Probabilistic Forecasting System Research Articles

A User-Focused Approach to Evaluating Probabilistic and Categorical Forecasts

Abstract A user-focused verification approach for evaluating probability forecasts of binary outcomes (also known as probabilistic classifiers) is demonstrated that (i) is based on proper scoring rules, (ii) focuses on user decision thresholds, and (iii) provides actionable insights. It is argued that when categorical performance diagrams and the critical success index are used to evaluate overall predictive performance, rather than the discrimination ability of probabilistic forecasts, they may produce misleading results. Instead, Murphy diagrams are shown to provide a better understanding of the overall predictive performance as a function of user probabilistic decision threshold. We illustrate how to select a proper scoring rule, based on the relative importance of different user decision thresholds, and how this choice impacts scores of overall predictive performance and supporting measures of discrimination and calibration. These approaches and ideas are demonstrated using several probabilistic thunderstorm forecast systems as well as synthetic forecast data. Furthermore, a fair method for comparing the performance of probabilistic and categorical forecasts is illustrated using the fixed risk multicategorical (FIRM) score, which is a proper scoring rule directly connected to values on the Murphy diagram. While the methods are illustrated using thunderstorm forecasts, they are applicable for evaluating probabilistic forecasts for any situation with binary outcomes. Significance Statement Recently, several papers have presented verification results for probabilistic forecasts using so-called categorical performance diagrams, which summarize multiple verification metrics. While categorical performance diagrams measure discrimination ability, we demonstrate how they can potentially lead to incorrect conclusions when evaluating overall predictive performance of probabilistic forecasts. By reviewing recent advances in the statistical literature, we show a comprehensive approach for the meteorological community that (i) does not reward a forecaster who “hedges” their forecast, (ii) focuses on the importance of the forecast user’s decision threshold(s), and (iii) provides actionable insights. Additionally, we present an approach for fairly comparing the skill of categorical forecasts to probabilistic forecasts.

Improving urban water demand forecast using conformal prediction-based hybrid machine learning models

This paper presents a probabilistic forecasting method that predicts the future water consumption patterns, taking into account the inherent uncertainty in the system. The proposed method leverages statistical techniques to estimate the probability distribution of future water demand scenarios. The findings of this study will provide decision-makers with a range of possible alternatives that facilitate better planning and management of water resources. We developed a conformal prediction-based hybrid demand forecast model, which combines convolutional neural network (CNN) and bidirectional long short-term memory (BiLSTM) while comparing it with other machine learning models for probabilistic hourly water demand forecasting. Additionally, we address crucial considerations when implementing a probabilistic forecasting system, including selecting appropriate data and choosing model parameters. The performance of the proposed model is validated for probabilistic water demand forecasting in real-world settings. Results indicate noteworthy improvement in deterministic and probabilistic predictions by 10 % and 26.7 %, respectively. The findings underscore the potential benefits of this approach for improved decision-making and resource management. The results also demonstrate that the proposed hybrid model outperforms the compared machine learning models with better prediction performance and increased robustness for handling prediction uncertainties.

Optimal Unit Commitment and Generation Scheduling of Integrated Power System with Plug-In Electric Vehicles and Renewable Energy Sources

The integration of wind energy sources and plug-in electric vehicles is essential for the efficient planning, reliability, and operation of modern electric power systems. Minimizing the overall operational cost of integrated power systems while dealing with wind energy sources and plug-in electric vehicles in integrated power systems using a chaotic zebra optimization algorithm (CZOA) is described. The proposed system deals with a probabilistic forecasting system for wind power generation and a realistic plug-in electric vehicle charging profile based on travel patterns and infrastructure characteristics. The objective is to identify the optimal scheduling and committed status of the generating unit for thermal and wind power generation while considering the system power demand, charging, and discharging of electric vehicles, as well as power available from wind energy sources. The proposed CZOA adeptly tackles the intricacies of the unit commitment problem by seamlessly integrating scheduling and the unit’s committed status, thereby enabling highly effective optimization. The proposed algorithm is tested for 10-, 20-, and 40-generating unit systems. The empirical findings pertaining to the 10-unit system indicate that the amalgamation of a thermal generating unit system with plug-in electric vehicles yields a 0.84% reduction in total generation cost. Furthermore, integrating the same system with a wind energy source results in a substantial 12.71% cost saving. Notably, the integration of the thermal generating system with both plug-in electric vehicles and a wind energy source leads to an even more pronounced overall cost reduction of 13.05%. The outcome of this study reveals competitive test results for 20- and 40-generating unit systems and contributes to the advancement of sustainable and reliable power systems, fostering the transition towards a greener energy future.

