Chemical Industry and Engineering Progress ›› 2025, Vol. 44 ›› Issue (10): 5673-5688.DOI: 10.16085/j.issn.1000-6613.2024-1284
• Energy processes and technology • Previous Articles
WANG Wenyang1,2(
), LUO Yuping1, YU Jiahuan1, ZHOU Jibin2, YE Mao2(
), LIU Zhongmin2
Received:2024-08-05
Revised:2024-11-12
Online:2025-11-10
Published:2025-10-25
Contact:
YE Mao
王文洋1,2(
), 罗玉平1, 余佳洹1, 周吉彬2, 叶茂2(
), 刘中民2
通讯作者:
叶茂
作者简介:王文洋(1991—),副教授,硕士生导师,研究方向为统计分析、机器学习与人工智能等。E-mail:wangwenyang@dlmu.edu.cn。
基金资助:CLC Number:
WANG Wenyang, LUO Yuping, YU Jiahuan, ZHOU Jibin, YE Mao, LIU Zhongmin. Analysis and forecasting of Chinese methanol price based on the intelligent chemical engineering large language model[J]. Chemical Industry and Engineering Progress, 2025, 44(10): 5673-5688.
王文洋, 罗玉平, 余佳洹, 周吉彬, 叶茂, 刘中民. 基于智能化工大模型的中国甲醇价格分析与预测[J]. 化工进展, 2025, 44(10): 5673-5688.
Add to citation manager EndNote|Ris|BibTeX
URL: https://hgjz.cip.com.cn/EN/10.16085/j.issn.1000-6613.2024-1284
| 研究对象 | 预测模型 | 模型类型 |
|---|---|---|
| 甲醇[ | LRSM | 计量经济学模型 |
| 甲醇[ | FIGARCH | 计量经济学模型 |
| 甲醇[ | GARCH+专家经验 | 混合模型 |
| 乙烯[ | VECM | 计量经济学模型 |
| 苯乙烯[ | ARIMA-ANN | 混合模型 |
| 聚乙烯[ | LSTM | ML模型 |
| 本文 | CEGPT-PF-M | 混合模型 |
| 研究对象 | 预测模型 | 模型类型 |
|---|---|---|
| 甲醇[ | LRSM | 计量经济学模型 |
| 甲醇[ | FIGARCH | 计量经济学模型 |
| 甲醇[ | GARCH+专家经验 | 混合模型 |
| 乙烯[ | VECM | 计量经济学模型 |
| 苯乙烯[ | ARIMA-ANN | 混合模型 |
| 聚乙烯[ | LSTM | ML模型 |
| 本文 | CEGPT-PF-M | 混合模型 |
| 应用领域 | 目标 | 模型名称 |
|---|---|---|
| 能源领域[ | 短期光伏功率预测 | VMD和Transformer混合模型 |
| 交通领域[ | 交通量预测 | LSTTN |
| 医学领域[ | 肾病病理分级预测 | CNN-Transformer |
| 生物领域[ | 毒性预测 | 基于Transformer架构的人工智能模型 |
| 环境领域[ | 水流预测 | MVMD-Transformer |
| 金融领域[ | 金融市场的波动性预测 | 基于Transformer架构的混合神经网络模型 |
| 工程领域[ | 大坝变形预测 | IPSO-LSTM-Transformer |
| 航空领域[ | 航空故障事件预测 | STL-transformer-ARIMA |
| 化工领域(本文) | 甲醇价格预测 | CEGPT-PF-M |
| 应用领域 | 目标 | 模型名称 |
|---|---|---|
| 能源领域[ | 短期光伏功率预测 | VMD和Transformer混合模型 |
| 交通领域[ | 交通量预测 | LSTTN |
| 医学领域[ | 肾病病理分级预测 | CNN-Transformer |
| 生物领域[ | 毒性预测 | 基于Transformer架构的人工智能模型 |
| 环境领域[ | 水流预测 | MVMD-Transformer |
| 金融领域[ | 金融市场的波动性预测 | 基于Transformer架构的混合神经网络模型 |
| 工程领域[ | 大坝变形预测 | IPSO-LSTM-Transformer |
| 航空领域[ | 航空故障事件预测 | STL-transformer-ARIMA |
| 化工领域(本文) | 甲醇价格预测 | CEGPT-PF-M |
算法1:CEGPT-PF-M时间序列预测算法 |
|---|
Input:外生变量 |
Step1:数据预处理,即外生变量进行Robust Standardization处理,目标变量取对数,并划分训练集和测试集。 |
Step2:定义Transformer模型结构,构建Transformer模型的编码层、解码层和线性层,选择MSE、MAE和RMSE等作为模型损失函数,Adam作为模型优化器。 |
Step3:输入嵌入层,将 |
Step4:位置编码,即求位置向量 |
对于每个位置 |
如果 |
如果 |
Step5:嵌入层向量与位置向量相加,将最终向量输入编码器。 |
Step6:编码器将输入序列转化为固定长度向量,映射为连续表示的上下文向量(能够概括整个输入序列语义信息),并输入至智能化工大模型的解码器①。 |
Step7:解码器将上下文向量转化为输出序列②(生成目标序列预测值,形式为非数值化信息)。 |
Step8:计算损失,选择最优损失函数进行损失计算。 |
Step9:模型训练。 ●初始化:动态参数 ●for t=1 to a.计算当前批次的损失函数 b.更新一阶矩估计 c.更新二阶矩估计 d.偏差校正 计算一阶矩估计的偏差校正 计算二阶矩估计的偏差校正 e.更新参数 end for |
Step10:保存模型,基于测试集进行测试。 |
Step11:智能化工大模型解码器的输出经过线性变换后,得到最终预测值。 |
算法1:CEGPT-PF-M时间序列预测算法 |
|---|
Input:外生变量 |
Step1:数据预处理,即外生变量进行Robust Standardization处理,目标变量取对数,并划分训练集和测试集。 |
Step2:定义Transformer模型结构,构建Transformer模型的编码层、解码层和线性层,选择MSE、MAE和RMSE等作为模型损失函数,Adam作为模型优化器。 |
Step3:输入嵌入层,将 |
Step4:位置编码,即求位置向量 |
对于每个位置 |
如果 |
如果 |
Step5:嵌入层向量与位置向量相加,将最终向量输入编码器。 |
Step6:编码器将输入序列转化为固定长度向量,映射为连续表示的上下文向量(能够概括整个输入序列语义信息),并输入至智能化工大模型的解码器①。 |
Step7:解码器将上下文向量转化为输出序列②(生成目标序列预测值,形式为非数值化信息)。 |
Step8:计算损失,选择最优损失函数进行损失计算。 |
Step9:模型训练。 ●初始化:动态参数 ●for t=1 to a.计算当前批次的损失函数 b.更新一阶矩估计 c.更新二阶矩估计 d.偏差校正 计算一阶矩估计的偏差校正 计算二阶矩估计的偏差校正 e.更新参数 end for |
Step10:保存模型,基于测试集进行测试。 |
Step11:智能化工大模型解码器的输出经过线性变换后,得到最终预测值。 |
| 变量 | LM统计量 | LM p值 | F检验统计量 | F检验p值 |
|---|---|---|---|---|
| 232 | 35.1702 | 0.0004399 | 3.2843 | 0.0002301 |
| 变量 | LM统计量 | LM p值 | F检验统计量 | F检验p值 |
|---|---|---|---|---|
| 232 | 35.1702 | 0.0004399 | 3.2843 | 0.0002301 |
| 数据总量 | 平均值 | 标准差 | 最小值 | 最大值 | 偏度 | 峰度 |
|---|---|---|---|---|---|---|
