Reliable Forecasts Research Articles

Quantifying circular economy pathways of decommissioned onshore wind turbines: The case of Denmark and Germany

Onshore wind turbines in Europe are increasingly reaching the end of their first lifecycle. Their pathways after decommissioning call for the establishment of circular supply chains (e.g. refurbishment or recycling facilities). Reliable component and material flow forecasts are particularly crucial for the development of blade-recycling capacity, as such facilities still need to be established. However, current forecasts assume a static decommissioning time and neglect a second lifecycle for the wind turbines and their blades, which has resulted in potential recycling quantities being over-estimated. This study overcomes these issues by (i) collecting empirical data on the circular economy pathways taken by decommissioned onshore wind turbines in the mature onshore wind markets of Denmark and Germany, and by (ii) proposing a new component and material flow forecasting model for the more reliable planning of blade-recycling capacity. The results reveal that ∼50–60 % of decommissioned onshore wind turbines in Denmark and Germany were exported mainly to other European countries. If the second lifecycle practices of the past are continued in the future, annual blade masses for domestic recycling are expected to range between ∼380–770 tonnes for Denmark and ∼4400–11,300 tonnes for Germany in the next ten years. This study finds that the threshold values of blade volumes for an economically viable blade-recycling facility can be reached in Germany with its large operating wind-turbine fleet, but the recycling of Danish wind turbine blades would have to rely on aggregating resource flows from other countries or industries. By modelling the cascading order of a sustainable circular economy and the EU Waste Hierarchy Directive, this study improves the decision-making basis for policy makers and companies to achieve sustainable resource use along the wind industry's entire value chain.

Reliability assessment and forecasting of moment ratio/factor of safety for sheet pile walls utilizing hybrid ANFIS enhanced by optimization techniques

Reliability assessment and forecasting of moment ratio/factor of safety for sheet pile walls utilizing hybrid ANFIS enhanced by optimization techniques

Impact of Feature Selection on the Prediction of Global Horizontal Irradiation under Ouarzazate City Climate

AbstractEnsuring accurate forecasts of Global Horizontal Irradiance (GHI) stands as a pivotal aspect in optimizing the efficient utilization of solar energy resources. Machine learning techniques offer promising prospects for predicting global horizontal irradiance. However, within the realm of machine learning,the importance of feature selection cannot be overestimated, as it is crucial in determining performance and reliability of predictive models. To address this, a comprehensive machine learning algorithm has been developed, leveraging advanced feature importance techniques to forecast GHI data with precision. The proposed models draw upon historical data encompassing solar irradiance characteristics and environmental variables within the Ouarzazate region, Morocco, spanning from 1st January 2018, to 31 December 2018, with readings taken at 60-minute intervals. The findings underscore the profound impact of feature selection on enhancing the predictive capabilities of machine learning models for GHI forecasting. By identifying and prioritizing the most informative features, the models exhibit significantly enhanced accuracy metrics, thereby bolstering the reliability, efficiency, and practical applicability of GHI forecasts. This advancement not only holds promise for optimizing solar energy utilization but also contributes to the broader discourse on leveraging machine learning for renewable energy forecasting and sustainability initiatives.

The operational CMEMS wind wave forecasting system of the Black Sea

ABSTRACT Accurate wind wave data is essential for ship navigation, safety at sea, and offshore energy production. Access to real-time and forecast wave data, with quality metrics, is vital for operational services. The Copernicus Marine Environment Monitoring Service (CMEMS) operational wave forecasting system of the Black Sea relies on the third-generation WAve Model (WAM). It offers a 1-day hindcast and 10-day forecast at 2.5 km spatial and 1-hour temporal resolutions. The model is driven by ECMWF IFS010 winds as well as surface currents and sea surface heights from a hydrodynamic model. The model's validation against satellite and buoy data indicates good overall performance. For two-year satellite significant wave heights, error statistics include a root mean square difference (RMSD) of 27 cm, a bias of – 16 cm, a scatter index of 0.22, and a correlation coefficient of 0.93. Wave buoy comparisons show a RMSD of 1.09 s for the TM02 period and a 30.5° standard deviation of the mean direction. The model performs well during storm events and provides reliable wave forecasts for up to four days, despite slightly underestimating the significant wave height. The driving wind fields generally perform well, but assessments reveal shortcomings along the coasts.