A novel probabilistic forecasting system based on quantile combination in electricity price

In the electricity market, the accuracy of electricity price forecasting is significant for real-time control; however, the complexity and volatility of electricity prices make this a challenge. Existing forecasting models focus on deterministic forecasting and rarely address the uncertainty in electricity price forecasting. Therefore, this study fills this knowledge gap by introducing a novel combined probability forecasting system (CPFS) and creatively incorporating probability density estimation based on kernel functions in a multi-objective optimization algorithm. In addition, to effectively integrate the forecasting components, the tuna optimization algorithm was enhanced to overcome the limitations of traditional multi-objective optimization algorithms. Finally, the validity of the CPFS is confirmed through two electricity price cases, considering three equally important aspects: reliability, resolution, and sharpness. From a comprehensive perspective, CPFS outperformed the most advanced benchmark by more than 5.66% and 38.93% in AIS and by more than 13.41% and 3.55% in quantile loss on the NSW and Singapore datasets, respectively. The experimental results demonstrate that the CPFS provides an effective range for electricity price fluctuations. Furthermore, given that probabilistic forecasting is essential for risk management, it offers important implications for the electricity market.

A novel combined probabilistic load forecasting system integrating hybrid quantile regression and knee improved multi-objective optimization strategy

In contemporary power systems, diverse energy integration and technological advancements have complicated load dynamics, revealing limitations in traditional deterministic forecasting techniques. Given the pivotal role of Probabilistic Load Forecasting (PLF) in mitigating risks in power system planning and operation, the pursuit of more accurate PLF techniques has evidently become a topic of paramount importance. Therefore, in this study, a novel hybrid Quantile Regression (QR) model integrating causal dilated convolution, residual connection, and Bidirectional Long Short-Term Memory (BiLSTM) is initially proposed for PLF, its core multi-scale feature extraction module can deeply mine the interactions between loads and their features. However, relying solely on a single-model often encounters inherent flaws in certain scenarios. Recognizing this, a Combined Probabilistic Load Forecasting System (CPLFS) is further proposed. This system incorporates the developed hybrid QR to establish a diversified candidate model pool, aiming to select the top-performing benchmarks in various scenarios and utilizes the Knee enhanced Pareto mechanism to adjust the combined weights of each benchmark, thereby bridging the gap created by the “No Free Lunch (NFL)” phenomenon. Taking the load uncertainty of Australia as an illustrative example, simulations indicate that the hybrid QR outperforms traditional models. Furthermore, CPLFS surpasses the best benchmarks and other combinations in both prediction accuracy and stability.

PROBABILISTIC FORECASTING SYSTEM OF STORM SURGE USING A STOCHASTIC TYPHOON MODEL

This paper presents a newly developed real time probabilistic forecasting system of storm surge at an arbitrary target location when a typhoon is formed and approaches to the location. Such probabilistic information may be essential for business or public entities along the coast who need to make appropriate decision to what extent they should break their daily operations and prepare for the upcoming coastal hazard. Developed system, based on Monte Carlo simulation, consists of two parts: (i) generation of a number of virtual typhoons; and (ii) storm surge prediction for each typhoon.

A Multimodel Real-Time System for Global Probabilistic Subseasonal Forecasts of Precipitation and Temperature

