Accuracy improvement of various short-term load forecasting models by a novel and unified statistical data-filtering method

ABSTRACT

Time-series and machine-learning methods are being strongly exploited to improve the accuracy of short-term load forecasting (STLF) results. In developing countries, power consumption behaviors could be suddenly changed by different customers, e.g. industrial customers, residential customers, so the load-demand dataset is often unstable. Therefore, reliability assessment of the load-demand dataset is obviously necessary for STLF models. Hence, this paper proposes a novel and unified statistical data-filtering method with the best confidence interval to eliminate unexpected noises/outliers of the input dataset before performing various short-term load forecasting models. This proposed novel data-filtering method, so-called the data pre-processing method, is also compared to other existing data-filtering methods (e.g. Kalman filter, Density-Based Spatial Clustering of Applications with Noise, Wavelet transform, and Singular Spectrum Analysis). By using an SCADA system?-based database of a typical 22kV distribution network in Vietnam, NYISO database, and PJM-RTO database, case studies of short-term load forecasting have been conducted with a conventional ARIMA model, an ANN forecasting model, an LSTM-RNN model, an LSTM-CNN combined model, a deep auto-encoder (DAE) network, a Wavenet-based model, a Wavenet and LSTM hybrid model, and a Wavelet Neural Network (WNN) model, which are to validate the novel and unified statistical data-filtering method proposed. The achieved numerical results demonstrate which the accuracy of the aforementioned STLF models can be significantly improved due to the proposed statistical data-filtering method with the best confidence interval of the input load dataset. The proposed statistical data-filtering method can considerably outperform the existing data-filtering methods.     

A Markov random field spatio-temporal analysis of ocean temperature

The National Oceanic Data Center (NODC) contains historical records from approximately 144,000 hydrographic stations in the North Atlantic. This data has been used by oceanographers to construct maps of point estimates of pressure, temperature, salinity and oxygen in the North Atlantic (Levitus (1994); Lozier et al. (1995)). Because data from any particular year are scarce, the previous maps have been for time-averaged values only. In addition, the maps have been reported without uncertainty estimates. This paper presents a Markov random field (MRF) analysis that can generate maps for specific time periods along with associated uncertainties. To estimate changes in oceanic properties over time previous oceanographic work has focused on differences between a few time periods each having many observations. Due to data scarcity this poses a severe restriction for both spatial and temporal coverage of climatic change. The MRF analysis provides a means for temporal modeling that does not require high data density at each time period. To demonstrate the usefulness of a MRF analysis of oceanic data we investigate the temporal variability along 24.5°N in the North Atlantic. Our results are compared to an earlier analysis (Parrilla et al. (1994)) where data from only three time periods was used. We obtain a more thorough understanding of the temperature change found by this previous study.     

基于CREAM和贝叶斯网络的航空维修人为差错概率预测

航空维修差错不仅严重威胁着飞行安全,同时也会增加航空公司的维修成本。针对航空维修人员发生差错成因的复杂性以及历史事故数据缺乏的情况下,将人因可靠性与失误分析方法（CREAM）和贝叶斯网络（BN）相结合,提出一种改进的维修差错分析模型。根据维修任务构建相应的贝叶斯网络模型,为各子节点设置条件概率表（CPT）;基于维修基地的实际维修环境,对行为形成因子（PSFs）进行评估,得到共同绩效条件（CPCs）的水平;利用各CPC因子下各个行为功能失效模式的权重因子,对各认知活动进行失效概率的修正;将修正概率作为贝叶斯网络根节点的输入,利用推理机制,得到差错发生概率。通过案例分析和计算,验证了所述方法的可行性和有效性。     

基于故障树与贝叶斯网络的钻井井塌事故的定量分析

井塌事故在钻井过程极其常见,由于钻井液、井身结构等设计不合理,以及钻井过程的误操作均会造成井塌的发生,其后果主要是造成卡钻、卡套管等事故,严重时造成井眼毁坏。为了对井塌事故进行危险性分析,采用贝叶斯网络与故障树相结合的方法,避免传统故障树方法的局限性。首先对井塌事故的所有危险因素进行辨识建立故障树,然后将故障树转化为贝叶斯网络,建立条件概率表,运用贝叶斯网络的推理能力对井塌事故进行危险性分析。对井塌事故各基本事件概率分布的计算分析结果表明,考虑事件的多态性,综合利用故障树分析法和贝叶斯网络法能有效提高井塌事故分析的有效性,推算出井塌事故概率分布更为准确。     

Analysis of factors that influence hazardous material transportation accidents based on Bayesian networks: A case study in China

In this study, we applied Bayesian networks to prioritize the factors that influence hazardous material (Hazmat) transportation accidents. The Bayesian network structure was built based on expert knowledge using Dempster-Shafer evidence theory, and the structure was modified based on a test for conditional independence. We collected and analyzed 94 cases of Chinese Hazmat transportation accidents to compute the posterior probability of each factor using the expectation-maximization learning algorithm. We found that the three most influential factors in Hazmat transportation accidents were human factors, the transport vehicle and facilities, and packing and loading of the Hazmat. These findings provide an empirically supported theoretical basis for Hazmat transportation corporations to take corrective and preventative measures to reduce the risk of accidents.     