A hybrid model coupling process-driven and data-driven models for improved real-time flood forecasting

Accurate and reliable incoming flood forecasting is an important prerequisite for flood warning, flood risk analysis and reservoir flood control operation. This paper proposes a hybrid model for real-time flood forecasting that couples process-driven hydrological models (HMs) with data-driven models (DDMs). The generic hybrid model framework adds DDMs as the post-processing procedure for residual correction to the original results of HMs, and considers multiple uncertainties in input data, parameter and model structure simultaneously. The performance of the hybrid model is evaluated comprehensively in terms of deterministic forecast accuracy, interval forecast reliability, and the reliability and sharpness of probabilistic forecast. Taking the multireservoir system at the east Pi River as a study case, the results indicate that: (1) Compared to the benchmark model (ensemble XAJ model), the hybrid model with additional residual analysis show a significant improvement. The average continuous ranked probability score (CRPS) metric values calculated by the Stacking-Hybrid model improved by 71.5 %, 67.0 % and 38.1 % in the three data samples. Furthermore, the adaptability of the Stacking-Hybrid model for residual correction during short-duration intense rainfall events has been validated, with the relative error of the peak discharge improved to within ± 10 %. (2) The Stacking-Hybrid model, which also takes into account structure uncertainty, is able to better exploit the combined advantages and improve the stability of the model performance compared to those that only apply a single DDM. (3) When the number of iterations within the BOA reaches 300, the parameter optimization process is capable to search for the hyperparameters that bring out the best performance of the DDMs. (4) When the ensemble size reaches 200, the uncertainty of HM parameters can be fully defined, and the consumed computational resources can be controlled within an acceptable bound while ensuring stable model performance. Overall, the hybrid model that takes into account multiple sources of uncertainty generates both interval and probabilistic forecast in addition to deterministic forecast, which can provide richer risk information for subsequent flood warning and reservoir operation, making the flood prevention decisions more reliable and scientific.

Assessment of stability and operability of skidder on slopes of thawing soil mass during its interaction with a wheeled forest machine

In the Russian Federation, a significant part of the forest fund is located on permafrost. Permafrost forests play a global ecological role by de-populating large amounts of carbon. However, when a forest reaches the age of obsolescence, this carbon is released into the atmosphere when the plantation decays. Therefore, from the ecological point of view, permafrost forests should be cut down when they reach the age of ripeness, in compliance with the requirements for their restoration. Many of the above-mentioned regions have large reserves of mature and overmature production forests, and a significant part of these reserves are located in hard-to-reach places, not only in relation to the development of the road network, but also by terrain relief. Currently, the vast majority of timber harvesting in Russia is done with modern machinery complexes, mainly including wheeled forest machines of different layout and purpose (harvesters, forwarders, skidders, etc.). With the help of special technical solutions, such as winches integrated into the transmission, or self-propelled winches (T-winch, ROB, Winch Assist), these machines can be successfully operated even on fairly steep slopes. The problem of negative effect of wheeled forest machines and skidding systems based on them not only stays actual, but also becomes even more urgent, as the ecosystems of mountain forests, forests on the slopes of hills, etc. are the most vulnerable and exposed to water and wind erosion. This article presents a mathematical model for reliable forecasting of rut depth and stability of a hairpin section during dynamic interaction of a forest machine with the soil on slopes, taking into account geometric parameters of the slope, the laws of the oscillating process of a forest machine with given values of speed and amortization characteristics of wheel-pair elements, the profile of the hairpin section and physical and mechanical properties of the soil in its boundaries, the weight of forest machines and the pressure in tires of wheel-pairs.