Abstract A global multimodel probabilistic subseasonal forecast system for precipitation and near-surface temperature is developed based on three NOAA ensemble prediction systems that make their forecasts available publicly in real time as part of the Subseasonal Experiment (SubX). The weekly and biweekly ensemble means of precipitation and temperature of each model are individually calibrated at each grid point using extended logistic regression, prior to forming equal-weighted multimodel ensemble (MME) probabilistic forecasts. Reforecast skill of week-3–4 precipitation and temperature is assessed in terms of the cross-validated ranked probability skill score (RPSS) and reliability diagram. The multimodel reforecasts are shown to be well calibrated for both variables. Precipitation is moderately skillful over many tropical land regions, including Latin America, sub-Saharan Africa and Southeast Asia, and over subtropical South America, Africa, and Australia. Near-surface temperature skill is considerably higher than for precipitation and extends into the extratropics as well. The multimodel RPSS skill of both precipitation and temperature is shown to exceed that of any of the constituent models over Indonesia, South Asia, South America, and East Africa in all seasons. An example real-time week-3–4 global forecast for 13–26 November 2021 is illustrated and shown to bear the hallmarks of the combined influences of a moderate Madden–Julian oscillation event as well as weak–moderate ongoing La Niña event. Significance Statement This paper develops a system for forecasting of precipitation and temperatures globally over land, several weeks in advance, with a focus on biweekly averaged conditions between three and four weeks ahead. The system provides the likelihood of biweekly and weekly conditions being below, near, or above their long-term averages, as well the probability of exceeding (or not exceeding) any threshold value. Using historical data, the precipitation forecasts are demonstrated to have skill in many tropical regions, and the temperature forecasts are more widely skillful. While weather and seasonal range forecasts have become quite generally available, this is one of the first examples of a publicly available, calibrated multimodel probabilistic real-time forecasting system for the subseasonal range.

Research of a novel combined deterministic and probabilistic forecasting system for air pollutant concentration

Hazy weather seriously threatens human health, providing accurate and reliable air pollutant concentration forecast will be of great scientific value and benefit to residents' livelihood and natural environment. However, existing forecasting models lack outlier detection, feature mapping and uncertainty forecast, which results in unreliable forecasting results. Thus, in this paper, a novel combined deterministic and probabilistic forecasting system based on meta-heuristic algorithm is proposed, which contains multiple data pre-processing, combined deterministic and probabilistic forecast. Firstly, the raw data is pre-processed and analyzed from multiple perspectives to fully extract the chaotic features. Then multi-objective white shark optimizer is utilized to combine the three deep learning models to enhance the forecasting accuracy and stability. Furthermore, the Pin-ball loss function is employed to yield prediction intervals at different confidence levels, and a novel interval pseudo-code is constructed to optimize the prediction intervals thus promoting the probabilistic forecasting reliability. The simulations indicated that the developed system enhances the forecasting accuracy by maximum 60.6772%, 54.9793% and 38.6876% in the three steps of PM2.5 prediction and up to 64.9001%, 68.4637% in the interval sharpness at significance level 0.05 and 0.1 than the tested models.

Tropical Cyclone Intensity Probabilistic Forecasting System Based on Deep Learning

Tropical cyclones (TC) are one of the extreme disasters that have the most significant impact on human beings. Unfortunately, intensity forecasting of TC has been a difficult and bottleneck in weather forecasting. Recently, deep learning-based intensity forecasting of TC has shown the potential to surpass traditional methods. However, due to the Earth system’s complexity, nonlinearity, and chaotic effects, there is inherent uncertainty in weather forecasting. Besides, previous studies have not quantified the uncertainty, which is necessary for decision-making and risk assessment. This study proposes an intelligent system based on deep learning, PTCIF, to quantify this uncertainty based on multimodal meteorological data, which, to our knowledge, is the first study to assess the uncertainty of TC based on a deep learning approach. In this study, probabilistic forecasts are made for the intensity of 6–24 hours. Experimental results show that our proposed method is comparable to the forecast performance of weather forecast centers in terms of deterministic forecasts. Moreover, reliable prediction intervals and probabilistic forecasts can be obtained, which is vital for disaster warning and is expected to be a complement to operational models.

Neighborhood-Based Ensemble Evaluation Using the CRPS

Abstract The neighborhood-based ensemble evaluation using the continuous ranked probability score is based on the pooling of the cumulative density function (CDF) for all the points inside a neighborhood. This methodology can be applied to the forecast CDF for measuring the predictive input of neighboring points in the center of the neighborhood. It can also be applied at the same time to forecast CDF and observed CDF so as to quantify the quality of the pooled ensemble forecast at the scale of the neighborhood. Fair versions of these two neighborhood scores are also defined in order to reduce their dependencies on the size of ensemble forecasts. The borderline case of deterministic forecasts is also explored so as to be able to compare them with ensemble forecasts. The information of these new scores is analyzed on idealized and real cases of rain accumulated during 3 h and of 2-m temperature forecast by four deterministic and probabilistic forecasting systems operational at Météo-France.

