In the complex and variable deep-sea environment,the compensation control of ship motion ensures the safety and efficiency of equipment installation and transportation in offshore wind farms.However,the ship motion po...In the complex and variable deep-sea environment,the compensation control of ship motion ensures the safety and efficiency of equipment installation and transportation in offshore wind farms.However,the ship motion posture compensation control system is severely affected by uncertainties,which significantly impact the accuracy of compensation control.In this paper,we propose a ship three-degree-of-freedom(3-DoF)motion posture stabilization control method based on the DTW-LSTM-MATD3 algorithm.We use the multi-agent twin delayed deep deterministic policy gradient(MATD3)to control a platform with six electric cylinders to achieve stable control.However,owing to random noise affecting the ship’s motion posture,we use a dynamic time warping(DTW)algorithm to distinguish between high-frequency noise and low-frequency tracking signals.Further,we embed a long short-term memory(LSTM)network into the MATD3 network to better align the Critic network’s training with the true Q-value.We use a combined reward function to enhance the agent’s exploration capability in complex dynamic environments.Finally,verification was conducted under sixth-level,abrupt sea conditions with high-frequency noise,as well as under real abrupt sea conditions,and a generalization test was also carried out.Simulation results show that the proposed DTW-LSTM-MATD3 method has great compensation control ability.展开更多
To address the shortcomings of existing fault diagnosis models for aero-engine gas path components,such as weak feature extraction capabilities and low diagnostic accuracy,this study proposes a fault diagnosis model u...To address the shortcomings of existing fault diagnosis models for aero-engine gas path components,such as weak feature extraction capabilities and low diagnostic accuracy,this study proposes a fault diagnosis model underpinned by the Fungal Growth Optimization(FGO)algorithm(FGO-1 DCNN-LSTM).This model integrates a 1D convolutional neural network(1 DCNN)and a long short-term memory network(LSTM).LSTM's strength in extracting temporal features compensates for the limitations of 1 DCNN in processing timeseries data.A split-path convolutional fusion module is introduced into the 1 DCNN,enabling parallel input of sensor data,thus enhancing both the network's extraction capabilities and efficiency.Simultaneously,the FGO algorithm is incorporated to tackle the issue of hyperparameter optimization.To validate the model's performance,training and validation experiments were conducted using the N-CMAPSS dataset,which covers fault data under three typical operating conditions:low-altitude low-velocity,higher-altitude and highervelocity,and high-altitude high-velocity.Experimental results indicate that the FGO-1 DCNNLSTM model attains fault diagnosis accuracies of 90.74%,91.67%,and 94.44%under three operating condition.In comparison with the unoptimized 1 DCNN-LSTM model,its diagnostic accuracy is improved by 2.78%,10.19%,and 3.7%,respectively.The results provide evidence that the FGO-1 DCNN-LSTM model can effectively achieve accurate identification and diagnosis of faults in aero-engine gas path components.展开更多
针对目前电池荷电状态(stage of charge,SOC)估计算法存在稳定性差、误差大等缺点,提出一种基于实车云端放电数据的自适应扩展卡尔曼滤波(adaptive extended Kalman filter,AEKF)与长短时记忆(long short term memory,LSTM)融合的算法,...针对目前电池荷电状态(stage of charge,SOC)估计算法存在稳定性差、误差大等缺点,提出一种基于实车云端放电数据的自适应扩展卡尔曼滤波(adaptive extended Kalman filter,AEKF)与长短时记忆(long short term memory,LSTM)融合的算法,预测小动力电动车的电池SOC。首先采用自适应遗忘因子最小二乘法(adaptive forgetting factor recursive least squares,AFFRLS)辨识电池的二阶RC等效电路模型参数。其次,将云端实时采集到的放电数据作为研究目标,通过AEKF-LSTM融合算法对小动力电动车的电池SOC进行预测实验,实验过程中AEKF-LSTM融合算法将当前时刻的端电压、电流、温度以及上一时刻电池的SOC作为输入,以更新的SOC作为输出训练估计模型。最后,将AEKF-LSTM融合算法和单一AEKF算法预测电池SOC的结果与实际SOC值进行比较,实验结果表明,AEKF-LSTM融合算法的均方根误差(root mean square error,RMSE)为0.0058 V,平均绝对误差(mean absolute error,MAE)为0.0041 V,比AEKF算法的RMSE减小0.0087 V,MAE减小0.1164 V,且AEKF-LSTM融合算法的RMSE和MAE均在0.6%以内,证明了该融合算法有较高的估计精度和较强的鲁棒性。展开更多
基金supported by the National Natural Science Foundation of China(No.52105466).
