GMDH方法在中长期电力负荷预测中的应用

在时间序列预测法的基础上,将数据处理组合方法GMDH应用于中长期电力负荷预测。介绍了GMDH的基本原理,根据历史数据建立GMDH模型,通过某地区电力负荷预测实例,对电力负荷值进行了计算分析,结果表明该方法可获得较高的预测精度。     

Bridging GPS outages of tightly-coupled GPS/SINS using GMDH neural network

A tightly-coupled GPS(global positioning system)/SINS(strap-down inertial navigation system)based on a GMDH(group method of data handling)neural network was presented to solve the problem of degraded accuracy for less than four visible GPS satellites with poor signal quality.Positions and velocities of the satellites were predicted by a GMDH neural network,and the pseudo-ranges and pseudo-range rates received by the GPS receiver were simulated to ensure the regular operation of the GPS/SINS Kalman filter during outages.In the mathematical simulation a tightly-coupled navigation system with a proposed approach has better navigation accuracy during GPS outages,and the anti-jamming ability is strengthened for the tightly-coupled navigation system.     

用自组织方法建立油田采收率模型的研究

数据组合处理方法（GMDH）是70年代发展起来的一种启发式自组织建立模型的方法，它能有将地解决复杂非线性系统的建模问题。文章简单介绍了此方法的基本原理。其思路是K－G多项式作为系统的完全描述，并分为若干层，每一层采用二次多项式作为基础函数，构成部分描述，用启发式规则筛选中间变量，逐层运算，得到最终所求模型。文章探讨了GMDH在建立油田采收率模型方面的应用，并以江汉油田的实际数据为例，建立了一个采收率预测模型，其计算结果令人满意，相对误差在5％以内。最后还讨论了该方法在油田开发中的应用前景。     

Design of enhanced MIA-GMDH learning networks

The article presents an enhanced multilayered iterative algorithm-group method of data handling (MIA-GMDH)-type network, discusses a comprehensive design methodology and carries out some numerical experiments which encompass system prediction and modelling. The method presented in this article is an enhancement of self-organising polynomial GMDH with several specific improved features – coefficient rounding and thresholding schemes and semi-randomised selection approach to pruning. The experiments carried out include representative time series prediction (gas furnace process data) and process modelling (investigating the milligrams of vitamin B2 per gram of turnip greens and drilling cutting force modelling). The results in this article show promising potential of self-organising network methodology in the field of both prediction and modelling applications.     

Data-driven modeling of H2 solubility in hydrocarbons using white-box approaches

As a result of technological advancements, reliable calculation of hydrogen (H₂) solubility in diverse hydrocarbons is now required for the design and efficient operation of processes in chemical and petroleum processing facilities. The accuracy of equations of state (EOSs) in estimating H₂ solubility is restricted, particularly in high-pressure or/and high-temperature conditions, which could result in energy loss and/or potential safety and environmental problem. Two strong machine learning techniques for building advanced correlation were used to evaluate H₂ solubility in hydrocarbons in this study which were Group method of data handling (GMDH) and genetic programming (GP). For that purpose, 1332 datasets from experimental results of H₂ solubility in 32 distinct hydrocarbons were collected from 68 various systems throughout a wide range of operating temperatures from 98 K to 701 K and pressures from 0.101325 MPa to 78.45 MPa. Hydrocarbons from two distinct classes include alkane, alkene, cycloalkane, aromatic, polycyclic aromatic, and terpene. Hydrocarbons have a molecular mass range of 28.054–647.2 g/mol, which corresponds to a carbon number of 2–46. Solvent molecular weight, critical pressure, and critical temperature, as well as pressure and temperature operational parameters, were used to create the features. With a regression coefficient (R²) which was equal to 0.986 and root mean square error (RMSE) which was 0.0132, the GP modeling approach estimated experimental solubility data more accurately than the GMDH approach. Operating pressure, followed by molecular weight of hydrocarbon solvents and temperature, had the greatest influence on estimation H₂ solubility, according to sensitivity analysis. The GP model shown in this paper is a reliable development that may be used in the chemical and petroleum sectors as a reliable and effective estimator for H₂ solubility in diverse hydrocarbons.     