Development of the Rodsim numerical simulator for the study of a sucker rod pump, comparing the Gibbs (1963) and Lea (1990) models

It is of great importance to be fully cognizant of the sucker rod pump system in order to make reliable production forecasts and to evaluate potential strategies for optimizing production processes. A frequently employed methodology in such circumstances is numerical simulation. A numerical sucker road pump simulator can provide valuable insights into production performance by generating dynamometric charts that represent the efficiency of this activity. Since the rod column behaves like a thin rod, it is possible to represent the mathematical model by a one-dimensional wave equation that describes the forces acting on this rod column. Lea (1990), proposing an improvement of the work of Gibbs (1963), considers the effects of the viscosity of the fluid produced in a vertical well without heat exchange, the piston diameter, the production column and the rods to obtain the damping factor, thus being more complete and producing a more accurate result in the dynamometric charts. Both models are implemented and discussed herein. The equations are discretized and solved in the Matlab® environment. The code includes a graphical user interface that generates the internal Rodsim application. The application is used to solve a typical case of a sucker rod pump and to obtain the surface and downhole dynamometric plots for certain scenarios.

QUALITY CONTROL METHODS AND MODELS OF POLYMER COMPOSITE MATERIALS

This article analyzes the traditional and modern methods of demolition control applied to structures made of polymer composite materials. Polymer composite materials are designed to ensure the strength of structures with a minimum mass, not subject to corrosion, etc. as there are advantages. But such materials require a special approach in the preparation and creation and the application of new methods, tools and solutions. This is due to the fact that there are many different types of polymer composite materials. Therefore, the development of criteria for the analysis of quality control methods and models, reliability forecasting in the technology of manufacturing products from polymer composite materials is an urgent issue. The authors give recommendations for the selection of methods for a complete assessment of the quality of compositional materials. The analysis of effective methods of quality control and forecasting of reliability of polymer composite materials was carried out. For this purpose, factors for the choice of the method and reliability prediction have been identified, taking into account its complexity for controlling polymer composite materials. Models for controlling polymer composite materials and products made on its basis were selected and on their basis the effectiveness of the application of low-frequency ultrasonic devices at the separation boundaries of controlled environments was determined.

A method for cleaning wind power anomaly data by combining image processing with community detection algorithms

Current methodologies for cleaning wind power anomaly data exhibit limited capabilities in identifying abnormal data within extensive datasets and struggle to accommodate the considerable variability and intricacy of wind farm data. Consequently, a method for cleaning wind power anomaly data by combining image processing with community detection algorithms (CWPAD-IPCDA) is proposed. To precisely identify and initially clean anomalous data, wind power curve (WPC) images are converted into graph structures, which employ the Louvain community recognition algorithm and graph- theoretic methods for community detection and segmentation. Furthermore, the mathematical morphology operation (MMO) determines the main part of the initially cleaned wind power curve images and maps them back to the normal wind power points to complete the final cleaning. The CWPAD-IPCDA method was applied to clean datasets from 25 wind turbines (WTs) in two wind farms in northwest China to validate its feasibility. A comparison was conducted using density-based spatial clustering of applications with noise (DBSCAN) algorithm, an improved isolation forest algorithm, and an image-based (IB) algorithm. The experimental results demonstrate that the CWPAD-IPCDA method surpasses the other three algorithms, achieving an approximately 7.23% higher average data cleaning rate. The mean value of the sum of the squared errors (SSE) of the dataset after cleaning is approximately 6.887 lower than that of the other algorithms. Moreover, the mean of overall accuracy, as measured by the F1-score, exceeds that of the other methods by approximately 10.49%; this indicates that the CWPAD-IPCDA method is more conducive to improving the accuracy and reliability of wind power curve modeling and wind farm power forecasting.