摘要In the complex and variable deep-sea environment,the compensation control of ship motion ensures the safety and efficiency of equipment installation and transportation in offshore wind farms.However,the ship motion posture compensation control system is severely affected by uncertainties,which significantly impact the accuracy of compensation control.In this paper,we propose a ship three-degree-of-freedom(3-DoF)motion posture stabilization control method based on the DTW-LSTM-MATD3 algorithm.We use the multi-agent twin delayed deep deterministic policy gradient(MATD3)to control a platform with six electric cylinders to achieve stable control.However,owing to random noise affecting the ship’s motion posture,we use a dynamic time warping(DTW)algorithm to distinguish between high-frequency noise and low-frequency tracking signals.Further,we embed a long short-term memory(LSTM)network into the MATD3 network to better align the Critic network’s training with the true Q-value.We use a combined reward function to enhance the agent’s exploration capability in complex dynamic environments.Finally,verification was conducted under sixth-level,abrupt sea conditions with high-frequency noise,as well as under real abrupt sea conditions,and a generalization test was also carried out.Simulation results show that the proposed DTW-LSTM-MATD3 method has great compensation control ability.
基金supported by the National Natural Science Foundation of China under Grant nos.62301341,62371315 and 62531017the Liaoning Provincial Department of Science and Technology(Applied Basic Research)under Grant nos.2025JH2/101300013the Youth Project of Liaoning Provincial Department of Education under Grant nos.200080762/070。
摘要To address the shortcomings of existing fault diagnosis models for aero-engine gas path components,such as weak feature extraction capabilities and low diagnostic accuracy,this study proposes a fault diagnosis model underpinned by the Fungal Growth Optimization(FGO)algorithm(FGO-1 DCNN-LSTM).This model integrates a 1D convolutional neural network(1 DCNN)and a long short-term memory network(LSTM).LSTM's strength in extracting temporal features compensates for the limitations of 1 DCNN in processing timeseries data.A split-path convolutional fusion module is introduced into the 1 DCNN,enabling parallel input of sensor data,thus enhancing both the network's extraction capabilities and efficiency.Simultaneously,the FGO algorithm is incorporated to tackle the issue of hyperparameter optimization.To validate the model's performance,training and validation experiments were conducted using the N-CMAPSS dataset,which covers fault data under three typical operating conditions:low-altitude low-velocity,higher-altitude and highervelocity,and high-altitude high-velocity.Experimental results indicate that the FGO-1 DCNNLSTM model attains fault diagnosis accuracies of 90.74%,91.67%,and 94.44%under three operating condition.In comparison with the unoptimized 1 DCNN-LSTM model,its diagnostic accuracy is improved by 2.78%,10.19%,and 3.7%,respectively.The results provide evidence that the FGO-1 DCNN-LSTM model can effectively achieve accurate identification and diagnosis of faults in aero-engine gas path components.
摘要针对目前电池荷电状态(stage of charge,SOC)估计算法存在稳定性差、误差大等缺点,提出一种基于实车云端放电数据的自适应扩展卡尔曼滤波(adaptive extended Kalman filter,AEKF)与长短时记忆(long short term memory,LSTM)融合的算法,预测小动力电动车的电池SOC。首先采用自适应遗忘因子最小二乘法(adaptive forgetting factor recursive least squares,AFFRLS)辨识电池的二阶RC等效电路模型参数。其次,将云端实时采集到的放电数据作为研究目标,通过AEKF-LSTM融合算法对小动力电动车的电池SOC进行预测实验,实验过程中AEKF-LSTM融合算法将当前时刻的端电压、电流、温度以及上一时刻电池的SOC作为输入,以更新的SOC作为输出训练估计模型。最后,将AEKF-LSTM融合算法和单一AEKF算法预测电池SOC的结果与实际SOC值进行比较,实验结果表明,AEKF-LSTM融合算法的均方根误差(root mean square error,RMSE)为0.0058 V,平均绝对误差(mean absolute error,MAE)为0.0041 V,比AEKF算法的RMSE减小0.0087 V,MAE减小0.1164 V,且AEKF-LSTM融合算法的RMSE和MAE均在0.6%以内,证明了该融合算法有较高的估计精度和较强的鲁棒性。