| 232 | 311.61 | 73.52 | 156.28 | 535.25 | 0.3029 | -0.1473 |
| 数据总量 | 平均值 | 标准差 | 最小值 | 最大值 | 偏度 | 峰度 |
|---|---|---|---|---|---|---|
| 232 | 311.61 | 73.52 | 156.28 | 535.25 | 0.3029 | -0.1473 |
| 变量 | MIC值 | 变量 | MIC值 | 变量 | MIC值 |
|---|---|---|---|---|---|
| 0.8241 | 0.4482 | 0.3733 | |||
| 0.7731 | 0.4330 | 0.3726 | |||
| 0.5783 | 0.4281 | 0.3688 | |||
| 0.5777 | 0.4227 | 0.3681 | |||
| 0.5614 | 0.4222 | 0.3592 | |||
| 0.5607 | 0.4175 | 0.3348 | |||
| 0.5509 | 0.4034 | 0.3342 | |||
| 0.5468 | 0.4005 | 0.3249 | |||
| 0.5329 | 0.3980 | 0.3182 | |||
| 0.5267 | 0.3970 | 0.3169 | |||
| 0.5191 | 0.3960 | 0.2981 | |||
| 0.4918 | 0.3889 | 0.2640 | |||
| 0.4587 | 0.3873 | 0.2513 | |||
| 0.4583 | 0.3849 | 0.2360 |
| 变量 | MIC值 | 变量 | MIC值 | 变量 | MIC值 |
|---|---|---|---|---|---|
| 0.8241 | 0.4482 | 0.3733 | |||
| 0.7731 | 0.4330 | 0.3726 | |||
| 0.5783 | 0.4281 | 0.3688 | |||
| 0.5777 | 0.4227 | 0.3681 | |||
| 0.5614 | 0.4222 | 0.3592 | |||
| 0.5607 | 0.4175 | 0.3348 | |||
| 0.5509 | 0.4034 | 0.3342 | |||
| 0.5468 | 0.4005 | 0.3249 | |||
| 0.5329 | 0.3980 | 0.3182 | |||
| 0.5267 | 0.3970 | 0.3169 | |||
| 0.5191 | 0.3960 | 0.2981 | |||
| 0.4918 | 0.3889 | 0.2640 | |||
| 0.4587 | 0.3873 | 0.2513 | |||
| 0.4583 | 0.3849 | 0.2360 |
| 宏观因素 | 微观因素 | 变量名 |
|---|---|---|
| 全球甲醇价格(C1) | 美国海湾地区甲醇价格 | |
| 台湾省甲醇价格 | ||
| 鹿特丹港甲醇价格 | ||
| 地缘政治风险(C2) | 中国地缘政治风险指数 | |
| 供需关系(C3) | 中国煤炭产量 | |
| 中国煤炭价格 | ||
| 中国煤炭销量 | ||
| 中国乙烯产量 | ||
| 中国丙烯产量 | ||
| 中国甲醛产量 | ||
| 中国二氯甲烷产量 | ||
| 中国乙二醇产量 | ||
| 中国苯乙烯产量 | ||
| 中国环氧乙烷产量 | ||
| 中国丙烯酸产量 | ||
| 能源市场(C4) | 全球石油产量 | |
| 布伦特原油价格 | ||
| 天然气价格指数 | ||
| 宏观经济(C5) | 中国制造业采购经理指数 | |
| 中国工业生产指数 | ||
| 中国汇率指数 | ||
| 中国经济政策不确定性指数 | ||
| 中国社会消费品零售额(月度) | ||
| 中国社会消费品零售额 | ||
| 中国工业生产者出厂价格指数(同比) | ||
| 中国采矿业工业生产者出厂价格指数(同比) | ||
| 其他大宗化工品市场(C6) | 中国大宗商品价格指数 | |
| 橡胶价格指数 | ||
| 中国环氧树脂价格 | ||
| 能源价格指数 | ||
| 聚丙烯价格 | ||
| 异丁醛价格 | ||
| 丙烯酸树脂价格 | ||
| 丁苯橡胶价格 | ||
| 顺丁橡胶价格 | ||
| 不饱和树脂价格 | ||
| 环氧氯丙烷价格 | ||
| 其他相关市场(C7) | 化工行业生产者价格指数 | |
| 煤炭行业生产者价格指数 | ||
| 能源行业生产者价格指数 | ||
| 纺织行业生产者价格指数 | ||
| 其他行业生产者价格指数 |
| 宏观因素 | 微观因素 | 变量名 |
|---|---|---|
| 全球甲醇价格(C1) | 美国海湾地区甲醇价格 | |
| 台湾省甲醇价格 | ||
| 鹿特丹港甲醇价格 | ||
| 地缘政治风险(C2) | 中国地缘政治风险指数 | |
| 供需关系(C3) | 中国煤炭产量 | |
| 中国煤炭价格 | ||
| 中国煤炭销量 | ||
| 中国乙烯产量 | ||
| 中国丙烯产量 | ||
| 中国甲醛产量 | ||
| 中国二氯甲烷产量 | ||
| 中国乙二醇产量 | ||
| 中国苯乙烯产量 | ||
| 中国环氧乙烷产量 | ||
| 中国丙烯酸产量 | ||
| 能源市场(C4) | 全球石油产量 | |
| 布伦特原油价格 | ||
| 天然气价格指数 | ||
| 宏观经济(C5) | 中国制造业采购经理指数 | |
| 中国工业生产指数 | ||
| 中国汇率指数 | ||
| 中国经济政策不确定性指数 | ||
| 中国社会消费品零售额(月度) | ||
| 中国社会消费品零售额 | ||
| 中国工业生产者出厂价格指数(同比) | ||
| 中国采矿业工业生产者出厂价格指数(同比) | ||
| 其他大宗化工品市场(C6) | 中国大宗商品价格指数 | |
| 橡胶价格指数 | ||
| 中国环氧树脂价格 | ||
| 能源价格指数 | ||
| 聚丙烯价格 | ||
| 异丁醛价格 | ||
| 丙烯酸树脂价格 | ||
| 丁苯橡胶价格 | ||
| 顺丁橡胶价格 | ||
| 不饱和树脂价格 | ||
| 环氧氯丙烷价格 | ||
| 其他相关市场(C7) | 化工行业生产者价格指数 | |
| 煤炭行业生产者价格指数 | ||
| 能源行业生产者价格指数 | ||
| 纺织行业生产者价格指数 | ||
| 其他行业生产者价格指数 |
| 划分形式 | MAE/% | MSE/% | RMSE/% | MAPE/% | SMAPE/% |
|---|---|---|---|---|---|
| 9∶1 | 0.59 | 12.16 | 6.87 | 1.53 | 6.86 |
| 8∶2 | 0.86 | 10.98 | 5.50 | 6.34 | 3.32 |
| 平均值 | 0.72 | 11.57 | 6.18 | 3.94 | 5.09 |
| 划分形式 | MAE/% | MSE/% | RMSE/% | MAPE/% | SMAPE/% |
|---|---|---|---|---|---|
| 9∶1 | 0.59 | 12.16 | 6.87 | 1.53 | 6.86 |
| 8∶2 | 0.86 | 10.98 | 5.50 | 6.34 | 3.32 |
| 平均值 | 0.72 | 11.57 | 6.18 | 3.94 | 5.09 |
| 划分形式 | 类型 | MAE | RMSE | MAPE/% | R2 | SDE |
|---|---|---|---|---|---|---|
| 9∶1 | 训练集 | 0.1108 | 0.1227 | 5.52 | 0.9132 | 0.0612 |
| 8∶2 | 0.0819 | 0.1147 | 3.28 | 0.9219 | 0.0758 | |
| 7∶3 | 0.0978 | 0.1230 | 4.21 | 0.9110 | 0.0915 | |
| 9∶1 | 微调集 | 0.3145 | 0.3564 | 10.61 | 0.7154 | 0.1201 |
| 8∶2 | 0.2982 | 0.3095 | 8.59 | 0.7529 | 0.1312 | |
| 7∶3 | 0.3047 | 0.3418 | 10.03 | 0.7391 | 0.1526 | |
| 9∶1 | 训练集+微调数据集 | 0.1040 | 0.1165 | 4.21 | 0.9478 | 0.0521 |
