Fuzzy Prediction Research Articles

Fuzzy Supervised Multi-Period Time Series Forecasting

Abstract The goal of this paper is to propose a new method for fuzzy forecasting of time series with supervised learning and k-order fuzzy relationships. In the training phase based on k previous historical periods, a multidimensional matrix of fuzzy dependencies is constructed. During the test stage, the fitted fuzzy model is run for validating the observations and each output value is predicted by using a fuzzy input vector of k previous intervals. The proposed algorithm is verified by a benchmark dataset for fuzzy time series forecasting. The results obtained are similar or better than those of other fuzzy time series prediction methods. Comparative analysis shows the high potential of the new algorithm as an alternative to fuzzy prediction and reveals some opportunities for its further improvement.

Experimental investigation and adaptive neural fuzzy inference system prediction of copper recovery from flotation tailings by acid leaching in a batch agitated tank

The potential of copper recovery from flotation tailings was experimentally investigated using a laboratory-mixing tank. The experiments were performed with solid weight percentages of 30wt%, 35wt%, 40wt% and 45wt% in water. The measurements revealed that adding sulfuric acid all at once to the tank rapidly increased the efficiency of the leaching process, which was attributed to the rapid change in the acid concentration. The rate of iron dissolution from tailings was less than when the acid was added gradually. The sample with 40wt% solid is recommended as an appropriate feed for the recovery of copper. The adaptive neural fuzzy system (ANFIS) was also used to predict the copper recovery from flotation tailings. The back-propagation algorithm and least squares method were applied for the training of ANFIS. The validation data was also applied to evaluate the performance of these models. Simulation results revealed that the testing results from these models were in good agreement with the experimental data.

Fuzzy logic-integrated PSO methodology for parameters optimization in end milling of Al/SiCp MMC

This paper deals with a novel integrated parametric optimization in end milling of Al–SiCp metal–matrix composites using carbide tool. Fuzzy logic modeling combined with particle swarm optimization (PSO) technique is applied as intelligent optimization methodology to obtain optimal process parameters in end milling process. Fuzzy rules-based surface roughness prediction model is developed with SiCp percentage as one the input parameter is addition to spindle speed, feed and depth of cut. Comparison of fuzzy rule model predictions with confirmation data sets showed 95.44% average prediction accuracy. The effect of process parameters on surface roughness is also studied. Analysis of variance established feed rate as the most significant process variable with 69.93% contribution followed by SiCp (19%) and spindle speed (9.73%), respectively. The integrated PSO optimizer optimizes the process parameters for an objective to minimize the machining time satisfying desired surface roughness to be obtained. A number of problems with different values of desired surface roughness are solved and algorithm converged in less than ten iterations.

Fuzzy logic method to investigate grinding of alumina ceramic using minimum quantity lubrication

AbstractThe use of nanofluid in lubrication during machining of advanced engineering ceramics has been found to be highly efficient and eco‐friendly. This work involves experimental investigation of grinding Alumina (Al2O3) ceramic to determine the effect of the grinding variables. The grinding variables considered include depth of cut, feed rate, type of diamond wheel, and lubrication type. Moreover, the response parameters considered include grinding power, coefficient of friction, and surface quality. The responses obtained during the experiments were used to develop a fuzzy logic prediction model. The findings from this work can be concluded as follows: (a) The depth of cut and feed rate have direct proportional relationship with the grinding power and coefficient of friction. (b) The metallic bonded diamond wheel was found to have higher machining efficiency than the resinoid bonded one. (c) Higher number of diamond grits produces lower frictional coefficient. (d) The carbon nanotube based nanofluid when used in the minimum quantity lubrication (MQL) process proffers better lubrication capability than conventional flood cooling system. (e) The developed fuzzy logic models were found to have high prediction accuracies of 97.22%, 98.60%, and 96.8%, respectively, for grinding power, grinding force ratio, and surface roughness.

Prediction of maximum oil-yield from almond seed in a chemical industry: A novel type-2 fuzzy logic approach

Almond (Termanilia catappa) is one of the most popular tree nuts on a world-wide basis due to its huge benefits in human health. Since it is not readily available in every country, research and investigations are still going on this particular. Therefore, here we have developed a unique model using type-2 fuzzy logic controller (T2FLC) approach to predict the optimum process conditions for the extraction of oil at maximum from almond seeds. Here, we have taken pressure, temperature, heating time and moisture content as input parameters and oil yield is taken as an output parameter to optimize the process. Though hundred percent recovery of oil by extraction of almond seeds is never possible, but controlling the process parameters at suitable condition, oil yield may vary within the range of 40%–45%. Using these four input and one output parameters, four Mamdani fuzzy inference systems are formed depending on the different membership functions of the variables. In T2FLC assists to trace inputs and outputs in a well-organized manner for building the inferences train so that various types of oil yield and its assessment can be predicted during extraction. Finally, a statistical analysis has been done using type-2 fuzzy data set to improve the control of process parameters that can be easily determined in the type-2 fuzzy prediction model to get high yield.