The Improvement of Short-Term Quantitative Precipitation Forecast in Mountainous Areas by the Assimilation of Meteorological State Variables Retrieved by Multiple Doppler Radar Data

Abstract This study presents a sequential procedure formulated by combining a multiple-Doppler radar wind synthesis technique with a thermodynamic retrieval method, which can be applied to retrieve the three-dimensional wind, pressure, temperature, rainwater mixing ratio, and moisture over complex terrain. The retrieved meteorological state variables are utilized to reinitialize a high-resolution numerical model, which then carries out time integration using four different microphysical (MP) schemes, including the Goddard Cumulus Ensemble (GCE), Morrison (MOR), WRF single-moment 6-class (WSM6), and WRF double-moment 6-class (WDM6) schemes. It is found that through this procedure, the short-term quantitative precipitation forecast (QPF) skill of a numerical model over mountainous areas can be significantly improved up to 6 h. The moisture field plays a crucial role in producing the correct rainfall forecast. Since no specific microphysical scheme outperforms the others, a combination of various rainfall scenarios forecasted by different MP schemes is suggested in order to provide a stable and reliable rainfall forecast. This work also demonstrates that, with the proposed approach, radar data from only two volume scans are sufficient to improve the rainfall forecasts. This is because the unobserved meteorological state variables are instantaneously retrieved and directly used to reinitialize the model, thereby the model spinup time can be effectively shortened.

Machine Learning Investigation of Downburst-Prone Environments in Canada

Abstract Thunderstorms are recognized as one of the most disastrous weather threats in Canada because of their power to cause substantial damage to human-made structures and even result in fatalities. It is therefore essential for operational forecasting to diagnose thunderstorms that generate damaging downdrafts of negatively buoyant air, known as downbursts. This study develops several machine learning models to identify environments supportive of downbursts in Canada. The models are trained and evaluated using 38 convective parameters calculated based on ERA5 reanalysis vertical profiles prior to thunderstorms with (306 cases) and without (19 132 cases) downbursts across Canada. Various resampling techniques are implemented to adjust data imbalance. An increase in the performance of the random forest (RF) model is observed when the support vector machine synthetic minority oversampling technique is utilized. The RF model outperforms other tested models, as indicated by model performance metrics and calibration. Several model interpretability methods highlight that the RF model has learned physical trends and patterns from the input variables. Moreover, the thermodynamic parameters are deemed to have higher impacts on the model outcomes compared to parcel, kinematic, and composite variables. For example, a considerable rise in the downburst probability is detected with an increase in cold pool strength. This study serves as one of the earliest attempts toward the fledgling field of machine learning applications in weather forecasting systems in Canada. The findings suggest that the developed model has the potential to enhance the effectiveness of issuing severe thunderstorm warnings in Canada, although further assessment with operational meteorologists is needed to validate its practical application. Significance Statement Severe thunderstorms demand particular attention in forecasting as their outflow can pose a serious threat to both structures and human life. This study uses machine learning techniques to predict whether or not a thunderstorm generates a damaging outflow in Canada. Atmospheric conditions that could trigger a severe thunderstorm are identified and discussed. Results show that the models have the potential to assist forecasters in better analyzing and predicting thunderstorms that generate destructive winds. Consequently, taking advantage of promising machine learning tools can yield more reliable forecasts of damaging thunderstorms, thereby mitigating the economic and societal burdens of these storms on Canadian communities.

Heatwave attribution based on reliable operational weather forecasts

The 2021 Pacific Northwest heatwave was so extreme as to challenge conventional statistical and climate-model-based approaches to extreme weather attribution. However, state-of-the-art operational weather prediction systems are demonstrably able to simulate the detailed physics of the heatwave. Here, we leverage these systems to show that human influence on the climate made this event at least 8 [2–50] times more likely. At the current rate of global warming, the likelihood of such an event is doubling every 20 [10–50] years. Given the multi-decade lower-bound return-time implied by the length of the historical record, this rate of change in likelihood is highly relevant for decision makers. Further, forecast-based attribution can synthesise the conditional event-specific storyline and unconditional event-class probabilistic approaches to attribution. If developed as a routine service in forecasting centres, it could provide reliable estimates of human influence on extreme weather risk, which is critical to supporting effective adaptation planning.

