigger a graded warning based on the comparison results between the predicted temperature and the warning threshold.

2. The high-power fan hoisting control system based on a small site in a low-temperature environment as claimed in claim 1 is characterized by: The specific implementation process of the temperature prediction module is as follows: Collect historical temperature time series data and historical environmental factor time series data related to temperature in the area where the construction site is located; Clean and standardize the collected historical temperature time series data and historical environmental factor time series data; Extract the periodic variation characteristics under low temperature environment from the historical temperature time series data; The correlation between temperature and environmental factors is analyzed by statistical methods, thereby selecting environmental factors that are correlated with temperature prediction in low temperature environments and recording them as correlated environmental factors; Based on the periodic variation characteristics and related environmental factors under low temperature environment, a prediction algorithm is selected to build a temperature prediction model; The historical data is divided into a training set and a test set. The training set is used to train the temperature prediction model. The input of the model includes the historical temperature sequence and its corresponding associated environmental factor sequence, and the output is the temperature prediction value within the target time period. After the model training is completed, the test set is used to validate it and the prediction performance is evaluated by calculating the prediction error; After the prediction performance is evaluated and meets the standards, the relevant environmental factors are selected based on the temperature data and environmental factor data collected in real time during the lifting process, and are combined with the temperature data as input to the trained temperature prediction model; The temperature prediction model generates a temperature prediction curve within a future set time window based on the input data.

3. The high-power fan hoisting control system based on a small site in a low-temperature environment as claimed in claim 1 is characterized by: The implementation process of the power adjustment module is as follows: The operating temperature range is divided into multiple temperature intervals according to the working characteristics of the fan hoisting equipment; Establish a suitable mapping relationship between each temperature range and the dynamic parameters; The temperature data collected in real time during the hoisting process and the temperature prediction curve within the future set time window provided by the temperature prediction module are integrated into the current expected temperature range; The current expected temperature range is compared and analyzed with the divided temperature intervals. If the current predicted temperature range is completely within a single temperature interval, the current power parameters of the lifting equipment are adjusted according to the appropriate power parameters mapped to the temperature interval.

4. The high-power fan hoisting control system based on a small site in a low-temperature environment as claimed in claim 3 is characterized by: The power adjustment module also includes the following contents: If the current expected temperature range spans multiple temperature intervals, these involved temperature intervals are recorded as covered temperature intervals; Selecting the covering temperature interval corresponding to the lowest temperature among all covering temperature intervals as the severe temperature interval; Extract appropriate power parameters corresponding to the severe temperature range based on the appropriate mapping relationship between the temperature range and the power parameters; The current power parameters of the lifting equipment are adjusted according to the appropriate power parameters mapped in the severe temperature range.

5. The high-power fan hoisting control system based on a small site in a low-temperature environment as claimed in claim 1 is characterized by: The generation of the initial lifting path refers to the following process: Construct a point cloud model of the construction site based on the three-dimensional point cloud data of the narrow site; Based on the size of the wind turbine, the minimum safe envelope space for wind turbine hoisting is simulated in the point cloud model; The geometric topological analysis method is used to identify the obstacle distribution of the construction site point cloud model containing the minimum safe envelope space of the wind turbine, thereby marking the passable area and obstacle area in the construction site. The initial lifting path is generated by combining the marked traversable area and obstacle area with the kinematic constraints of the lifting equipment.

6. The high-power fan hoisting control system based on a small site in a low-temperature environment as claimed in claim 1, characterized in that: The dynamic optimization of the path nodes according to the temperature prediction curve includes the following contents: Mark low-temperature sensitive areas in the point cloud model of the construction site; Allocate buffer zones for low-temperature sensitive areas in route planning based on temperature prediction curves; Identify path nodes on the initial lifting path that are located in or near low-temperature sensitive buffer zones; When the temperature forecast shows that the temperature within a future set time window is lower than the set warning threshold, the turning radius is shortened for the route nodes located in or close to the low temperature sensitive buffer zone.

7. The high-power fan hoisting control system based on a small site in a low-temperature environment as claimed in claim 6, characterized in that: The process of marking low temperature sensitive areas in the construction site point cloud model is as follows: Collect historical performance failure records of hoisting equipment operating in low temperature environments, and extract the locations where hoisting equipment failures occurred from the historical performance failure records; The location of the hoisting equipment failure in each historical performance failure record is classified into a low-temperature sensitive area using a spatial clustering algorithm; Use visualization tools to mark low-temperature sensitive areas in the construction site cloud model; For each low-temperature sensitive area, its proportion in all historical performance failure records is calculated, and a priority label is assigned to each low-temperature sensitive area based on the proportion.

8. The high-power fan hoisting control system based on a small site in a low-temperature environment as claimed in claim 7, characterized in that: The triggering of graded warning includes the following process: When the predicted temperature is lower than the primary warning threshold, a primary warning is triggered and the device status check instruction is pushed to the visualization interface; When the predicted temperature is lower than the intermediate warning threshold, an intermediate warning is triggered and the hoisting speed is automatically reduced; When the predicted temperature exceeds the advanced warning threshold, an advanced warning is triggered and emergency braking is performed.

9. The high-power fan hoisting control system based on a small site in a low-temperature environment as claimed in claim 8, characterized in that: The buffer includes the following allocation process: For each low-temperature sensitive area, an initial buffer is allocated according to the priority label; The trigger conditions for dynamic adjustment at the buffer boundary are as follows: a) When the predicted temperature obtained according to the temperature prediction curve is higher than the primary warning threshold, the initial buffer zone is reduced according to the set ratio; b) maintaining the initial buffer zone when the forecast temperature is below the primary alert threshold and above the intermediate alert threshold; c) When the predicted temperature is lower than the intermediate warning threshold and higher than the advanced warning threshold, the initial buffer zone is expanded according to the set ratio.

10. A high-power fan hoisting control method based on a small site in a low-temperature environment, characterized in that: The steps include: Step 1: Use environmental sensors deployed at the edge of the hoisting site and key parts of the equipment to collect real-time temperature data and environmental factor data in a low-temperature environment, and use a 3D laser scanner to obtain 3D spatial point cloud data of a small site; Step 2: Build a temperature prediction model based on the historical temperature time series data and related environmental factor data collected in step 1, and output the temperature prediction curve within the future set time window; Step 3: Match the preset temperature range and power parameter mapping relationship according to the temperature prediction curve generated in step 2, so as to adjust the power parameters of the lifting equipment; Step 4: Based on the three-dimensional point cloud data obtained in step 1, identify the distribution of obstacles on the construction site and mark the low-temperature sensitive areas, and generate the initial lifting path in combination with the size parameters of the wind turbine; allocate a dynamic buffer zone for the low-temperature sensitive area according to the temperature prediction curve in step 2, thereby dynamically optimizing the initial lifting path; Step 5: During the power adjustment and route optimization process, it is configured to trigger a graded warning based on the comparison result between the predicted temperature and the warning threshold.

Images