A platform for research: civil engineering, architecture and urbanism
Full-intelligent cutter suction dredger
The invention relates to a full-intelligent cutter suction dredger. The full-intelligent cutter suction dredger is characterized in that on the basis of the embodiment 1, a traditional cutter suctiondredger is transformed or the cutter suction dredger is newly built through electricity and informatization to form a control system from top to bottom and an information acquisition system from bottom to top, meanwhile, electrical operation state data and yield data in the service process are obtained through the acquisition system and used for forming a sample database, a neural network online prediction model is constructed, and the design of "input of five control variables and output of one yield" serves as a mathematical model of a neural network, the mathematical model is updated afterthe neural network is trained by using a sample, when a ship body control variable at the current moment is input, a predicted yield value is output through online solving of the neural network, the optimal control parameter is output by using an optimizer optimization algorithm, and a control parameter is continuously close to the optimal process control parameter in an ideal state after iteration, so that intelligent operation can be realized, and the optimal yield value can be maintained.
一种全智能绞吸挖泥船,其特征在于,基于实施例1,通过电气和信息化改造传统绞吸挖泥船或者新建绞吸挖泥船,以形成自上而下的控制系统和自下而上信息采集系统;同时,通过采集系统获得业务过程中的电气作业状态数据及产量数据,用于形成样本数据库;构建神经网络在线预测模型,以“五个控制变量输入,一个产量输出”设计为神经网络的数学模型;用样本训练神经网络后更新数学模型,当输入当前时刻船体控制变量,经神经网络在线求解输出预测产量值,采用优化器寻优算法输出最佳控制参数,迭代后不断接近理想状态的最佳工艺控制参数,实现智能作业且保持最优产值。
Full-intelligent cutter suction dredger
The invention relates to a full-intelligent cutter suction dredger. The full-intelligent cutter suction dredger is characterized in that on the basis of the embodiment 1, a traditional cutter suctiondredger is transformed or the cutter suction dredger is newly built through electricity and informatization to form a control system from top to bottom and an information acquisition system from bottom to top, meanwhile, electrical operation state data and yield data in the service process are obtained through the acquisition system and used for forming a sample database, a neural network online prediction model is constructed, and the design of "input of five control variables and output of one yield" serves as a mathematical model of a neural network, the mathematical model is updated afterthe neural network is trained by using a sample, when a ship body control variable at the current moment is input, a predicted yield value is output through online solving of the neural network, the optimal control parameter is output by using an optimizer optimization algorithm, and a control parameter is continuously close to the optimal process control parameter in an ideal state after iteration, so that intelligent operation can be realized, and the optimal yield value can be maintained.
一种全智能绞吸挖泥船,其特征在于,基于实施例1,通过电气和信息化改造传统绞吸挖泥船或者新建绞吸挖泥船,以形成自上而下的控制系统和自下而上信息采集系统;同时,通过采集系统获得业务过程中的电气作业状态数据及产量数据,用于形成样本数据库;构建神经网络在线预测模型,以“五个控制变量输入,一个产量输出”设计为神经网络的数学模型;用样本训练神经网络后更新数学模型,当输入当前时刻船体控制变量,经神经网络在线求解输出预测产量值,采用优化器寻优算法输出最佳控制参数,迭代后不断接近理想状态的最佳工艺控制参数,实现智能作业且保持最优产值。
Full-intelligent cutter suction dredger
一种全智能绞吸挖泥船
ZHANG QINGBO (author) / DAI WENBO (author) / HOU XIAOMING (author) / XIAO YE (author) / LIN TING (author) / CHEN XINHUA (author) / FENG BO (author) / HU JINGZHAO (author) / WANG HAIRONG (author) / CUI PENGFEI (author)
2020-06-26
Patent
Electronic Resource
Chinese
IPC:
E02F
Baggern
,
DREDGING
/
G01D
MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE
,
Anzeigen oder Aufzeichnen in Verbindung mit Messen allgemein
/
G05B
Steuer- oder Regelsysteme allgemein
,
CONTROL OR REGULATING SYSTEMS IN GENERAL
/
G06N
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
,
Rechnersysteme, basierend auf spezifischen Rechenmodellen
European Patent Office | 2024
|British Library Online Contents | 1995
|Full-automatic control dredging and cutter-suction dredger
European Patent Office | 2020
|