ПЕРСПЕКТИВИ ЗАСТОСУВАННЯ РЕКУРЕНТНОГО АНАЛІЗУ ДЛЯ ЦИКЛІЧНИХ ПРОЦЕСІВ У ВИРОБНИЦТВІ ЕЛЕКТРОЕНЕРГІЇ

The increasing demand for flexible and dynamic power systems drives the search for new modeling and forecasting methods. Recurrent analysis, which has gained wide recognition in natural language processing and machine learning, has the potential to become a promising tool for research and optimization of cyclic processes in electricity generation. This paper explores the potential applications of recurrent analysis for addressing various energy-related tasks including electricity demand forecasting, dispatch optimization, power plant scheduling, and consumption data analysis. The paper discusses the advantages of recurrent analysis, such as flexibility, learning ability, and high accuracy, as well as the challenges associated with large data volumes, computational resources, and result interpretation. It emphasizes the prospects of recurrent analysis for developing more efficient, reliable, and sustainable power systems. The article also explores the integration of recurrence analysis with other data-driven techniques such as machine learning and statistical modeling. This integration enhances the predictive capabilities of recurrence analysis, allowing for more accurate and reliable forecasts of energy production and demand cycles. The synergy between recurrence analysis and machine learning algorithms can lead to the development of advanced control systems that dynamically adjust to changing conditions in real-time, thus ensuring a more stable and efficient power supply. In conclusion, the article posits that recurrence analysis holds significant potential for enhancing the understanding and management of cyclic processes in power generation. By providing detailed insights into the temporal dynamics of energy production systems, recurrence analysis can contribute to more efficient and reliable power generation, ultimately supporting the transition to sustainable energy systems. The adoption of recurrence analysis in the energy sector represents a step forward in leveraging advanced mathematical techniques to address the complexities of modern power generation.

Exploring the Efficacy of Random Linear Parameter Models for Forecasting Heating Demand in District Heating Networks

The heating and cooling sector accounts for approximately half of global energy consumption, making it a pivotal focus for energy transition efforts. District heating and cooling networks have emerged as efficient, innovative, and dominant technologies in various regions. Accurate forecasting of heating demand within these networks is imperative for effective energy transition and responsive demand management. While existing forecasting techniques primarily concentrate on limited parameters such as ambient temperature and time of day, this paper explores the multifaceted impact of various factors, including ambient temperature, energy prices, number of consumers, building types, and solar radiation, on heating demand. Through a comprehensive study of these determinants, a robust and efficient forecasting model is developed in this research. Specifically, a novel Random Linear Parameter Model (RLPM) is introduced for hourly heating demand forecasting within district heating networks. The model's performance is assessed and compared with existing forecasting models, namely Linear Regression, ARIMA, and RLPM. A notable advantage of the Random Parameter Linear Regression model lies in its capability to recognize unobserved parameters influencing demand and account for unobserved heterogeneity, providing a robust framework for incorporating these factors into the forecasting process. This study aims to underscore the proposed approach's advantages in terms of accuracy and its ability to estimate the determinants of heating load, thereby offering reliable future forecasting. The findings have the potential to significantly contribute to optimizing district heating systems by delivering more nuanced and dependable predictions of heating demand.

Parabolic Modeling Forecasts of Space and Time European Hydropower Production

Renewable sources of energy production are some of the main targets today to protect the environment through reduced fossil fuel consumption and CO2 emissions. Alongside wind, solar, marine, biomass and nuclear sources, hydropower is among the oldest but still not fully explored renewable energy sources. Compared with other sources like wind and solar, hydropower is more stable and consistent, offering increased predictability. Even so, it should be analyzed considering water flow, dams capacity, climate change, irrigation, navigation, and so on. The aim of this study is to propose a forecast model of hydropower production capacity and identify long-term trends. The curve fit forecast parabolic model was applied to 33 European countries for time series data from 1990 to 2021. Space-time cube ArcGIS representation in 2D and 3D offers visualization of the prediction and model confidence rate. The quadratic trajectory fit the raw data for 14 countries, validated by visual check, and in 20 countries, validated by FMRSE 10% threshold from the maximal value. The quadratic model choice is good for forecasting future values of hydropower electric capacity in 22 countries, with accuracy confirmed by the VMRSE 10% threshold from the maximal value. Seven local outliers were identified, with only one validated as a global outlier based on the Generalized Extreme Studentized Deviate (GESD) test at a 5% maximal number of outliers and a 90% confidence level. This result achieves our objective of estimating a level with a high degree of occurrence and offering a reliable forecast of hydropower production capacity. All European countries show a growing trend in the short term, but the trends show a stagnation or decrease if policies do not consider intensive growth through new technology integration and digital adoption. Unfortunately, Europe does not have extensive growth potential compared with Asia–Pacific. Public policies must boost hybrid hydro–wind or hydro–solar systems and intensive technical solutions.