| 8∶2 | 0.0732 | 0.1023 | 3.08 | 0.9656 | 0.0364 | |
| 7∶3 | 0.0817 | 0.1107 | 3.26 | 0.9526 | 0.0880 |
| 划分形式 | 类型 | MAE | RMSE | MAPE/% | R2 | SDE |
|---|---|---|---|---|---|---|
| 9∶1 | 训练集 | 0.1108 | 0.1227 | 5.52 | 0.9132 | 0.0612 |
| 8∶2 | 0.0819 | 0.1147 | 3.28 | 0.9219 | 0.0758 | |
| 7∶3 | 0.0978 | 0.1230 | 4.21 | 0.9110 | 0.0915 | |
| 9∶1 | 微调集 | 0.3145 | 0.3564 | 10.61 | 0.7154 | 0.1201 |
| 8∶2 | 0.2982 | 0.3095 | 8.59 | 0.7529 | 0.1312 | |
| 7∶3 | 0.3047 | 0.3418 | 10.03 | 0.7391 | 0.1526 | |
| 9∶1 | 训练集+微调数据集 | 0.1040 | 0.1165 | 4.21 | 0.9478 | 0.0521 |
| 8∶2 | 0.0732 | 0.1023 | 3.08 | 0.9656 | 0.0364 | |
| 7∶3 | 0.0817 | 0.1107 | 3.26 | 0.9526 | 0.0880 |
| 模型 | 指标名称 | MAE | RMSE | MAPE/% | R2 | 误差均值标准差 |
|---|---|---|---|---|---|---|
| CEGPT-PF-M | 乙醇 | 0.1600 | 0.1385 | 5.12 | 0.9123 | 0.0936 |
| 乙烯 | 0.1257 | 0.1425 | 5.5 | 0.8814 | 0.1089 | |
| 丙烯 | 0.1565 | 0.1307 | 4.99 | 0.9201 | 0.0806 | |
| 聚乙烯 | 0.1790 | 0.1423 | 5.35 | 0.9005 | 0.0968 |
| 模型 | 指标名称 | MAE | RMSE | MAPE/% | R2 | 误差均值标准差 |
|---|---|---|---|---|---|---|
| CEGPT-PF-M | 乙醇 | 0.1600 | 0.1385 | 5.12 | 0.9123 | 0.0936 |
| 乙烯 | 0.1257 | 0.1425 | 5.5 | 0.8814 | 0.1089 | |
| 丙烯 | 0.1565 | 0.1307 | 4.99 | 0.9201 | 0.0806 | |
| 聚乙烯 | 0.1790 | 0.1423 | 5.35 | 0.9005 | 0.0968 |
| 模型 | 规模 | 相对误差/% | MAPE/% |
|---|---|---|---|
| CEGPT-PF-M | 8∶2 | 0.22 | 3.08 |
| 9∶1 | 0.61 | 4.21 | |
| 极端学习机算法[ | 长期 | 0.67~3.78 | — |
| 短期 | 1.6 | — | |
| 季节指数法[ | — | — | 4.62 |
| 指数平滑法[ | — | — | 5.19 |
| 模型 | 规模 | 相对误差/% | MAPE/% |
|---|---|---|---|
| CEGPT-PF-M | 8∶2 | 0.22 | 3.08 |
| 9∶1 | 0.61 | 4.21 | |
| 极端学习机算法[ | 长期 | 0.67~3.78 | — |
| 短期 | 1.6 | — | |
| 季节指数法[ | — | — | 4.62 |
| 指数平滑法[ | — | — | 5.19 |
| 宏观因素 | 重要度值 |
|---|---|
| 全球甲醇价格(C1) | 34.3 |
| 供需关系(C3) | 27.4 |
| 其他大宗化工品市场(C6) | 24.2 |
| 其他相关市场(C7) | 22.6 |
| 能源市场(C4) | 19.5 |
| 宏观经济(C5) | 15.4 |
| 地缘政治风险(C2) | 14.5 |
| 宏观因素 | 重要度值 |
|---|---|
| 全球甲醇价格(C1) | 34.3 |
| 供需关系(C3) | 27.4 |
| 其他大宗化工品市场(C6) | 24.2 |
| 其他相关市场(C7) | 22.6 |
| 能源市场(C4) | 19.5 |
| 宏观经济(C5) | 15.4 |
| 地缘政治风险(C2) | 14.5 |
| 变量 | 重要度值 | 变量 | 重要度值 | 变量 | 重要度值 |
|---|---|---|---|---|---|
| 55.5 | 25.0 | 23.0 | |||
| 37.0 | 25.0 | 22.5 | |||
| 36.0 | 25.0 | 19.5 | |||
| 34.0 | 24.5 | 16.0 | |||
| 31.5 | 24.5 | 14.5 | |||
| 30.0 | 24.5 | 14.5 | |||
| 28.0 | 24.5 | 13.5 | |||
| 28.0 | 24.5 | 13.5 | |||
| 28.0 | 24.0 | 13.5 | |||
| 27.0 | 24.0 | 13.0 | |||
| 26.5 | 24.0 | 13.0 | |||
| 26.0 | 24.0 | 9.5 | |||
| 25.5 | 23.5 | 8.0 | |||
| 25.5 | 23.5 | 7.0 |
| 变量 | 重要度值 | 变量 | 重要度值 | 变量 | 重要度值 |
|---|---|---|---|---|---|
| 55.5 | 25.0 | 23.0 | |||
| 37.0 | 25.0 | 22.5 | |||
| 36.0 | 25.0 | 19.5 | |||
| 34.0 | 24.5 | 16.0 | |||
| 31.5 | 24.5 | 14.5 | |||
| 30.0 | 24.5 | 14.5 | |||
| 28.0 | 24.5 | 13.5 | |||
| 28.0 | 24.5 | 13.5 | |||
| 28.0 | 24.0 | 13.5 | |||
| 27.0 | 24.0 | 13.0 | |||
| 26.5 | 24.0 | 13.0 | |||
| 26.0 | 24.0 | 9.5 | |||
| 25.5 | 23.5 | 8.0 | |||
| 25.5 | 23.5 | 7.0 |
| 类别 | 模型 | 模型评判标准 | |||
|---|---|---|---|---|---|
| MAE | RMSE | MAPE | R2 | ||
| SM | ARIMA | 0.1927 | 0.1850 | 3.57% | 0.7683 |
| SARIMA | 0.1523 | 0.1652 | 3.75% | 0.8195 | |
| ML | LSTM | 0.1479 | 0.1899 | 8.58% | 0.8277 |
| GRU | 0.1224 | 0.1518 | 7.49% | 0.8616 | |
| XGBoost | 0.1632 | 0.1977 | 8.41% | 0.8110 | |
| ANN | 0.1065 | 0.1159 | 3.49% | 0.8720 | |
| CNN | 0.1163 | 0.1245 | 6.16% | 0.8657 | |
| SVM | 0.1150 | 0.1181 | 5.71% | 0.8669 | |
| Transformer | LogSparse-Transformer | 0.1019 | 0.1093 | 3.87% | 0.8864 |
| Autoformer | 0.1090 | 0.1171 | 4.56% | 0.8701 | |
| Informer | 0.1047 | 0.1159 | 3.97% | 0.8805 | |
| LogTrans | 0.1052 | 0.1131 | 3.69% | 0.8774 | |
| 本文 | CEGPT-PF-M | 0.0859 | 0.1072 | 3.53% | 0.9541 |
| 类别 | 模型 | 模型评判标准 | |||
|---|---|---|---|---|---|
| MAE | RMSE | MAPE | R2 | ||
