Deterministic and probabilistic ship pitch prediction using a multi-predictor integration model based on hybrid data preprocessing,reinforcement learning and improved QRNN

The deterministic and probabilistic prediction of ship motion is important for safe navigation and stable real-time operational control of ships at sea. However, the volatility and randomness of ship motion, the non-adaptive nature of single predictors and the poor coverage of quantile regression pose serious challenges to uncertainty prediction, making research in this field limited. In this paper, a multi-predictor integration model based on hybrid data preprocessing, reinforcement learning and improved quantile regression neural network (QRNN) is proposed to explore the deterministic and probabilistic prediction of ship pitch motion. To validate the performance of the proposed multi-predictor integrated prediction model, an experimental study is conducted with three sets of actual ship longitudinal motions during sea trials in the South China Sea. The experimental results indicate that the root mean square errors (RMSEs) of the proposed model of deterministic prediction are 0.0254°, 0.0359°, and 0.0188°, respectively. Taking series #2 as an example, the prediction interval coverage probabilities (PICPs) of the proposed model of probability predictions at 90%, 95%, and 99% confidence levels (CLs) are 0.9400, 0.9800, and 1.0000, respectively. This study signifies that the proposed model can provide trusted deterministic predictions and can effectively quantify the uncertainty of ship pitch motion, which has the potential to provide practical support for ship early warning systems.     

基于多通道的ICU脑血管疾病死亡风险预测模型

死亡风险预测指根据病人临床体征监测数据来预测未来一段时间的死亡风险。对于ICU病患，通过死亡风险预测可以有针对性地对病人做出临床诊断，以及合理安排有限的医疗资源。基于临床使用的MEWS和Glasgow昏迷评分量表，针对ICU病人临床监测的17项生理参数，提出一种基于多通道的ICU脑血管疾病死亡风险预测模型。引入多通道概念应用于BiLSTM模型，用于突出每个生理参数对死亡风险预测的作用。采用Attention机制用于提高模型预测精度。实验数据来自MIMIC [Ⅲ]数据库，从中提取3?080位脑血管疾病患者的16?260条记录用于此次研究，除了六组超参数实验之外，将所提模型与LSTM、Multichannel-BiLSTM、逻辑回归（logistic regression）和支持向量机（support vector machine, SVM）四种模型进行了对比分析，准确率Accuracy、灵敏度Sensitive、特异性Specificity、AUC-ROC和AUC-PRC作为评价指标，实验结果表明，所提模型性能优于其他模型，AUC值达到94.3%。     

Dynamic shovel allocation approach to short-term production planning in open-pit mines

The complexity and uncertainties associated with mining operations often lead to deviations from short-term plans. A proactive approach will predict such deviations and help minimise opportunity losses, providing economic and operational gains. This paper presents an optimisation tool as a dynamic shovel allocation decision-maker for mining operation simulation models. The objectives of the optimisation tool are to provide shovel allocations to mining faces in order to maximise production, meet desired head grade and tonnage at the crushers, and minimise shovel movements. This paper presents the development and implementation of the optimisation tool with an iron ore mine case study.     

How Far Are We from Prescribing Fasting as Anticancer Medicine?

(1) Background: the present review provides a comprehensive and up-to date overview of the potential exploitation of fasting as an anticancer strategy. The rationale for this concept is that fasting elicits a differential stress response in the setting of unfavorable conditions, empowering the survival of normal cells, while killing cancer cells. (2) Methods: the present narrative review presents the basic aspects of the hormonal, molecular, and cellular response to fasting, focusing on the interrelationship of fasting with oxidative stress. It also presents nonclinical and clinical evidence concerning the implementation of fasting as adjuvant to chemotherapy, highlighting current challenges and future perspectives. (3) Results: there is ample nonclinical evidence indicating that fasting can mitigate the toxicity of chemotherapy and/or increase the efficacy of chemotherapy. The relevant clinical research is encouraging, albeit still in its infancy. The path forward for implementing fasting in oncology is a personalized approach, entailing counteraction of current challenges, including: (i) patient selection; (ii) fasting patterns; (iii) timeline of fasting and refeeding; (iv) validation of biomarkers for assessment of fasting; and (v) establishment of protocols for patients’ monitoring. (4) Conclusion: prescribing fasting as anticancer medicine may not be far away if large randomized clinical trials consolidate its safety and efficacy.     

Operational time-series data modeling via LSTM network integrating principal component analysis based on human experience

Today’s information technologies involve increasingly intelligent systems, which come at the cost of increasingly complex equipment. Modern monitoring systems collect multi-measuring-point and long-term data which make equipment health prediction a “big data” problem. It is difficult to extract information from such condition monitoring data to accurately estimate or predict health statuses. Deep learning is a powerful tool for big data processing that is widely utilized in image and speech recognition applications, and can also provide effective predictions in industrial processes. This paper proposes the Long Short-term Memory Integrating Principal Component Analysis based on Human Experience (HEPCA-LSTM), which uses operational time-series data for equipment health prognostics. Principal component analysis based on human experience is first conducted to extract condition parameters from the condition monitoring system. The long short-term memory (LSTM) framework is then constructed to predict the target status. Finally, a dynamic update of the prediction model with incoming data is performed at a certain interval to prevent any model misalignment caused by the drifting of relevant variables. The proposed model is validated on a practical case and found to outperform other prediction methods. It utilizes a powerful deep learning analysis method, the LSTM, to fully process big condition monitoring series data; it effectively extracts the features involved with human experience and takes dynamic updates into consideration.     