Container flow forecasting through neural networks based on metaheuristics

In this paper we propose a fuzzy neural network prediction approach based on metaheuristics for container flow forecasting. The approach uses fuzzy if–then rules for selection between two different heuristics for developing neural network architecture, simulated annealing and genetic algorithm, respectively. These non-parametric models are compared with traditional parametric ARIMA technique. Time series composed from monthly container traffic observations for Port of Barcelona are used for model developing and testing. Models are compared based on the most important criteria for performance evaluation and for each of the data sets (total container traffic, loaded, unloaded, transit and empty) the appropriate model is selected.

Fuzzy ridgelet neural network prediction model trained by improved particle swarm algorithm for maintenance decision of polypropylene plant

AbstractThe proper maintenance plan should be made for ensuring the safety and reliability of polypropylene plant and improve economic benefits of petrochemical enterprise. To meet the requirement, a novel maintenance prediction model of polypropylene plant based on fuzzy theory, ridgelet an artificial neural network is constructed. The economy and reliability models of polypropylene plant maintenance are established through comprehensively considering the reliability and economy. The basic structure of fuzzy ridgelet neural network is designed, and the training algorithm is improved through combining the traditional particle swarm algorithm and bacterial foraging algorithm, and the corresponding algorithm flow is confirmed. Finally, prediction simulation analysis is carried out using a polypropylene plant as research object, and analysis results show that the fuzzy ridgelet neural network has best prediction effect, and the optimal maintenance plan can be confirmed to ensure security and reduce maintenance cost of polypropylene plant.

Fuzzy Prediction of Power Lithium Ion Battery State of Function Based on the Fuzzy c-Means Clustering Algorithm

Following the widespread and large-scale application of power lithium ion battery, State of Function (SOF) estimation technology of power lithium ion batteries has gained an increasing amount of attention from both scientists and engineers. During the lifetime of the power lithium ion battery, SOF reflects the maximum instantaneous output power of the battery. When discarded, it is able to show the degree of performance degradation of the power battery when also taken as a performance evaluation parameter. In this paper, the variables closely related to SOF have been selected to conduct the fuzzy inference system, which is optimized by the fuzzy c-means clustering algorithm, to estimate the SOF of the power lithium ion battery, whose relations can be proved by experimental data. Our simulation results and experimental results demonstrate the feasibility and advantages of the estimation strategy.

Fuzzy prediction and early detection of stomach diseases by means of combined iteration fuzzy models

Fuzzy prediction and early detection of stomach diseases by means of combined iteration fuzzy models

Fuzzy prediction and early detection of stomach diseases by means of combined iteration fuzzy models

The work discusses aspects of decision rule synthesis for prediction and early diagnostics of stomach diseases. The distinguishing feature of heterogeneous fuzzy rules of decision making is the fact that they use information about the energetic condition of biologically active points and also features traditionally used in medical practice such as alcohol consumption, smoking tobacco, inheritance, etc. Use of different types of the original data allows us to provide diagnostic efficiency in decisions at the level 0.9 or greater, which makes it possible to recommend the research outcome for medical practice.

Using a Digital Camera Combined With Fitting Algorithm and T-S Fuzzy Neural Network to Determine the Turbidity in Water

We developed a method using a digital camera combined with fitting algorithm and T-S fuzzy neural network (T-S fnn) to measure the water turbidity accurately and efficiently. A turbidity-measuring device and image-processing software based on the common camera were designed to obtain the image of a standard solution after a constant light source passed through the sample and the RGB values, Lab values corresponding to the image were obtained. These RGB values were used as the input of the fuzzy neural network prediction model, the corresponding standard turbidity values were used as the output, and a camera-based fuzzy neural network turbidity measurement method was established. Simultaneously, the standard curve of turbidity measurement is established by fitting the turbidity using color component and color difference. The proposed method was applied to the measurement of standard turbidity solution and the results were compared with those of turbidmeter. The accuracy of the fuzzy neural network and the fitting algorithm is higher than that of the turbidimeter, and the accuracy of the fuzzy neural network method is the highest, the measurement error was only ±0.89%, and the accuracy much higher than that of an ordinary turbidimeter. By comparing the independent sample t-tests of the actual water samples, the fuzzy neural network method had the same trend as the turbidimeter and there was no significant difference of the results. The water turbidity measurement with the camera-T-S fuzzy neural network assembly can be applied to the actual water samples. This method can replace the traditional photoelectric detection method to measure turbidity, and it can reduce measurement error and cost. It may be used in environmental detection, biomedicine, and other fields.