| SM | ARIMA | 0.1927 | 0.1850 | 3.57% | 0.7683 |
| SARIMA | 0.1523 | 0.1652 | 3.75% | 0.8195 | |
| ML | LSTM | 0.1479 | 0.1899 | 8.58% | 0.8277 |
| GRU | 0.1224 | 0.1518 | 7.49% | 0.8616 | |
| XGBoost | 0.1632 | 0.1977 | 8.41% | 0.8110 | |
| ANN | 0.1065 | 0.1159 | 3.49% | 0.8720 | |
| CNN | 0.1163 | 0.1245 | 6.16% | 0.8657 | |
| SVM | 0.1150 | 0.1181 | 5.71% | 0.8669 | |
| Transformer | LogSparse-Transformer | 0.1019 | 0.1093 | 3.87% | 0.8864 |
| Autoformer | 0.1090 | 0.1171 | 4.56% | 0.8701 | |
| Informer | 0.1047 | 0.1159 | 3.97% | 0.8805 | |
| LogTrans | 0.1052 | 0.1131 | 3.69% | 0.8774 | |
| 本文 | CEGPT-PF-M | 0.0859 | 0.1072 | 3.53% | 0.9541 |
| 变量 | 敏感度 | 变量 | 敏感度 | 变量 | 重要度值 |
|---|---|---|---|---|---|
| 2.5671 | 2.5018 | 2.4578 | |||
| 2.5987 | 2.3096 | 2.3663 | |||
| 2.3595 | 2.5335 | 2.4174 | |||
| 2.3458 | 2.4381 | 2.3979 | |||
| 2.5493 | 2.4843 | 2.5229 | |||
| 2.3397 | 2.5287 | 2.5014 | |||
| 2.4868 | 2.3395 | 2.4910 | |||
| 2.5910 | 2.3136 | 2.4386 | |||
| 2.5914 | 2.3228 | 2.4644 | |||
| 2.5892 | 2.3024 | 2.5453 | |||
| 2.5674 | 2.3592 | 2.5893 | |||
| 2.4561 | 2.3329 | 2.3222 | |||
| 2.4519 | 2.4787 | 2.3321 | |||
| 2.5232 | 2.4695 | 2.4644 |
| 变量 | 敏感度 | 变量 | 敏感度 | 变量 | 重要度值 |
|---|---|---|---|---|---|
| 2.5671 | 2.5018 | 2.4578 | |||
| 2.5987 | 2.3096 | 2.3663 | |||
| 2.3595 | 2.5335 | 2.4174 | |||
| 2.3458 | 2.4381 | 2.3979 | |||
| 2.5493 | 2.4843 | 2.5229 | |||
| 2.3397 | 2.5287 | 2.5014 | |||
| 2.4868 | 2.3395 | 2.4910 | |||
| 2.5910 | 2.3136 | 2.4386 | |||
| 2.5914 | 2.3228 | 2.4644 | |||
| 2.5892 | 2.3024 | 2.5453 | |||
| 2.5674 | 2.3592 | 2.5893 | |||
| 2.4561 | 2.3329 | 2.3222 | |||
| 2.4519 | 2.4787 | 2.3321 | |||
| 2.5232 | 2.4695 | 2.4644 |
| [1] | BROWN Tom B, MANN Benjamin, RYDER Nick, et al. Language models are few-shot learners[EB/OL]. (2020-05-28) [2024-08-15]. . |
| [2] | VASWANI Ashish, SHAZEER Noam, PARMAR Niki, et al. Attention is all you need[EB/OL]. (2017-06-12) [2024-08-15]. . |
| [3] | DEVLIN Jacob, CHANG Mingwei, LEE Kenton, et al. BERT: Pre-training of deep bidirectional transformers for language understanding[EB/OL]. (2018-10-11) [2024-08-15]. . |
| [4] | KUNZ Manuel, BIRR Stefan, RASLAN Mones, et al. Deep learning based forecasting: A case study from the online fashion industry[M]//Palgrave advances in the economics of innovation and technology. Cham: Springer Nature Switzerland, 2023: 279-311. |
| [5] | 张真, 张凡, 云祉婷. 绿氢在石化和化工行业的减碳经济性分析[J]. 化工进展, 2024, 43(6): 3021-3028. |
| ZHANG Zhen, ZHANG Fan, YUN Zhiting. Carbon reduction and techno-economic analysis of using green hydrogen in chemical and petrochemical industry[J]. Chemical Industry and Engineering Progress, 2024, 43(6): 3021-3028. | |
| [6] | WANG Xiaomeng, DEMIREL Yaşar. Feasibility of power and methanol production by an entrained-flow coal gasification system[J]. Energy & Fuels, 2018, 32(7): 7595-7610. |
| [7] | CHEN Zhuo, SHEN Qun, SUN Nannan, et al. Life cycle assessment of typical methanol production routes: The environmental impacts analysis and power optimization[J]. Journal of Cleaner Production, 2019, 220: 408-416. |
| [8] | Judit NYÁRI, MAGDELDIN Mohamed, LARMI Martti, et al. Techno-economic barriers of an industrial-scale methanol CCU-plant[J]. Journal of CO2 Utilization, 2020, 39: 101166. |
| [9] | SU Liwang, LI Xiangrong, SUN Zuoyu. The consumption, production and transportation of methanol in China: A review[J]. Energy Policy, 2013, 63: 130-138. |
| [10] | 中国产业信息网. 2017年中国甲醇行业发展现状及价格走势分析[EB/OL]. (2018-05-15) [2024-11-08]. . |
| China Industry Information Network. Analysis of the development status and price trends of China’s methanol industry in 2017[EB/OL]. (2018-05-15)[2024-11-08]. . | |