改进Elman神经网络在短期热负荷预测中的应用

热电厂的短期热负荷预测在城市集中供暖中起着至关重要的作用,直接影响热电厂的经济效益和热能利用率。电厂的短期热负荷一般采用神经网络预测模型进行预测,而BP神经网络应用最为广泛。Elman神经网络算法在BP神经网络基础上加入了承接层,作为一步延时算子,实现记忆能力,使系统具备适应时变能力,增强系统全局稳定性。但Elman神经网络算法模型的构造依然需要大量样本的支撑,而且输入层的变量多,导致预测时间依然很长,收敛速度慢。该文在Elman神经网络预测前,进行了相关系数预处理和对样本中异常值的平均化预处理,通过数据归一化运算,使Elman神经网络输入层变量大幅减少。仿真实验表明,改进的Elman神经网络算法使预测模型快速寻优,减少预测时间的同时明显提高预测精度。     

基于IFOA-GRNN的电力系统短期负荷预测

考虑到电网负荷与诸多因素有关，设计了一种带有温度、气象、日期类型的广义回归神经网络（GRNN）负荷预测模型。为了提高该模型的预测精度，提出了一种改进果蝇优化算法优化广义回归神经网络（IFOA-GRNN）的方法，即在利用果蝇优化算法（FOA）进入迭代寻优时，通过改进搜索距离优化该算法的性能和稳定性。利用改进的FOA优化GRNN的光滑参数，然后利用训练好的预测模型对甘肃省某地区进行了短期负荷预测，并与FOA-GRNN和误差反向传播神经网络(BPNN)模型结果进行了误差比较。结果表明， IFOA-GRNN具有较高的预测精度，能够满足电力系统短期负荷预测的要求。     

考虑CEEMDAN样本熵和SVR的短期风速预测

为降低由于风速信号的非线性和非平稳性带来的风速预测难度，提高短期风速预测的准确性，提出一种考虑样本熵的组合分解模式和支持向量回归（SVR）相结合的预测模型。首先采用自适应噪声的完全集合经验模态分解（CEEMDAN）方法分解风速历史数据，并计算各模态分量的样本熵；然后采用变分模态分解（VMD）方法对样本熵最大的模态分量进行二次分解，充分削弱风速分量的非平稳性；接着对分解得到所有模态分量分别建立SVR预测模型；最后将各分量的预测值求和完成最终风速预测。实例分析表明，所提模型对比其他模型的预测误差最小，预测精度最高，可有效预测短期风速。     

基于SARIMA-GRNN-SVM的短期商业电力负荷组合预测方法

针对短期商业电力负荷预测准确性与周期难以满足现有电力现货市场的问题,提出了一种基于SARIMAGRNN-SVM(seasonal autoregressive integrated moving average-generalized regression neural network-support vector machine)的商业电力负荷组合预测模型。首先,对商业电力负荷变化的周期规律与随机因素的复杂影响进行了分析;然后,结合以上分析,选用SARIMA和GRNN为单一预测模型对商业电力负荷进行预测,并利用SVM进行组合,实现日前商业电力负荷预测;最后,通过某商业综合体的电力负荷数据进行验证。所提组合预测模型较单一预测模型拥有更优的预测精度与鲁棒性,可以为短期商业电力负荷预测提供借鉴。     

Genetic algorithm to simultaneously optimise stope sequencing and equipment dispatching in underground short-term mine planning under time uncertainty

ABSTRACT

The main issue in short-term planning optimisation for underground mining is organising the mining process with limited resources in the form of equipment and materials to satisfy production targets and stable feed grade requirements. In this paper, an integrated optimisation model is proposed based on an individual generation algorithm and an improved Genetic Algorithm to simultaneously optimise stope extraction sequencing and timing, extracted ore grade and equipment dispatching. The model objectives are to shorten the time gap between the stope mining processes and the overall working time. When the uncertainty of equipment working time is taken into account in a short-term scheduling model, the Monte Carlo simulation is applied to evaluate the risk of not meeting the production target. A modification strategy is defined to evaluate equipment failure. Consequently, any available equipment is automatically reassigned to the mining site to replace the broken-down equipment. A case study is used to validate the model in the Sanshandao gold mine of China to formulate an optimal monthly schedule. Compared with the conventional approach, the new model could reduce the variance of ore tonnage and feed grade and improve the equipment allocation efficiency. Discussions are presented to address the uncertainty.