水电厂电力设备事故应急响应FPN-MC模型及效能分析

为更好地利用信息技术研究水电厂电力设备事故应急效能,提高应急救援效率和应急处置能力,综合考虑应急响应过程中各个环节时间的随机性与模糊性,运用模糊Petri网（FPN）和马尔科夫链（MC）理论,构建水电厂电力设备事故应急响应FPN-MC模型,求解模糊稳态概率值并分析系统效能指标。最后,以ZLB水电厂电力设备事故Ⅲ级应急响应作为实例,进一步剖析应急流程信息拥挤的瓶颈环节及事件应急处置的关键节点。结果表明:应急资源使用完毕、相关人员现场处置到位、确定完善的处置方案时的状态最为繁忙;利用资源实施救援、相关人员进行现场处置、完善处置方案等实施过程需要重点监督和管理。     

基于贝叶斯网络的气井井筒完整性风险评价

井筒完整性失效是气井生产中的主要风险,为有效评价井筒完整性风险,应用贝叶斯网络的推理与学习能力,建立了基于贝叶斯网络和Noisy-OR gate模型的井筒完整性失效概率计算方法和风险评价模型。由故障树分析将井筒分为管柱、水泥环密封性、井口装置、水力屏障和其他部件5个评价单元,确定了各单元的主要风险因素,建立了井筒完整性失效的贝叶斯网络拓扑结构;由Noisy-OR gate模型和历史数据,确定了贝叶斯网络的条件概率参数;将基于贝叶斯网络的失效概率与层次分析法相结合,确定了风险评价指标和等级划分标准;建立了气井井筒完整性风险评价方法。结果表明,该方法实现了井筒完整性失效概率的定量计算、风险的定量评价和主要风险因素的反向推理,可为预防和控制井筒完整性失效提供决策依据,有助于降低井筒完整性失效风险。     

基于随机Petri网的人道主义救援供应链业务流程协同建模及效能分析

伴随自然灾害频发而来的人道主义救援规模越来越大,针对救援业务流程的复杂性特征,从人道主义救援供应链业务流程分析入手,运用随机Petri网,对灾后响应阶段的救援供应链业务流程协同进行过程建模,并通过同构马尔科夫链,构建其可达图;应用Matlab软件,仿真人道主义救援供应链业务流程协同时间效能及运营效率。结果表明:模型能够检验救援业务流程中存在的瓶颈环节和耗时的主要活动,为提高人道主义救援供应链的运营效率提供理论支持及实践指导。     

近50年贵州净生态系统生产力时空分布特征

利用大气植被相互作用模型AVIM2(Atmosphere-Vegetation Interaction Model 2)估算分析了时间长度为50年、空间分辨率为0.02°×0.02°的贵州净生态系统生产力(NEP),分析了其对气候变化的响应。结果表明,(1) AVIM2模型能够模拟出贵州森林净初级生产力(NPP)的变化,模拟偏差随着树龄的增大而不断减小,其模拟效果优于综合模型。(2)1961-2010年,贵州NEP(以C计)平均值为23.9 g·m-2·a-1,碳源区面积比例仅为5%,且植被覆盖类型为南部部分常绿阔叶林。NEP总量的变动范围为-7.0~11.5 Tg ·a-1,平均每年吸收碳4.87Tg,碳汇量占中国区域的3~7%。(3)贵州境内31%的区域固碳能力下降明显(P<0.05)且主要集中在植被类型为常绿针叶林及农作物的北部地区,还有7%的区域固碳能力升高明显(P<0.05)且位于南部部分常绿阔叶林地区。(4)贵州NEP与气温显著负相关(P<0.01),与降水量显著正相关(P<0.05),气温对NEP的影响大于降水。     

基于区间活性数据的定量结构-活性关系

利用贝叶斯统计方法构建了基于区间活性数据的取代苯胺和苯酚类化合物对大型溞(Daphniamagna)24 h急性毒性的定量结构-活性关系模型,并与基于平均数和中位数的点估计活性数据的定量结构-活性关系模型进行了比较.结果表明,前者可以充分利用化合物的活性数据信息,模型具有更好的拟合效果与预测能力以及较宽的应用范围.基于区间活性数据的定量结构-活性关系模型可为生态风险评价等提供更加可靠的预测数据.