Development and Evaluation of Global Korean Aviation Turbulence Forecast Systems Based on an Operational Numerical Weather Prediction Model and In Situ Flight Turbulence Observation Data

Abstract A global Korean deterministic aviation turbulence guidance (G-KTG) system and a global Korean probabilistic turbulence forecast (G-KPT) system are developed using outputs from the operational Global Data Assimilation and Prediction System of the Korea Meteorological Administration, and the performance skill of the systems are evaluated against in situ flight eddy dissipation rates (EDRs) recorded for one year (September 2018–August 2019). G-KTG and G-KPT consider clear-air turbulence (CAT) and mountain wave turbulence diagnostics, while G-KTG additionally considers near-cloud turbulence (NCT) diagnostics. In the G-KTG system, the various combinations of deterministic EDR forecasts are tested by different ensemble means of individual turbulence diagnostics. In the G-KPT system, the probabilistic forecast is established by counting the number of diagnostics that exceed a certain threshold for strong intensity turbulence on the given model grid. The evaluation results of the G-KTG system based on the area under the relative operating characteristic curve (AUC) reveal that G-KTG, which consists of CAT and NCT diagnostics, shows the highest AUC value among the various G-KTG combinations; in addition, the summertime performance is significantly improved when NCT diagnostics are included. In the evaluation results of the G-KTG system over the globe, U.S., and East Asia regions, the recent graphical turbulence guidance system–based G-KTG shows better performance than the regional KTG–based G-KTG for all three regions. For all altitude bands, the G-KPTs with 40% probability as the minimal threshold for alerting forecasters of strong turbulence show higher values of true skill statistic than the G-KTGs.

A novel ensemble probabilistic forecasting system for uncertainty in wind speed

The quantification of wind speed uncertainty is of great significance for real-time control of wind turbines and power grid dispatching. However, the intermittence and fluctuation of wind energy present great challenges in modeling its uncertainty; research in this field is limited. A quantile regression bi-directional long short-term memory network (QrBiLStm) and a novel ensemble probabilistic forecasting strategy are proposed in this study to explore ensemble probabilistic forecasting. To verify the reliability of the proposed ensemble probabilistic forecasting system, the uncertainties of wind speed at wind farms in China were modeled as a case study. The results of comparative experiments including 15 other models demonstrate the superiority of this ensemble probabilistic forecasting system in terms of sharpness while maintaining high interval coverage. More specifically, it was observed that the prediction interval coverage probability obtained by the proposed system is above 97%, and the sharpness is improved by at least 24.21% as compared with the commonly used single models. The proposed ensemble probabilistic forecasting system can accurately quantify the uncertainty of wind speed, and also reduce the operation cost of power systems by improving the efficiency of wind energy utilization.

Comparing Partial and Continuously Cycling Ensemble Kalman Filter Data Assimilation Systems for Convection-Allowing Ensemble Forecast Initialization

Abstract Several limited-area 80-member ensemble Kalman filter (EnKF) data assimilation systems with 15-km horizontal grid spacing were run over a computational domain spanning the conterminous United States (CONUS) for a 4-week period. One EnKF employed continuous cycling, where the prior ensemble was always the 1-h forecast initialized from the previous cycle’s analysis. In contrast, the other EnKFs used a partial cycling procedure, where limited-area states were discarded after 12 or 18 h of self-contained hourly cycles and reinitialized the next day from global model fields. “Blended” states were also constructed by combining large scales from global ensemble initial conditions (ICs) with small scales from limited-area continuously cycling EnKF analyses using a low-pass filter. Both the blended states and EnKF analysis ensembles initialized 36-h, 10-member ensemble forecasts with 3-km horizontal grid spacing. Continuously cycling EnKF analyses initialized ∼1–18-h precipitation forecasts that were comparable to or somewhat better than those with partial cycling EnKF ICs. Conversely, ∼18–36-h forecasts with partial cycling EnKF ICs were comparable to or better than those with unblended continuously cycling EnKF ICs. However, blended ICs yielded ∼18–36-h forecasts that were statistically indistinguishable from those with partial cycling ICs. ICs that more closely resembled global analysis spectral characteristics at wavelengths > 200 km, like partial cycling and blended ICs, were associated with relatively good ∼18–36-h forecasts. Ultimately, findings suggest that EnKFs employing a combination of continuous cycling and blending can potentially replace the partial cycling assimilation systems that currently initialize operational limited-area models over the CONUS without sacrificing forecast quality. SIGNIFICANCE STATEMENT Numerical weather prediction models (i.e., weather models) are initialized through a process called data assimilation, which combines real atmospheric observations with a previous short-term weather model forecast using statistical techniques. The overarching data assimilation strategy currently used to initialize operational regional weather models over the United States has several disadvantages that ultimately limit progress toward improving weather model forecasts. Thus, we suggest an alternative data assimilation strategy be adopted to initialize a next-generation, high-resolution (∼3 km) probabilistic forecast system currently being developed. This alternative approach preserves forecast quality while fostering a framework that can accelerate weather model improvements, which in turn will lead to better weather forecasts.