An efficient two-stage approach for structural damage detection using meta-heuristic algorithms and group method of data handling surrogate model

In this study, the performance of an efficient two-stage methodology which is applied in a damage detection system using a surrogate model of the structure has been investigated. In the first stage, in order to locate the damage accurately, the performance of the modal strain energy based index for using different numbers of natural mode shapes has been evaluated using the confusion matrix. In the second stage, to estimate the damage extent, the sensitivity of most used modal properties due to damage, such as natural frequency and flexibility matrix is compared with the mean normalized modal strain energy (MNMSE) of suspected damaged elements. Moreover, a modal property change vector is evaluated using the group method of data handling (GMDH) network as a surrogate model during damage extent estimation by optimization algorithm; in this part of methodology, the performance of the three popular optimization algorithms including particle swarm optimization (PSO), bat algorithm (BA), and colliding bodies optimization (CBO) is examined and in this regard, root mean square deviation (RMSD) based on the modal property change vector has been proposed as an objective function. Furthermore, the effect of noise in the measurement of structural responses by the sensors has also been studied. Finally, in order to achieve the most generalized neural network as a surrogate model, GMDH performance is compared with a properly trained cascade feed-forward neural network (CFNN) with log-sigmoid hidden layer transfer function. The results indicate that the accuracy of damage extent estimation is acceptable in the case of integration of PSO and MNMSE. Moreover, the GMDH model is also more efficient and mimics the behavior of the structure slightly better than CFNN model.     

Implementation of an optimized binary classification by GMDH-type neural network algorithm for predicting the blast produced ground vibration

Ground vibration is one of the most important undesired phenomena resulting from blasting operations imposing damages to facilities and buildings on the one hand, and creating environmental problems in open pit mining on the other. Therefore, the present study aims to provide an optimized classification binary model to identify the blasting patterns with an acceptable ground vibration intensity to reduce the damages resulting from this artificial phenomenon. This study uses a binary method to provide an optimized classification model for predicting and evaluating the blasting patterns with the minimum ground vibration. Group Method of Data Handling-Type Neural Network is used as one of the most practical optimization algorithms to solve complicated and uncertain problems in this modelling. In this study, by collecting the data of 52 different blasting patterns from Soungun copper mine, some of the most important geometric properties and the amount of ammonium nitrate fuel oil consumed in each blasting pattern are recorded. In addition, based on expertise and experience of experts, the degree of ground vibration produced by each blasting is qualitatively classified into four different ranges of very high, high, normal and low in the form of unacceptable (very high and High) and acceptable (normal and low) clusters. Based on the results obtained from the analyses, the developed model has a high flexibility and ability in the binary prediction of blasting patterns with an acceptable vibration magnitude.     

Path synthesis of four-bar mechanisms using synergy of polynomial neural network and Stackelberg game theory

In this article, a novel approach based on game theory is presented for multi-objective optimal synthesis of four-bar mechanisms. The multi-objective optimization problem is modelled as a Stackelberg game. The more important objective function, tracking error, is considered as the leader, and the other objective function, deviation of the transmission angle from 90° (TA), is considered as the follower. In a new approach, a group method of data handling (GMDH)-type neural network is also utilized to construct an approximate model for the rational reaction set (RRS) of the follower. Using the proposed game-theoretic approach, the multi-objective optimal synthesis of a four-bar mechanism is then cast into a single-objective optimal synthesis using the leader variables and the obtained RRS of the follower. The superiority of using the synergy game-theoretic method of Stackelberg with a GMDH-type neural network is demonstrated for two case studies on the synthesis of four-bar mechanisms.     

GMDH-based networks for intelligent intrusion detection

Network intrusion detection has been an area of rapid advancement in recent times. Similar advances in the field of intelligent computing have led to the introduction of several classification techniques for accurately identifying and differentiating network traffic into normal and anomalous. Group Method for Data Handling (GMDH) is one such supervised inductive learning approach for the synthesis of neural network models. Through this paper, we propose a GMDH-based technique for classifying network traffic into normal and anomalous. Two variants of the technique, namely, Monolithic and Ensemble-based, were tested on the KDD-99 dataset. The dataset was preprocessed and all features were ranked based on three feature ranking techniques, namely, Information Gain, Gain Ratio, and GMDH by itself. The results obtained proved that the proposed intrusion detection scheme yields high attack detection rates, nearly 98%, when compared with other intelligent classification techniques for network intrusion detection.     

Cluster ensemble framework based on the group method of data handling

Cluster ensemble is a powerful method for improving both the robustness and the stability of unsupervised classification solutions. This paper introduced group method of data handling (GMDH) to cluster ensemble, and proposed a new cluster ensemble framework, which named cluster ensemble framework based on the group method of data handling (CE-GMDH). CE-GMDH consists of three components: an initial solution, a transfer function and an external criterion. Several CE-GMDH models can be built according to different types of transfer functions and external criteria. In this study, three novel models were proposed based on different transfer functions: least squares approach, cluster-based similarity partitioning algorithm and semidefinite programming. The performance of CE-GMDH was compared among different transfer functions, and with some state-of-the-art cluster ensemble algorithms and cluster ensemble frameworks on synthetic and real datasets. Experimental results demonstrate that CE-GMDH can improve the performance of cluster ensemble algorithms which used as the transfer functions through its unique modelling process. It also indicates that CE-GMDH achieves a better or comparable result than the other cluster ensemble algorithms and cluster ensemble frameworks.