Fuzzy Predictor With Additive Learning for Very Short-Term PV Power Generation

Photovoltaic (PV) power generation is highly intermittent in nature and any accurate very short-term prediction can decrease the impact of its uncertainties and operation costs and boost the reliable and efficient integration of PV systems into micro/smart grids. This work develops a new generalized technique for very short-term prediction of PV power generation from the lagged power generation data using fuzzy techniques. A preprocessor extracts relevant statistical features from the PV data which are fed to the fuzzy predictor. A modified version of Wang-Mendel training algorithm is employed to directly extract the fuzzy rules from the training data pairs. This methodology exploits the limited training data more efficiently. In addition, an online additive learning routine is proposed, which enables the predictor to learn from new data while running the predictions. So, the prediction accuracy increases over time and the predictor updates to account for long-term changing conditions of weather and PV system performance and its surroundings. Numerical results of the comparison of the proposed approach with simple fuzzy and traditional artificial neural network methods on a live PV system in the United Kingdom demonstrate its improved prediction accuracy, outperforming the benchmark approaches with a normalized mean absolute error (NMAE) of 3.6%.

Heat transfer dynamics modelling by means of clustering and swarm methods

This paper deals with the modelling problem of heat transfer dynamics in thermal exchanger process by using fuzzy prediction approaches. Clustering and swarm-based optimisation methods are used to derive heat transfer dynamical models to predict temperature variations of hot and cold fluids in the exchanger. The clustering method relies on a one-shot potential calculating strategy to extract the fuzzy sets distribution from the data space. However, the swarm optimisation method employs a subject function to optimise the premise and conclusion parameters of the fuzzy structure. Experimental data extracted from a pilot exchanger process is used to learn the fuzzy models, and their performances are compared on both training and testing measurement data.

Heat transfer dynamics modelling by means of clustering and swarm methods

This paper deals with the modelling problem of heat transfer dynamics in thermal exchanger process by using fuzzy prediction approaches. Clustering and swarm-based optimisation methods are used to derive heat transfer dynamical models to predict temperature variations of hot and cold fluids in the exchanger. The clustering method relies on a one-shot potential calculating strategy to extract the fuzzy sets distribution from the data space. However, the swarm optimisation method employs a subject function to optimise the premise and conclusion parameters of the fuzzy structure. Experimental data extracted from a pilot exchanger process is used to learn the fuzzy models, and their performances are compared on both training and testing measurement data.

Fuzzy Modelling for Prediction of Bursting Strength of Knitted Cotton Fabric using Bleaching Process Variables

A fuzzy prediction model has been built based on hydrogen peroxide concentration, bleaching temperature, and time of bleaching as the input variables and knitted cotton fabric bursting strength as the output variable. Fuzzy expert systems can map efficiently in nonlinear domains with minimal experimental data. The model developed in the present study has been validated by new experimental data. The root mean square, mean absolute error percentage, and coefficient of determination ( R 2 ) between the predicted and experimental values were found to be 4.89, 0.707, and 0.965 respectively. The results confirm that the model can be applied successfully for the prediction of fabric bursting strength in textile dye houses.

Research on Maintenance Spares Support Based on Combat Phrase

Aiming at the problem that maintenance spares are difficult to be accurately guaranteed in each combat phrase due to multiple influencing factors of maintenance spares consumption in the combat support process, strong randomness, constantly changing and difficult to be accurately quantified, a maintenance spares support method based on fuzzy data fusion is proposed. And a spares consumption fuzzy prediction model based on expert experience prediction value and Markov prediction value is established. On this basis, the influence factor is introduced to make the prediction results more realistic. The method is verified by an example. And the results show that the method can effectively realize the precise guarantee of maintenance spares in each combat phrase.

Ensuring the Security by Using Fuzzy Based Trust Routing Scheme in MANET

MANETs are substantially more defenseless to different attacks due to receptiveness in network topology and being endless of a preferred performance in the network. As a result of that, more malicious nodes are regularly intruding without being recognized from the network topology. MANET needs exceptionally particular security techniques to detach the false passage. Also, no single solution is fit into various sorts of the network. The networks perform admirably if the nodes are trusty and act appropriately helpfully. The goal is to enhance the security of the network. This paper begins the new intriguing way to deal with assessing the trustworthiness of the nodes. Fuzzy Trust-based Secured Routing (FTSR) approach presented to deal with a select the trusted route to meet the necessity of the security of the information transmission. In this, the fuzzy logic rule prediction component is adapted to see the future conduct of the node by refreshing the node’s trust. We have additionally broken down the execution measurements, for example, the packet transmission rate, end-to-end delay and average throughput which can likewise increment in most up to date approach.