Weather Forecasting

My Project is on implementation of a weather forecasting using python and machine learning. real-time weather data from locations across the world. Through Python scripting, the project generates accurate forecasts for diverse regions, providing essential meteorological parameters such as temperature, wind speed, humidity, and weather conditions. Weather forecasting plays a crucial role in various aspects of human life, from planning outdoor activities to making strategic decisions in sectors like agriculture, transportation, and emergency management. In this project, we explore the application of Python programming language in weather forecasting, employing both open-source libraries and APIs to access and analyze weather data. Through this project, we aim to demonstrate the versatility of Python in harnessing weather data for forecasting purposes, showcasing its capabilities in data retrieval, analysis, visualization, and predictive modelling. By empowering users with accessible tools and methodologies, we envision a future where weather forecasting becomes more accurate, reliable, and actionable, contributing to informed decision-making and enhanced societal resilience in the face of changing weather patterns. Weather forecasting is the scientific process of predicting atmospheric conditions at a specific location and time. It involves the collection and analysis of data from various sources, such as satellites, weather stations, and radar systems. This data is then input into numerical weather prediction models, which use mathematical equations to simulate the behavior of the atmosphere. Advances in technology, such as high-performance computing and machine learning, have significantly improved the accuracy and reliability of weather forecasts. These predictions are crucial for a wide range of applications, including agriculture, disaster management, transportation, and daily planning, helping to mitigate the impacts of severe weather events and enhance societal preparedness and safety. Despite significant progress, challenges remain due to the inherently chaotic nature of the atmosphere, which limits the precision of long-term forecasts

Tailoring Waterborne Coating Rheology with Hydrophobically Modified Ethoxylated Urethanes (HEURs): Molecular Architecture Insights Supported by CG-MD Simulations.

A novel investigation of the effects of the hydrophilic and hydrophobic segments of hydrophobically modified ethoxylated urethanes (HEURs) on the rheological properties of their aqueous solutions, latex-based emulsions, and waterborne paints is demonstrated. Different HEUR thickeners were produced by varying the poly(ethylene glycol) (PEG) molecular weight and terminal hydrophobic size. Results reveal that the strength of hydrophobic associations and, consequently, the rheological properties of HEUR formulations can be effectively controlled by modifying the structure of the hydrophobic segment, specifically, the combination of diisocyanate and monoalcohol. This allows for the on-demand attainment of diverse rheological behaviors ranging from predominantly Newtonian profiles exhibiting lower viscosities to markedly pseudoplastic behaviors with significantly higher viscosities. The length of the hydrophilic group appears to affect viscosity only marginally up to a molecular weight of 23,000 g/mol, with more notable effects at 33,000 g/mol. Additionally, it was indicated that the rheological responses observed in water solutions provide a reliable forecast of their behavior in latex-based emulsions and waterborne paints. Coarse-grained molecular dynamics (CG-MD) simulations were also applied to gain insight into HEUR micelle dynamics in aqueous solutions. Guided by the DBSCAN algorithm, the simulations successfully captured the concentration-dependent behavior and the impact of hydrophilic chain length, aligning with the experimental viscosity trends. Various metrics were employed to provide a comprehensive analysis of the micellization process, including the hydrophobic cluster volume, the total micellar volume, the aggregation number, and the number of chains interconnecting with other micelles.