| [11] | AL-SHARRAH Ghanima K, ALATIQI Imad, ELKAMEL Ali. Modeling and identification of economic disturbances in the planning of the petrochemical industry[J]. Industrial & Engineering Chemistry Research, 2003, 42(20): 4678-4688. |
| [12] | 原野. 2009年甲醇市场分析及2010年预测[J]. 化工管理, 2010(2): 56-58. |
| YUAN Ye. Analysis of methanol market in 2009 and forecast in 2010[J]. Chemical Enterprise Management, 2010(2): 56-58. | |
| [13] | 殷红, 张霞, 王长波. 基于组合模型的大宗商品价格预测与可视分析——以甲醇价格为例[J]. 东华大学学报(自然科学版), 2017, 43(4): 541-546, 551. |
| YIN Hong, ZHANG Xia, WANG Changbo. Price forecasting and visual analysis of bulk commodities based on combined model: A case study of methanol prices[J]. Journal of Donghua University (Natural Science Edition), 2017, 43(4): 541-546, 551. | |
| [14] | NOWNEOW Anuwat, RUNGREUNGANUN Vilas. Poly vinyl chloride pellet price forecasting using ARIMA model[J]. International Journal of Mechanical Engineering and Technology, 2018, 9(13): 224-232. |
| [15] | 赵鲁涛, 郑志益, 邢悦悦, 等. 2021年国际原油价格分析与趋势预测[J]. 北京理工大学学报(社会科学版), 2021, 23(2): 25-29. |
| ZHAO Lutao, ZHENG Zhiyi, XING Yueyue, et al. International crude oil price analysis and projection in 2021[J]. Journal of Beijing Institute of Technology (Social Sciences Edition), 2021, 23(2): 25-29. | |
| [16] | 张金岱. 记忆性特征驱动的成品油价格预测研究[J]. 系统科学与数学, 2022, 42(5): 1300-1313. |
| ZHANG Jindai. Memory-trait-driven refined oil price forecasting[J]. Journal of Systems Science and Mathematical Sciences, 2022, 42(5): 1300-1313. | |
| [17] | 李慧, 李威龙, 胡一鸣, 等. 2024年成品油价格分析与趋势预测[J]. 北京理工大学学报(社会科学版), 2024, 26(2): 59-67. |
| LI Hui, LI Weilong, HU Yiming, et al. Analysis and trend prediction of refined oil price in 2024[J]. Journal of Beijing Institute of Technology (Social Sciences Edition), 2024, 26(2): 59-67. | |
| [18] | STOPFORD Martin. Maritime economics [M]. 3rd Ed. London: Routledge, 2008. |
| [19] | WANG Wenyang, HE Nan, CHEN Muxin, et al. Freight rate index forecasting with Prophet model based on multi-dimensional significant events[J]. Expert Systems with Applications, 2024, 249: 123451. |
| [20] | LU Minrong, XU Xuerong. TRNN: An efficient time-series recurrent neural network for stock price prediction[J]. Information Sciences, 2024, 657: 119951. |
| [21] | CHENG Jiyang, TIWARI Sunil, KHALED Djebbouri, et al. Forecasting Bitcoin prices using artificial intelligence: Combination of ML, SARIMA, and Facebook Prophet models[J]. Technological Forecasting and Social Change, 2024, 198: 122938. |
| [22] | REZAEI Hadi, FAALJOU Hamidreza, MANSOURFAR Gholamreza. Stock price prediction using deep learning and frequency decomposition[J]. Expert Systems with Applications, 2021, 169: 114332. |
| [23] | ZHANG Xin, XUE Tianyuan, EUGENE STANLEY H. Comparison of econometric models and artificial neural networks algorithms for the prediction of Baltic dry index[J]. IEEE Access, 2018, 7: 1647-1657. |
| [24] | SAEED Naima, NGUYEN Su, CULLINANE Kevin, et al. Forecasting container freight rates using the Prophet forecasting method[J]. Transport Policy, 2023, 133: 86-107. |
| [25] | SAEED Naima, NGUYEN Su, CULLINANE Kevin, et al. Forecasting container freight rates using the Prophet forecasting method[J]. Transport Policy, 2023, 133: 86-107. |
| [26] | LUNDBERG Scott, LEE Su-In. A unified approach to interpreting model predictions[EB/OL]. (2017-05-22) [2024-11-08]. . |
| [27] | 孙善辉, 李鸿, 张祖峰. 相空间重构和参数统一求解的石油价格预测[J]. 计算机工程与应用, 2013, 49(23): 247-251. |
| SUN Shanhui, LI Hong, ZHANG Zufeng. Oil price predicting based on unified solving by phase space reconstruction and parameters[J]. Computer Engineering and Applications, 2013, 49(23): 247-251. | |
| [28] | 张学龙, 王云峰, 谢廷宇. 灰色预测模型在石油化工原料价格预测中的应用[J]. 数学的实践与认识, 2014, 44(16): 31-38. |
| ZHANG Xuelong, WANG Yunfeng, XIE Tingyu. Application of grey forecasting model on price forecasting of petrochemical materials[J]. Mathematics in Practice and Theory, 2014, 44(16): 31-38. | |