Intelligent model for solar energy forecasting and its implementation for solar photovoltaic applications

As the demand for renewable energy is surging day-by-day, the solar energy data are important for applications in the field of solar photovoltaic (PV) systems. However, there exists a challenge in the collection of data owing to expensive instruments and a limited number of meteorological stations. In addition, the output of the system is largely affected due to variation in sky-conditions; therefore, an intelligent model based on sky-conditions is essential for estimating global solar energy so as to meet the energy requirements. In this work, the sky-based model employing fuzzy logic modelling has been developed and presented to forecast global solar energy using the dew-point as the meteorological parameter along with other known available parameters, namely, sunshine duration, wind speed, ambient temperature, and relative humidity for different sky-conditions, namely, clear sky (type-a), hazy sky (type-b), partially foggy/cloudy sky (type-c), and fully foggy/cloudy sky (type-d) respectively. Simulations have been performed for five meteorological stations across India that represents distinct climate zones such as composite, warm and humid, hot and dry, cold and cloudy, and moderate climate zone respectively, and the performance of the proposed model has been evaluated by using statistical indicators. The applicability of the proposed sky-based model employing fuzzy logic modelling can further be exploited for solar PV systems. The model is implemented in 210 W PV modules in forecasting the power output of solar photovoltaic systems in different sky-conditions. The obtained results reveal that the systems employing fuzzy logic modelling can be implemented for a wide range of applications and provide benefits. Furthermore, to check for accuracy of the proposed model, a comparative analysis has been carried out with the Angstrom model using statistical indicators. The value of the results, however, shows the supremacy of the proposed fuzzy logic prediction model.

Long-term electricity consumption forecasting based on expert prediction and fuzzy Bayesian theory

Long-term electricity consumption (EC) forecasting is a very important part for the expansion planning of power system. Instead of point forecasting, based on fuzzy Bayesian theory and expert prediction, a novel long-term probability forecasting model is proposed to predict the Chinese per-capita electricity consumption (PEC) and its variation interval over the period 2010–2030. The special model structure can improve the reliability and accuracy of expert prediction through econometric methodology. It contains three components: fuzzy relation matrix, prior prediction, and fuzzy Bayesian formula. To contend with the long-term uncertainty, the prior prediction is implemented to combine the advantages of expert's experience with other time-based methods from the perspective of probability. With the utilization of fuzzy technique, the multiple effects of influencing factors (IFs) on PEC can be expressed as a fuzzy relation matrix. It can rule the results of prior prediction to obey the long-run equilibrium relationship of natural evolution thorough probability calibration. To demonstrate its efficiency and applicability, the result of this method is compared with that of other 6 approaches and 4 agencies. The case study shows that the proposed methodology has higher accuracy and adaptability.

Prediction interval methodology based on fuzzy numbers and its extension to fuzzy systems and neural networks

Prediction interval modelling has been proposed in the literature to characterize uncertain phenomena and provide useful information from a decision-making point of view. In most of the reported studies, assumptions about the data distribution are made and/or the models are trained at one step ahead, which can decrease the quality of the interval in terms of the information about the uncertainty modelled for a higher prediction horizon. In this paper, a new prediction interval modelling methodology based on fuzzy numbers is proposed to solve the abovementioned drawbacks. Fuzzy and neural network prediction interval models are developed based on this proposed methodology by minimizing a novel criterion that includes the coverage probability and normalized average width. The fuzzy number concept is considered because the affine combination of fuzzy numbers generates, by definition, prediction intervals that can handle uncertainty without requiring assumptions about the data distribution. The developed models are compared with a covariance-based prediction interval method, and high-quality intervals are obtained, as determined by the narrower interval width of the proposed method. Additionally, the proposed prediction intervals are tested by forecasting up to two days ahead of the load of the Huatacondo microgrid in the north of Chile and the consumption of the residential dwellings in the town of Loughborough, UK. The results show that the proposed models are suitable alternatives to electrical consumption forecasting because they obtain the minimum interval widths that characterize the uncertainty of this type of stochastic process. Furthermore, the information provided by the obtained prediction interval could be used to develop robust energy management systems that, for example, consider the worst-case scenario.