Air quality deterministic and probabilistic forecasting system based on hesitant fuzzy sets and nonlinear robust outlier correction

Performing scientific and accurate forecasting and realizing the quantitative description of uncertainties in air quality remain challenging prospects. Because of the strong volatility and uncertainty of air pollutant concentrations, this problem increases in difficulty when multiple requirements are considered. In this study, a novel air quality deterministic and probabilistic forecasting system based on hesitant fuzzy sets and nonlinear robust outlier correction was proposed to realize air quality early warning. The proposed system solves the non-stochastic non-deterministic problem in air quality forecasting considering a novel hesitant fuzzy time series forecasting model wherein the intervals are partitioned by different approaches with optimal weights determined by the multi-objective JAYA algorithm. The forecasting performance is further enhanced with the introduction of a new nonlinear error correction model based on an outlier robust extreme learning machine and multi-objective JAYA algorithm, and the quality of the solution obtained is verified by sensitivity analysis. However, point forecast information alone is not sufficient to facilitate the rational integration of pollution control measures. Therefore, this study conducts probabilistic forecasting and constructs proper prediction intervals based on the optimal distribution of the forecasting residuals. By comparing the results with typical counterparts and comparison models considering multiple metrics, the experimental results confirmed the improvement scheme proposed in this study on the traditional fuzzy time series forecasting method while the effectiveness of applying the proposed system to air quality early warning was confirmed as well.

Calibrated ensemble forecasts of the height of new snow using quantile regression forests and ensemble model output statistics

Abstract. Height of new snow (HN) forecasts help to prevent critical failures of infrastructures in mountain areas, e.g. transport networks and ski resorts. The French national meteorological service, Météo-France, operates a probabilistic forecasting system based on ensemble meteorological forecasts and a detailed snowpack model to provide ensembles of HN forecasts. These forecasts are, however, biased and underdispersed. As for many weather variables, post-processing methods can be used to alleviate these drawbacks and obtain meaningful 1 to 4 d HN forecasts. In this paper, we compare the skill of two post-processing methods. The first approach is an ensemble model output statistics (EMOS) method, which can be described as a nonhomogeneous regression with a censored shifted Gamma distribution. The second approach is based on quantile regression forests, using different meteorological and snow predictors. Both approaches are evaluated using a 22 year reforecast. Thanks to a larger number of predictors, the quantile regression forest is shown to be a powerful alternative to EMOS for the post-processing of HN ensemble forecasts. The gain of performance is large in all situations but is particularly marked when raw forecasts completely miss the snow event. This type of situation happens when the rain–snow transition elevation is overestimated by the raw forecasts (rain instead of snow in the raw forecasts) or when there is no precipitation in the forecast. In that case, quantile regression forests improve the predictions using the other weather predictors (wind, temperature, and specific humidity).

From Random Forests to Flood Forecasts: A Research to Operations Success Story

AbstractExcessive rainfall is difficult to forecast, and there is a need for tools to aid Weather Prediction Center (WPC) forecasters when generating Excessive Rainfall Outlooks (EROs), which are issued for the contiguous United States at lead times of 1–3 days. To address this need, a probabilistic forecast system for excessive rainfall, known as the Colorado State University Machine Learning Probabilities (CSU-MLP) system, was developed based on ensemble reforecasts, precipitation observations, and machine-learning algorithms, specifically random forests. The CSU-MLP forecasts were designed to emulate the EROs, with the goal being a tool that forecasters can use as a “first guess” in the ERO forecast process. Resulting from close collaboration between CSU and WPC and evaluation at the Flash Flood and Intense Rainfall Experiment, iterative improvements were made to the forecast system and it was transitioned into operational use at WPC. Quantitative evaluation shows that the CSU-MLP forecasts are skillful and reliable, and they are now being used as a part of the WPC forecast process. This project represents an example of a successful research-to-operations transition, and highlights the potential for machine learning and other postprocessing techniques to improve operational predictions.