| [29] | 董振宇, 冯恩民, 尹洪超, 等. 国际原油价格预测的双层随机整数规划模型、算法及应用[J]. 运筹学学报, 2015, 19(3): 18-25. |
| DONG Zhenyu, FENG Enmin, YIN Hongchao, et al. A bilevel stochastic integer programming model, algorithm, and application for international crude oil price forecasting[J]. Operations Research Transactions, 2015, 19(3): 18-25. | |
| [30] | Nimish JHA, KUMAR TANNERU Hemanth, PALLA Sridhar, et al. Multivariate analysis and forecasting of the crude oil prices: Part Ⅰ—Classical machine learning approaches[J]. Energy, 2024, 296: 131185. |
| [31] | LIU Longlong, ZHOU Suyu, Qian JIE, et al. A robust time-varying weight combined model for crude oil price forecasting[J]. Energy, 2024, 299: 131352. |
| [32] | LI Jingjing, HONG Zhanjiang, ZHANG Chengyuan, et al. A novel hybrid model for crude oil price forecasting based on MEEMD and Mix-KELM[J]. Expert Systems with Applications, 2024, 246: 123104. |
| [33] | 吴东武, 朱帮助. 基于HAR-RV-CJ模型的天然气价格预测[J]. 统计与决策, 2017, 33(23): 83-87. |
| WU Dongwu, ZHU Bangzhu. Natural gas price forecast based on HAR-RV-CJ model[J]. Statistics & Decision, 2017, 33(23): 83-87. | |
| [34] | 王建良, 雷昌然. 基于数据挖掘技术的天然气价格预测方法研究[J]. 中国矿业, 2020, 29(2): 52-58. |
| WANG Jianliang, LEI Changran. Research on the forecasting method for natural gas price based on the data mining technique[J]. China Mining Magazine, 2020, 29(2): 52-58. | |
| [35] | 裴莹, 李天祥, 王鏖清, 等. 基于新闻的国际天然气价格趋势预测方法[J]. 计算机科学, 2021, 48(S1): 235-239. |
| PEI Ying, LI Tianxiang, WANG Aoqing, et al. Prediction method of international natural gas price trends based on news[J]. Computer Science, 2021, 48(S1): 235-239. | |
| [36] | 周游, 张佳佳, 张良恺, 等. 2024年国际天然气市场分析与趋势预测[J]. 北京理工大学学报(社会科学版), 2024, 26(2): 76-83. |
| ZHOU You, ZHANG Jiajia, ZHANG Liangkai, et al. Analysis and trend forecasting prediction of the international natural gas markets in 2024[J]. Journal of Beijing Institute of Technology (Social Sciences Edition), 2024, 26(2): 76-83. | |
| [37] | TIWARI Aviral Kumar, SHARMA Gagan Deep, RAO Amar, et al. Unraveling the crystal ball: Machine learning models for crude oil and natural gas volatility forecasting[J]. Energy Economics, 2024, 134: 107608. |
| [38] | 李亚鹏, 韩旭, 于旭光, 等. 模型和数据混合驱动的双边协商电力市场合约价格预测方法[J]. 电力系统自动化, 2022, 46(18): 179-189. |
| LI Yapeng, HAN Xu, YU Xuguang, et al. Hybrid model-driven and data-driven approach to price forecasting in bilateral contract electricity markets[J]. Automation of Electric Power Systems, 2022, 46(18): 179-189. | |
| [39] | LOIZIDIS Stylianos, KYPRIANOU Andreas, GEORGHIOU George E. Electricity market price forecasting using ELM and bootstrap analysis: A case study of the German and Finnish day-ahead markets[J]. Applied Energy, 2024, 363: 123058. |
| [40] | MASIH A Mansur M, ALBINALI Khaled, DEMELLO Lurion. Price dynamics of natural gas and the regional methanol markets[J]. Energy Policy, 2010, 38(3): 1372-1378. |
| [41] | DELAVARI Majid, KHANI Nadiya Gandali ALI, NADERI Esmaeil. Oil and methanol price volatility[J]. Australian Journal of Business and Management Research, 2013, 3(8): 1-10. |
| [42] | ZHANG Xia, YIN Hong, WANG Changbo, et al. Forecast the price of chemical products with multivariate data[C]//2015 International Conference on Behavioral, Economic and Socio-cultural Computing (BESC). Nanjing, China: IEEE, 2015: 76-82. |
| [43] | Myriam THÖMMES, WINKER Peter. Multivariate modelling of cross-commodity price relations along the petrochemical value chain[M]//LAUSEN Berthold, VAN DEN POEL Dirk, ULTSCH Alfred, et al. Studies in classification, data analysis, and knowledge organization. Cham: Springer International Publishing, 2013: 427-435. |