Intrusive mechanism of the 2018 Sierra Negra Galápagos eruption, constrained by 4D tomographic images

Understanding the mechanisms of magma intrusion underpins our ability to interpret geophysical monitoring signals at volcanoes and thus issue reliable forecasts of future activity. The basaltic caldera volcanoes of the western Galápagos Islands, Ecuador, exhibit exceptionally high rates of deformation driven by shallow magma accumulation and migration. However, the nature and evolution of magma storage at these volcanoes is poorly constrained by earthquake hypocentre locations or geodetic inversions. Here we show that transient variations in seismic velocity before, during, and after the 2018 eruption of the Sierra Negra volcano track the accumulation of magma in a shallow sill complex and the emplacement of a lateral flank intrusion. A four-dimensional tomographic technique applied to the P-wave arrivals of local earthquakes provides high spatiotemporal resolution of changes in the physical properties of the shallow volcanic system. In the month before the eruption, the expansion of a low-velocity zone above the sub-caldera sill complex coincides with caldera uplift and near-surface fracturing, driven by persistent shallow magma accumulation. A new low-velocity anomaly appeared progressively in the western flank of the volcano in the days after the onset of the eruption, coinciding with the opening of a curved sill that supplied magma to secondary eruptive fissures. The anomaly disappeared as the curved sill deflated after the initial opening, despite it remaining the conduit for magma from the caldera complex to the flank fissures. Low velocities across the shallow caldera after the end of the eruption likely result from rapid inflation due to recharge with fresh magma from depth. These results indicate a previously unknown complexity to the magmatic plumbing system at Sierra Negra and suggest velocity changes resulting from an interplay of thermal and stress perturbations.

IS THE PHILLIPS CURVE USEFUL FOR FORECASTING INFLATION IN RUSSIA?

The paper analyzes the experience of using the Phillips curve to model inflation in Russia, taking into account the specific features of domestic labor market. Based on Russian data for the period from 2000 to 2022, a wide range of inflation forecasting models has been built: both based on the Phillips curve and alternative ones. The econometric tools used are autoregression models with a moving average in the residuals taking into account seasonality (SARIMA) and their generalizations; autoregression models of distributed lags (ADL) and their generalizations; as well as other estimation methods. During the simulation, we use data on inflation, inflation expectations, production dynamics, unemployment, wages, exchange rates, money supply and other variables. The models are compared drawing on the accuracy of single-period and multi-period out-of-sample forecasts. The modelling results allows us to conclude that onedimensional models work well during the periods of stable economic dynamics, but lose in their predictive power to the “triangular” Phillips curve in crisis years. Comparison of models for forecasting inflation shows that in a stable economic situation, one-dimensional models provide a more reliable forecast. However, in the context of structural transformation faced by the Russian economy in 2022, the “triangular” models of the Phillips curve demonstrate maximum quality of the forecast. Although the acceleration of inflation in 2022 obviously reduces the accuracy of any forecast equations, however, the “triangular” model based on the lags in inflation, unemployment and the index of industrial production demonstrates the best results. This conclusion remains stable to changes in the length of the time series used for forecasting, as well as to changes in forecasting horizon.

The Use of Artificial Intelligence and Machine Learning in Forecasting the Financial Growth of Automobile Industries

The automotive industry's financial forecasting is vital for growth and competitiveness in today's rapidly evolving market landscape. Accurate and reliable forecasting is essential for strategic planning, resource allocation, and decision-making processes. However, traditional forecasting methods often struggle to adapt to the dynamic nature of the automotive sector, characterized by fluctuating consumer demands, evolving market trends, and complex supply chain dynamics. The integration of Artificial Intelligence (AI) and Machine Learning (ML) technologies has revolutionized financial forecasting in the automotive industry, transforming forecasting accuracy, efficiency, and decision-making processes. By leveraging AI algorithms and ML models, automotive companies can gain deeper insights into market dynamics, predict consumer behavior more accurately, optimize production schedules, and streamline distribution processes. This paper explores the transformative impact of AI and ML technologies on financial forecasting in the automotive industry, delving into key applications, benefits, challenges, and future directions of integrating these advanced technologies into forecasting processes. The adoption of AI and ML empowers automotive companies to make data-driven decisions with greater precision and agility, driving innovation, growth, and competitiveness in this dynamic sector. Key Words: Artificial Intelligence, Machine Learning, Financial Forecasting, Automotive Industry, Predictive Maintenance, Hybrid Decision Support Systems, Production Planning and Control Systems, Intelligent Transport Logistics.