Beyond Strictly Proper Scoring Rules: The Importance of Being Local

AbstractThe evaluation of probabilistic forecasts plays a central role both in the interpretation and in the use of forecast systems and their development. Probabilistic scores (scoring rules) provide statistical measures to assess the quality of probabilistic forecasts. Often, many probabilistic forecast systems are available while evaluations of their performance are not standardized, with different scoring rules being used to measure different aspects of forecast performance. Even when the discussion is restricted to strictly proper scoring rules, there remains considerable variability between them; indeed strictly proper scoring rules need not rank competing forecast systems in the same order when none of these systems are perfect. The locality property is explored to further distinguish scoring rules. The nonlocal strictly proper scoring rules considered are shown to have a property that can produce “unfortunate” evaluations, particularly the fact that the continuous rank probability score prefers the outcome close to the median of the forecast distribution regardless of the probability mass assigned to the value at/near the median raises concern to its use. The only local strictly proper scoring rule, the logarithmic score, has direct interpretations in terms of probabilities and bits of information. The nonlocal strictly proper scoring rules, on the other hand, lack meaningful direct interpretation for decision support. The logarithmic score is also shown to be invariant under smooth transformation of the forecast variable, while the nonlocal strictly proper scoring rules considered may, however, change their preferences due to the transformation. It is therefore suggested that the logarithmic score always be included in the evaluation of probabilistic forecasts.

Comprehensive assessment of water supply and sanitation in single-industry towns

The article presents the results of evaluating the main direction characterizing the state of health and quality of life of the population of single-industry towns: the sphere of water supply and sanitation. To improve the health status and quality of life of the population in single-industry towns, an action plan is proposed to bring the quality of drinking water in accordance with the established requirements for the period from 2017 to 2022. Prediction of improving the quality of drinking water, including lowering the overall hardness, as a result of the installation of a drinking water treatment system as part of the study, was based on the retrospective results of the introduction of similar drinking water treatment systems in municipalities of the constituent entities of the Federation and a probabilistic forecasting system. The results of the calculations showed that the average expected value of improving the quality of drinking water will be 7.039 %.

Study on the applicability of a stochastic typhoon model for probabilistic forecasting of storm surge induced by a typhoon

ABSTRACT This study investigates the applicability of a stochastic typhoon model, STM, for probabilistic forecasting of storm surge induced by a typhoon. When a typhoon forms and approaches the coast, the present STM generates a number of virtual typhoons with initial conditions specified by the recent history of the actual typhoon. Storm surge is then computed for each of the generated typhoons, and the probabilistic characteristics of the estimated storm surges are investigated. The present STM is based on the higher-order autoregressive model, which can account for the history of the characteristics of individual typhoons. The concept of minimal pressure was introduced to improve the predictive skills of the statistical characteristics of extreme typhoons. The model was applied to the cases of Typhoons Jebi and Trami, which hit Japan in 2018. The present system reasonably explained the contrasting characteristics of Jebi and Trami in terms of their peak storm surge levels in the inner part of Osaka Bay. Furthermore, the sensitivity analysis of the present probabilistic forecasting system suggested that a relatively large number of virtual typhoons are needed for forecasting of the worst storm surge conditions induced by the arbitrary actual typhoons approaching the target coast.

A novel system for multi-step electricity price forecasting for electricity market management

Electricity price forecasting is an important and challenging issue for all participants in the power market because of the wide application of electricity in our society and its inherent features. In this context, some current forecasting systems use data preprocessing and optimization for theoretical and practical achievements. However, some limitations to these systems exist which need to be urgently solved. First, future information is overdrawn in the data preprocessing stage of these forecasting systems, which is actually unknown in practical applications. The crucial question, therefore, is how to develop a forecasting system without using any future information. Second, the complex features of original nonlinear and nonstationary electricity price have a negative influence on the generalization ability of these previously developed models. To decrease the negative effects on management, a method to develop a forecasting system to improve the model’s generalization ability is required. Therefore, in this study, we developed an adaptive deterministic and probabilistic interval forecasting system for multi-step electricity price forecasting, which can present more valuable information to power market decision makers. Two cases and one comparative study are provided and analyzed to validate the performance of the developed system in multi-step electricity price forecasting. Furthermore, further discussions are presented to illustrate the significance of this study, thus proving that the results of the present study fill the present knowledge gap and provide some new future directions for related studies.