| [44] | GHAHNAVIEH Ali Ebrahimi. Time series forecasting of styrene price using a hybrid ARIMA and neural network model[J]. Independent Journal of Management & Production, 2019, 10(3): 915-933. |
| [45] | LU Yachen, TENG Yufan, ZHANG Qi, et al. Prediction model for the chemical futures price using improved genetic algorithm based long short-term memory[J]. Processes, 2023, 11(1): 238. |
| [46] | 季晨洋, 林杰. 基于随机森林方法的甲醇期货价格预测与交易策略研究[J]. 上海管理科学, 2023, 45(1): 113-118. |
| JI Chenyang, LIN Jie. Research on methanol futures price forecast and trading strategy based on random forest method[J]. Shanghai Management Science, 2023, 45(1): 113-118. | |
| [47] | POVEY Daniel, HADIAN Hossein, GHAHREMANI Pegah, et al. A time-restricted self-attention layer for ASR[C]//2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Calgary, AB, Canada: IEEE, 2018: 5874-5878. |
| [48] | PARMAR Niki, VASWANI Ashish, USZKOREIT Jakob, et al. Image transformer[C]//Proceedings of the International Conference on Machine Learning. PMLR, 2018: 4055-4064. |
| [49] | KIM Jimin, OBREGON Josue, PARK Hoonseok, et al. Multi-step photovoltaic power forecasting using transformer and recurrent neural networks[J]. Renewable and Sustainable Energy Reviews, 2024, 200: 114479. |
| [50] | WANG Xinyu, MA Wenping. A hybrid deep learning model with an optimal strategy based on improved VMD and transformer for short-term photovoltaic power forecasting[J]. Energy, 2024, 295: 131071. |
| [51] | MO Site, WANG Haoxin, LI Bixiong, et al. Powerformer: A temporal-based transformer model for wind power forecasting[J]. Energy Reports, 2024, 11: 736-744. |
| [52] | 牛昊天, 林宇轩, 蔡念, 等. CNN-Transformer交互模型预测IgA肾病病理分级[J]. 计算机工程与应用, 2025, 61(10): 331-340. |
| NIU Haotian, LIN Yuxuan, CAI Nian, et al. CNN-Transformer interactive model for predicting IgA nephropathy pathological grading[J]. Computer Engineering and Applications, 2025, 61(10): 331-340. | |
| [53] | LUO Qinyao, HE Silu, HAN Xing, et al. LSTTN: A long-short term transformer-based spatiotemporal neural network for traffic flow forecasting[J]. Knowledge-Based Systems, 2024, 293: 111637. |
| [54] | NGUYEN Duc, FABLET Ronan. TrAISformer —A generative transformer for AIS trajectory prediction[EB/OL]. (2021-09-08) [2024-11-08]. . |
| [55] | 冼浩然, 江昊, 廖娟, 等. 基于Transformer的出租车轨迹预测方法研究[J]. 武汉大学学报(工学版), 2025, 58(2): 306-315. |
| XIAN Haoran, JIANG Hao, LIAO Juan, et al. Research on Taxi trajectory prediction method based on Transformer[J]. Engineering Journal of Wuhan University, 2025, 58(2): 306-315. | |
| [56] | GUSTAVSSON Mikael, Styrbjörn KÄLL, SVEDBERG Patrik, et al. Transformers enable accurate prediction of acute and chronic chemical toxicity in aquatic organisms[J]. Science Advances, 2024, 10(10): eadk6669. |
| [57] | 刘敏毅, 崔博文, 王宇坤, 等. 基于注意力机制的Transformer模型预测PM2.5浓度[J]. 环境科学, 2024, 45(12): 6993-7002. |
| LIU Minyi, CUI Bowen, WANG Yukun, et al. PM2.5 Concentration prediction based on transformer model with attention mechanism[J]. Environmental Science, 2024, 45(12): 6993-7002. | |
| [58] | SUN Wei, CHANG Li-Chiu, CHANG Fi-John. Deep dive into predictive excellence: Transformer’s impact on groundwater level prediction[J]. Journal of Hydrology, 2024, 636: 131250. |
| [59] | FANG Jinjie, YANG Linshan, WEN Xiaohu, et al. Ensemble learning using multivariate variational mode decomposition based on the Transformer for multi-step-ahead streamflow forecasting[J]. Journal of Hydrology, 2024, 636: 131275. |
| [60] | MISHRA Aswini Kumar, RENGANATHAN Jayashree, GUPTA Aaryaman. Volatility forecasting and assessing risk of financial markets using multi-transformer neural network based architecture[J]. Engineering Applications of Artificial Intelligence, 2024, 133: 108223. |
| [61] | 翁鸣昊, 项兴华, 陈俊涛, 等. 基于LSTM与Transformer的大坝变形预测研究[J]. 中国农村水利水电, 2024(4): 250-257. |
| WENG Minghao, XIANG Xinghua, CHEN Juntao, et al. Dam deformation prediction research based on LSTM and transformer[J]. China Rural Water and Hydropower, 2024(4): 250-257. | |
| [62] | ZENG Hang, ZHANG Hongmei, GUO Jiansheng, et al. A novel hybrid STL-transformer-ARIMA architecture for aviation failure events prediction[J]. Reliability Engineering & System Safety, 2024, 246: 110089. |
| [63] | 李明阳, 鲁之君, 曹东晶, 等. 一种基于Transformer编码器与LSTM的飞机轨迹预测方法[J]. 航天返回与遥感, 2024, 45(2): 163-176. |
| LI Mingyang, LU Zhijun, CAO Dongjing, et al. A predictive aircraft trajectory prediction method based on transformer encoder and LSTM[J]. Spacecraft Recovery & Remote Sensing, 2024, 45(2): 163-176. | |
| [64] | 翟文鹏, 宋一峤, 张兆宁. 基于Transformer-GRU网络的4D航迹预测[J]. 重庆交通大学学报(自然科学版), 2024, 43(6): 94-101. |
| ZHAI Wenpeng, SONG Yiqiao, ZHANG Zhaoning. 4D trajectory prediction based on transformer-GRU network[J]. Journal of Chongqing Jiaotong University (Natural Science), 2024, 43(6): 94-101. | |
| [65] | WU Haixu, XU Jiezhong, WANG Jianmin, et al. Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting[C]//Advances in Neural Information Processing Systems, 2021, 34: 22419-22430. |
| [66] | ZHOU Haoyi, ZHANG Shanghang, PENG Jieqi, et al. Informer: Beyond efficient transformer for long sequence time-series forecasting[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(12): 11106-11115. |
| [67] | LI Shiyang, JIN Xiaoyong, XUAN Yao, et al. Enhancing the locality and breaking the memory bottleneck of transformer on time series forecasting[EB/OL]. (2019-06-29) [2024-11-08]. . |
| [68] | NIE Xingqing, ZHOU Xiaogen, LI Zhiqiang, et al. LogTrans: Providing efficient local-global fusion with transformer and CNN parallel network for biomedical image segmentation[C]//2022 IEEE 24th Int Conf on High Performance Computing & Communications; 8th Int Conf on Data Science & Systems; 20th Int Conf on Smart City; 8th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys). Hainan, China: IEEE, 2022: 769-776. |
| [69] | WANG Huan, YUAN Zhaolin, CHEN Yibin, et al. An industrial missing values processing method based on generating model[J]. Computer Networks, 2019, 158: 61-68. |
| [70] | DEMIRHAN Haydar, RENWICK Zoe. Missing value imputation for short to mid-term horizontal solar irradiance data[J]. Applied Energy, 2018, 225: 998-1012. |
| [71] | ENGLE Robert F. Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation[J]. Econometrica, 1982, 50(4): 987. |
| [72] | DEKKING Frederik Michel, KRAAIKAMP Cornelis, LOPUHAÄ Hendrik Paul, et al. A modern introduction to probability and statistics: Understanding why and how[M]//Springer texts in statistics. London: Springer 2006. |
| [73] | 裴钦, 雷昭. 极端学习机算法下甲醇价格的短期与长期预测研究[J]. 黑龙江工业学院学报(综合版), 2022, 22(1): 127-132. |
| PEI Qin, LEI Zhao. On short-term and long-term prediction of methanol price based on extreme learning machine algorithm[J]. Journal of Heilongjiang University of Technology (Comprehensive Edition), 2022, 22(1): 127-132. |
| [1] | QIN Muxuan, ZHANG Wei, WANG Yingjin, LI Ziliang. Construction of Aspen model for large gas-liquid ratio in microreactors based on transfer learning [J]. Chemical Industry and Engineering Progress, 2025, 44(9): 4908-4916. |
| [2] | ZHENG Kexin, JIANG Yuxin, BI Kexin, ZHAO Qiming, CHEN Shaochen, WANG Bingbing, REN Junyu, JI Xu, QIU Tong, DAI Yiyang. Ensemble transfer learning framework for outflow compositions prediction in steam cracking process [J]. Chemical Industry and Engineering Progress, 2024, 43(5): 2880-2889. |
| Viewed | ||||||
|
Full text |
|
|||||
|
Abstract |
|
|||||
|
京ICP备12046843号-2;京公网安备 11010102001994号 Copyright © Chemical Industry and Engineering Progress, All Rights Reserved. E-mail: hgjz@cip.com.cn Powered by Beijing Magtech Co. Ltd |